W0521 16:05:17.164000 81005 torch/distributed/run.py:774] 
W0521 16:05:17.164000 81005 torch/distributed/run.py:774] *****************************************
W0521 16:05:17.164000 81005 torch/distributed/run.py:774] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 
W0521 16:05:17.164000 81005 torch/distributed/run.py:774] *****************************************
Distributed init debug info:
RANK: 1
LOCAL_RANK: 1
WORLD_SIZE: 2
MASTER_ADDR: 127.0.0.1
MASTER_PORT: 2207
torch.distributed.is_initialized: True
torch.distributed.get_rank(): 1
torch.distributed.get_world_size(): 2
Distributed init debug info:
RANK: 0
LOCAL_RANK: 0
WORLD_SIZE: 2
MASTER_ADDR: 127.0.0.1
MASTER_PORT: 2207
torch.distributed.is_initialized: True
torch.distributed.get_rank(): 0
torch.distributed.get_world_size(): 2
[2026-05-21 16:05:21,502] INFO [src.utils.basic_utils:10] rank0: init wandb
wandb: [wandb.login()] Loaded credentials for https://api.wandb.ai from WANDB_API_KEY.
[2026-05-21 16:05:21,616] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.
[2026-05-21 16:05:21,616] WARNING [huggingface_hub.utils._http:904] Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.
[2026-05-21 16:05:21,624] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:21,711] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:21,719] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[transformers] The fast path is not available because one of the required library is not installed. Falling back to torch implementation. To install follow https://github.com/fla-org/flash-linear-attention#installation and https://github.com/Dao-AILab/causal-conv1d

Loading weights:   0%|          | 0/473 [00:00<?, ?it/s]
Loading weights: 100%|██████████| 473/473 [00:00<00:00, 8521.49it/s]
[2026-05-21 16:05:21,918] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/generation_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:21,927] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/generation_config.json "HTTP/1.1 200 OK"
wandb: Currently logged in as: infinitylogesh to https://api.wandb.ai. Use `wandb login --relogin` to force relogin
[2026-05-21 16:05:22,769] INFO [src.utils.basic_utils:10] rank1: model_backbone: qwen3_5
[2026-05-21 16:05:22,849] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:22,859] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:22,944] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/tree/main/additional_chat_templates?recursive=false&expand=false "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:23,024] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:23,034] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
wandb: setting up run bx5yye6h
[2026-05-21 16:05:23,114] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/chat_template.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:23,199] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/chat_template.jinja "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:23,209] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/chat_template.jinja "HTTP/1.1 200 OK"
[2026-05-21 16:05:23,294] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/audio_tokenizer_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:23,386] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/tree/main/additional_chat_templates?recursive=false&expand=false "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:23,466] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:23,475] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:23,554] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/chat_template.json "HTTP/1.1 404 Not Found"
wandb: Tracking run with wandb version 0.26.1
wandb: Run data is saved locally in /workspace/VLM2Vec/outputs/Qwen3.5-08b-continued-mixed-alltasks/wandb/run-20260521_160521-bx5yye6h
wandb: Run `wandb offline` to turn off syncing.
wandb: Syncing run Qwen3.5-08b-continued-mixed-alltasks
wandb: ⭐️ View project at https://wandb.ai/infinitylogesh/vlm2vec-debug
wandb: 🚀 View run at https://wandb.ai/infinitylogesh/vlm2vec-debug/runs/bx5yye6h
[2026-05-21 16:05:23,658] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/chat_template.jinja "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:23,667] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/chat_template.jinja "HTTP/1.1 200 OK"
[2026-05-21 16:05:23,747] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/audio_tokenizer_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:23,750] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:23,761] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:23,766] INFO [src.utils.basic_utils:19] Loading backbone [qwen3_5] from infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750
[2026-05-21 16:05:23,829] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:23,837] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:23,848] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.
[2026-05-21 16:05:23,848] WARNING [huggingface_hub.utils._http:904] Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.
[2026-05-21 16:05:23,859] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[transformers] The fast path is not available because one of the required library is not installed. Falling back to torch implementation. To install follow https://github.com/fla-org/flash-linear-attention#installation and https://github.com/Dao-AILab/causal-conv1d

Loading weights:   0%|          | 0/473 [00:00<?, ?it/s][2026-05-21 16:05:23,927] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/preprocessor_config.json "HTTP/1.1 404 Not Found"

Loading weights: 100%|██████████| 473/473 [00:00<00:00, 23639.88it/s]
[2026-05-21 16:05:24,014] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:24,023] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:24,027] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/generation_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:24,035] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/generation_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:24,037] INFO [src.utils.basic_utils:19] Loading lora adapter from Qwen3_5ForConditionalGeneration(
  (model): Qwen3_5Model(
    (visual): Qwen3_5VisionModel(
      (patch_embed): Qwen3_5VisionPatchEmbed(
        (proj): Conv3d(3, 768, kernel_size=(2, 16, 16), stride=(2, 16, 16))
      )
      (pos_embed): Embedding(2304, 768)
      (rotary_pos_emb): Qwen3_5VisionRotaryEmbedding()
      (blocks): ModuleList(
        (0-11): 12 x Qwen3_5VisionBlock(
          (norm1): LayerNorm((768,), eps=1e-06, elementwise_affine=True)
          (norm2): LayerNorm((768,), eps=1e-06, elementwise_affine=True)
          (attn): Qwen3_5VisionAttention(
            (qkv): Linear(in_features=768, out_features=2304, bias=True)
            (proj): Linear(in_features=768, out_features=768, bias=True)
          )
          (mlp): Qwen3_5VisionMLP(
            (linear_fc1): Linear(in_features=768, out_features=3072, bias=True)
            (linear_fc2): Linear(in_features=3072, out_features=768, bias=True)
            (act_fn): GELUTanh()
          )
        )
      )
      (merger): Qwen3_5VisionPatchMerger(
        (norm): LayerNorm((768,), eps=1e-06, elementwise_affine=True)
        (linear_fc1): Linear(in_features=3072, out_features=3072, bias=True)
        (act_fn): GELU(approximate='none')
        (linear_fc2): Linear(in_features=3072, out_features=1024, bias=True)
      )
    )
    (language_model): Qwen3_5TextModel(
      (embed_tokens): Embedding(248320, 1024)
      (layers): ModuleList(
        (0-2): 3 x Qwen3_5DecoderLayer(
          (linear_attn): Qwen3_5GatedDeltaNet(
            (act): SiLUActivation()
            (conv1d): Conv1d(6144, 6144, kernel_size=(4,), stride=(1,), padding=(3,), groups=6144, bias=False)
            (norm): Qwen3_5RMSNormGated()
            (out_proj): Linear(in_features=2048, out_features=1024, bias=False)
            (in_proj_qkv): Linear(in_features=1024, out_features=6144, bias=False)
            (in_proj_z): Linear(in_features=1024, out_features=2048, bias=False)
            (in_proj_b): Linear(in_features=1024, out_features=16, bias=False)
            (in_proj_a): Linear(in_features=1024, out_features=16, bias=False)
          )
          (mlp): Qwen3_5MLP(
            (gate_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (up_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (down_proj): Linear(in_features=3584, out_features=1024, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
          (post_attention_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
        )
        (3): Qwen3_5DecoderLayer(
          (self_attn): Qwen3_5Attention(
            (q_proj): Linear(in_features=1024, out_features=4096, bias=False)
            (k_proj): Linear(in_features=1024, out_features=512, bias=False)
            (v_proj): Linear(in_features=1024, out_features=512, bias=False)
            (o_proj): Linear(in_features=2048, out_features=1024, bias=False)
            (q_norm): Qwen3_5RMSNorm((256,), eps=1e-06)
            (k_norm): Qwen3_5RMSNorm((256,), eps=1e-06)
          )
          (mlp): Qwen3_5MLP(
            (gate_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (up_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (down_proj): Linear(in_features=3584, out_features=1024, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
          (post_attention_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
        )
        (4-6): 3 x Qwen3_5DecoderLayer(
          (linear_attn): Qwen3_5GatedDeltaNet(
            (act): SiLUActivation()
            (conv1d): Conv1d(6144, 6144, kernel_size=(4,), stride=(1,), padding=(3,), groups=6144, bias=False)
            (norm): Qwen3_5RMSNormGated()
            (out_proj): Linear(in_features=2048, out_features=1024, bias=False)
            (in_proj_qkv): Linear(in_features=1024, out_features=6144, bias=False)
            (in_proj_z): Linear(in_features=1024, out_features=2048, bias=False)
            (in_proj_b): Linear(in_features=1024, out_features=16, bias=False)
            (in_proj_a): Linear(in_features=1024, out_features=16, bias=False)
          )
          (mlp): Qwen3_5MLP(
            (gate_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (up_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (down_proj): Linear(in_features=3584, out_features=1024, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
          (post_attention_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
        )
        (7): Qwen3_5DecoderLayer(
          (self_attn): Qwen3_5Attention(
            (q_proj): Linear(in_features=1024, out_features=4096, bias=False)
            (k_proj): Linear(in_features=1024, out_features=512, bias=False)
            (v_proj): Linear(in_features=1024, out_features=512, bias=False)
            (o_proj): Linear(in_features=2048, out_features=1024, bias=False)
            (q_norm): Qwen3_5RMSNorm((256,), eps=1e-06)
            (k_norm): Qwen3_5RMSNorm((256,), eps=1e-06)
          )
          (mlp): Qwen3_5MLP(
            (gate_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (up_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (down_proj): Linear(in_features=3584, out_features=1024, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
          (post_attention_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
        )
        (8-10): 3 x Qwen3_5DecoderLayer(
          (linear_attn): Qwen3_5GatedDeltaNet(
            (act): SiLUActivation()
            (conv1d): Conv1d(6144, 6144, kernel_size=(4,), stride=(1,), padding=(3,), groups=6144, bias=False)
            (norm): Qwen3_5RMSNormGated()
            (out_proj): Linear(in_features=2048, out_features=1024, bias=False)
            (in_proj_qkv): Linear(in_features=1024, out_features=6144, bias=False)
            (in_proj_z): Linear(in_features=1024, out_features=2048, bias=False)
            (in_proj_b): Linear(in_features=1024, out_features=16, bias=False)
            (in_proj_a): Linear(in_features=1024, out_features=16, bias=False)
          )
          (mlp): Qwen3_5MLP(
            (gate_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (up_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (down_proj): Linear(in_features=3584, out_features=1024, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
          (post_attention_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
        )
        (11): Qwen3_5DecoderLayer(
          (self_attn): Qwen3_5Attention(
            (q_proj): Linear(in_features=1024, out_features=4096, bias=False)
            (k_proj): Linear(in_features=1024, out_features=512, bias=False)
            (v_proj): Linear(in_features=1024, out_features=512, bias=False)
            (o_proj): Linear(in_features=2048, out_features=1024, bias=False)
            (q_norm): Qwen3_5RMSNorm((256,), eps=1e-06)
            (k_norm): Qwen3_5RMSNorm((256,), eps=1e-06)
          )
          (mlp): Qwen3_5MLP(
            (gate_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (up_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (down_proj): Linear(in_features=3584, out_features=1024, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
          (post_attention_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
        )
        (12-14): 3 x Qwen3_5DecoderLayer(
          (linear_attn): Qwen3_5GatedDeltaNet(
            (act): SiLUActivation()
            (conv1d): Conv1d(6144, 6144, kernel_size=(4,), stride=(1,), padding=(3,), groups=6144, bias=False)
            (norm): Qwen3_5RMSNormGated()
            (out_proj): Linear(in_features=2048, out_features=1024, bias=False)
            (in_proj_qkv): Linear(in_features=1024, out_features=6144, bias=False)
            (in_proj_z): Linear(in_features=1024, out_features=2048, bias=False)
            (in_proj_b): Linear(in_features=1024, out_features=16, bias=False)
            (in_proj_a): Linear(in_features=1024, out_features=16, bias=False)
          )
          (mlp): Qwen3_5MLP(
            (gate_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (up_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (down_proj): Linear(in_features=3584, out_features=1024, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
          (post_attention_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
        )
        (15): Qwen3_5DecoderLayer(
          (self_attn): Qwen3_5Attention(
            (q_proj): Linear(in_features=1024, out_features=4096, bias=False)
            (k_proj): Linear(in_features=1024, out_features=512, bias=False)
            (v_proj): Linear(in_features=1024, out_features=512, bias=False)
            (o_proj): Linear(in_features=2048, out_features=1024, bias=False)
            (q_norm): Qwen3_5RMSNorm((256,), eps=1e-06)
            (k_norm): Qwen3_5RMSNorm((256,), eps=1e-06)
          )
          (mlp): Qwen3_5MLP(
            (gate_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (up_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (down_proj): Linear(in_features=3584, out_features=1024, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
          (post_attention_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
        )
        (16-18): 3 x Qwen3_5DecoderLayer(
          (linear_attn): Qwen3_5GatedDeltaNet(
            (act): SiLUActivation()
            (conv1d): Conv1d(6144, 6144, kernel_size=(4,), stride=(1,), padding=(3,), groups=6144, bias=False)
            (norm): Qwen3_5RMSNormGated()
            (out_proj): Linear(in_features=2048, out_features=1024, bias=False)
            (in_proj_qkv): Linear(in_features=1024, out_features=6144, bias=False)
            (in_proj_z): Linear(in_features=1024, out_features=2048, bias=False)
            (in_proj_b): Linear(in_features=1024, out_features=16, bias=False)
            (in_proj_a): Linear(in_features=1024, out_features=16, bias=False)
          )
          (mlp): Qwen3_5MLP(
            (gate_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (up_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (down_proj): Linear(in_features=3584, out_features=1024, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
          (post_attention_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
        )
        (19): Qwen3_5DecoderLayer(
          (self_attn): Qwen3_5Attention(
            (q_proj): Linear(in_features=1024, out_features=4096, bias=False)
            (k_proj): Linear(in_features=1024, out_features=512, bias=False)
            (v_proj): Linear(in_features=1024, out_features=512, bias=False)
            (o_proj): Linear(in_features=2048, out_features=1024, bias=False)
            (q_norm): Qwen3_5RMSNorm((256,), eps=1e-06)
            (k_norm): Qwen3_5RMSNorm((256,), eps=1e-06)
          )
          (mlp): Qwen3_5MLP(
            (gate_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (up_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (down_proj): Linear(in_features=3584, out_features=1024, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
          (post_attention_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
        )
        (20-22): 3 x Qwen3_5DecoderLayer(
          (linear_attn): Qwen3_5GatedDeltaNet(
            (act): SiLUActivation()
            (conv1d): Conv1d(6144, 6144, kernel_size=(4,), stride=(1,), padding=(3,), groups=6144, bias=False)
            (norm): Qwen3_5RMSNormGated()
            (out_proj): Linear(in_features=2048, out_features=1024, bias=False)
            (in_proj_qkv): Linear(in_features=1024, out_features=6144, bias=False)
            (in_proj_z): Linear(in_features=1024, out_features=2048, bias=False)
            (in_proj_b): Linear(in_features=1024, out_features=16, bias=False)
            (in_proj_a): Linear(in_features=1024, out_features=16, bias=False)
          )
          (mlp): Qwen3_5MLP(
            (gate_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (up_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (down_proj): Linear(in_features=3584, out_features=1024, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
          (post_attention_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
        )
        (23): Qwen3_5DecoderLayer(
          (self_attn): Qwen3_5Attention(
            (q_proj): Linear(in_features=1024, out_features=4096, bias=False)
            (k_proj): Linear(in_features=1024, out_features=512, bias=False)
            (v_proj): Linear(in_features=1024, out_features=512, bias=False)
            (o_proj): Linear(in_features=2048, out_features=1024, bias=False)
            (q_norm): Qwen3_5RMSNorm((256,), eps=1e-06)
            (k_norm): Qwen3_5RMSNorm((256,), eps=1e-06)
          )
          (mlp): Qwen3_5MLP(
            (gate_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (up_proj): Linear(in_features=1024, out_features=3584, bias=False)
            (down_proj): Linear(in_features=3584, out_features=1024, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
          (post_attention_layernorm): Qwen3_5RMSNorm((1024,), eps=1e-06)
        )
      )
      (norm): Qwen3_5RMSNorm((1024,), eps=1e-06)
      (rotary_emb): Qwen3_5TextRotaryEmbedding()
    )
  )
  (lm_head): Linear(in_features=1024, out_features=248320, bias=False)
)
[2026-05-21 16:05:24,104] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/preprocessor_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:24,187] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:24,197] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:24,282] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/tokenizer_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:24,290] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/tokenizer_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:24,372] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/tokenizer_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:24,381] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/tokenizer_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:24,464] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/tree/main/additional_chat_templates?recursive=false&expand=false "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:24,565] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/tree/main?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:24,801] INFO [src.utils.basic_utils:10] rank0: model_backbone: qwen3_5
[2026-05-21 16:05:24,801] INFO [src.utils.basic_utils:19] Loading processor from: infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750
[2026-05-21 16:05:24,882] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:24,891] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:25,010] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/tree/main/additional_chat_templates?recursive=false&expand=false "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:25,089] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:25,098] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:25,186] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/chat_template.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:25,270] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/chat_template.jinja "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:25,278] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/chat_template.jinja "HTTP/1.1 200 OK"
[2026-05-21 16:05:25,367] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/audio_tokenizer_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:25,450] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/tree/main/additional_chat_templates?recursive=false&expand=false "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:25,524] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750 "HTTP/1.1 200 OK"
[2026-05-21 16:05:25,530] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:25,539] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:25,621] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:25,622] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/chat_template.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:25,629] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:25,705] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/chat_template.jinja "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:25,712] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/video_preprocessor_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:25,714] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/chat_template.jinja "HTTP/1.1 200 OK"
[2026-05-21 16:05:25,794] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/preprocessor_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:25,796] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/audio_tokenizer_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:25,874] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:25,877] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:25,883] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:25,886] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:25,962] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/video_preprocessor_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:25,967] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/preprocessor_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:26,055] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:26,064] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:26,068] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/preprocessor_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:26,143] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/preprocessor_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:26,157] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:26,165] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:26,225] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:26,234] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:26,247] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:26,320] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/tokenizer_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:26,331] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/tokenizer_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:26,456] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/tokenizer_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:26,465] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/tokenizer_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:26,550] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/tree/main/additional_chat_templates?recursive=false&expand=false "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:26,636] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/tree/main?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:27,190] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:27,289] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/revision/76dd0a440b6d4c02776830a804443fffbb2d0bfa "HTTP/1.1 200 OK"
[2026-05-21 16:05:27,379] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:27,493] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750 "HTTP/1.1 200 OK"
[2026-05-21 16:05:27,540] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:27,588] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:27,597] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:27,632] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/A-OKVQA?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:27,676] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/video_preprocessor_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:27,712] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa?recursive=false&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:27,756] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/preprocessor_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:27,792] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:27,835] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/processor_config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:27,843] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/processor_config.json "HTTP/1.1 200 OK"
[2026-05-21 16:05:27,886] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/ImageNet_1K?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:27,929] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/video_preprocessor_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:27,996] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:28,005] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:28,010] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/preprocessor_config.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:28,088] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:28,093] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:28,102] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:28,160] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:28,181] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:28,256] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:28,356] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:28,439] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:28,528] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/N24News?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:28,624] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:28,632] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:28,713] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:28,786] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:28,885] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:28,987] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:29,072] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:29,160] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/HatefulMemes?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:29,249] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:29,257] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:29,337] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:29,411] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:29,441] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:29,504] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:29,535] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/revision/76dd0a440b6d4c02776830a804443fffbb2d0bfa "HTTP/1.1 200 OK"
[2026-05-21 16:05:29,600] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:29,617] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:29,692] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:29,739] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:29,779] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/VOC2007?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:29,826] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/A-OKVQA?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:29,910] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa?recursive=false&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:29,926] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:29,938] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:29,991] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,013] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,078] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/ImageNet_1K?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:30,086] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,091] INFO [src.utils.basic_utils:19] Loaded mmeb/ImageNet_1K dataset with 100000 samples
[2026-05-21 16:05:30,091] INFO [src.utils.basic_utils:19] 		Dataset#0 (dataset_parser=mmeb): ImageNet_1K, num_rows=100000, prob=1.9230769230769231
[2026-05-21 16:05:30,169] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:30,178] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:30,181] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,256] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,274] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:30,327] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,363] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,424] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,452] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/SUN397?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:30,519] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:30,537] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:30,547] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:30,599] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,632] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,699] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/N24News?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:30,706] INFO [src.utils.basic_utils:19] Loaded mmeb/N24News dataset with 48988 samples
[2026-05-21 16:05:30,706] INFO [src.utils.basic_utils:19] 		Dataset#1 (dataset_parser=mmeb): N24News, num_rows=48988, prob=1.9230769230769231
[2026-05-21 16:05:30,707] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,795] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:30,804] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:30,813] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,887] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:30,923] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:30,961] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,017] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,061] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,103] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/OK-VQA?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:31,159] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:31,193] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:31,201] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:31,240] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,280] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,341] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/HatefulMemes?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:31,348] INFO [src.utils.basic_utils:19] Loaded mmeb/HatefulMemes dataset with 8500 samples
[2026-05-21 16:05:31,348] INFO [src.utils.basic_utils:19] 		Dataset#2 (dataset_parser=mmeb): HatefulMemes, num_rows=8500, prob=0.9615384615384616
[2026-05-21 16:05:31,354] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,425] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:31,433] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:31,452] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,544] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:31,568] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,628] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,639] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,723] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:31,731] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:31,735] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,810] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,829] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:31,882] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,913] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:31,979] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:32,037] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/VOC2007?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:32,043] INFO [src.utils.basic_utils:19] Loaded mmeb/VOC2007 dataset with 7844 samples
[2026-05-21 16:05:32,043] INFO [src.utils.basic_utils:19] 		Dataset#3 (dataset_parser=mmeb): VOC2007, num_rows=7844, prob=0.9615384615384616
[2026-05-21 16:05:32,074] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:32,120] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:32,128] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:32,156] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:32,210] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:32,242] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/DocVQA?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:32,282] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:32,352] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:32,361] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:32,374] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:32,445] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:32,469] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:32,518] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:32,549] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:32,622] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:32,635] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/SUN397?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:32,641] INFO [src.utils.basic_utils:19] Loaded mmeb/SUN397 dataset with 19850 samples
[2026-05-21 16:05:32,641] INFO [src.utils.basic_utils:19] 		Dataset#4 (dataset_parser=mmeb): SUN397, num_rows=19850, prob=0.9615384615384616
[2026-05-21 16:05:32,716] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:32,718] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:32,727] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:32,800] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:32,814] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:32,885] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:32,885] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/InfographicsVQA?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:32,988] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,016] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:33,026] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:33,088] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:33,106] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,171] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,180] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,262] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/OK-VQA?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:33,271] INFO [src.utils.basic_utils:19] Loaded mmeb/OK-VQA dataset with 9009 samples
[2026-05-21 16:05:33,271] INFO [src.utils.basic_utils:19] 		Dataset#5 (dataset_parser=mmeb): OK-VQA, num_rows=9009, prob=0.9615384615384616
[2026-05-21 16:05:33,273] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,353] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:33,362] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:33,369] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:33,443] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,457] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,516] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,556] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/ChartQA?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:33,636] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,677] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:33,685] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:33,732] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:33,768] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,816] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,824] INFO [src.utils.basic_utils:19] Loaded mmeb/A-OKVQA dataset with 17056 samples
[2026-05-21 16:05:33,824] INFO [src.utils.basic_utils:19] 		Dataset#6 (dataset_parser=mmeb): A-OKVQA, num_rows=17056, prob=0.9615384615384616
[2026-05-21 16:05:33,840] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,904] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:33,913] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:33,935] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:33,996] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:34,029] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:34,071] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:34,113] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:34,165] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:34,200] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/Visual7W?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:34,260] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:34,295] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:34,304] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:34,385] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:34,385] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:34,458] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:34,474] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/DocVQA?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:34,482] INFO [src.utils.basic_utils:19] Loaded mmeb/DocVQA dataset with 39463 samples
[2026-05-21 16:05:34,483] INFO [src.utils.basic_utils:19] 		Dataset#7 (dataset_parser=mmeb): DocVQA, num_rows=39463, prob=1.9230769230769231
[2026-05-21 16:05:34,555] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:34,561] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:34,569] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:34,647] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:34,651] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:34,720] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:34,730] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:34,815] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/VisDial?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:34,824] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:34,923] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:34,924] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:34,932] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:35,007] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,015] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,089] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,092] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/InfographicsVQA?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:35,100] INFO [src.utils.basic_utils:19] Loaded mmeb/InfographicsVQA dataset with 23946 samples
[2026-05-21 16:05:35,100] INFO [src.utils.basic_utils:19] 		Dataset#8 (dataset_parser=mmeb): InfographicsVQA, num_rows=23946, prob=0.9615384615384616
[2026-05-21 16:05:35,186] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:35,187] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,194] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:35,273] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,283] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:35,346] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,365] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,444] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,453] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/CIRR?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:35,539] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:35,559] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:35,568] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:35,622] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,652] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,707] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/ChartQA?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:35,716] INFO [src.utils.basic_utils:19] Loaded mmeb/ChartQA dataset with 28000 samples
[2026-05-21 16:05:35,716] INFO [src.utils.basic_utils:19] 		Dataset#9 (dataset_parser=mmeb): ChartQA, num_rows=28000, prob=0.9615384615384616
[2026-05-21 16:05:35,726] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,793] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:35,801] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:35,821] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,890] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,917] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:35,963] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:35,999] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:36,060] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:36,087] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/VisualNews_t2i?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:36,155] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:36,197] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:36,212] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:36,238] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:36,310] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:36,327] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/Visual7W?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:36,336] INFO [src.utils.basic_utils:19] Loaded mmeb/Visual7W dataset with 69817 samples
[2026-05-21 16:05:36,336] INFO [src.utils.basic_utils:19] 		Dataset#10 (dataset_parser=mmeb): Visual7W, num_rows=69817, prob=1.9230769230769231
[2026-05-21 16:05:36,384] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:36,415] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:36,424] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:36,482] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:36,503] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:36,574] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:36,582] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:36,662] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:36,670] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:36,749] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/VisualNews_i2t?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:36,769] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:36,847] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:36,866] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:36,874] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:36,931] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/VisDial?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:36,941] INFO [src.utils.basic_utils:19] Loaded mmeb/VisDial dataset with 123287 samples
[2026-05-21 16:05:36,941] INFO [src.utils.basic_utils:19] 		Dataset#11 (dataset_parser=mmeb): VisDial, num_rows=123287, prob=1.9230769230769231
[2026-05-21 16:05:36,956] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,018] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:37,026] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:37,027] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,104] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,121] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,177] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,215] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:37,273] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,293] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,367] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:37,380] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MSCOCO_t2i?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:37,448] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,502] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:37,511] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:37,533] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/CIRR?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:37,541] INFO [src.utils.basic_utils:19] Loaded mmeb/CIRR dataset with 26116 samples
[2026-05-21 16:05:37,541] INFO [src.utils.basic_utils:19] 		Dataset#12 (dataset_parser=mmeb): CIRR, num_rows=26116, prob=0.9615384615384616
[2026-05-21 16:05:37,591] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,624] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:37,633] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:37,665] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,712] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,763] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,784] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,864] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:37,880] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,947] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:37,980] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:38,035] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MSCOCO_i2t?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:38,098] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:38,146] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:38,156] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:38,182] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/VisualNews_t2i?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:38,190] INFO [src.utils.basic_utils:19] Loaded mmeb/VisualNews_t2i dataset with 99903 samples
[2026-05-21 16:05:38,190] INFO [src.utils.basic_utils:19] 		Dataset#13 (dataset_parser=mmeb): VisualNews_t2i, num_rows=99903, prob=1.9230769230769231
[2026-05-21 16:05:38,236] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:38,273] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:38,282] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:38,307] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:38,362] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:38,402] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:38,434] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:38,495] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:38,529] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:38,575] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:38,625] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:38,661] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/NIGHTS?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:38,713] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:38,755] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:38,764] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:38,802] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/VisualNews_i2t?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:38,811] INFO [src.utils.basic_utils:19] Loaded mmeb/VisualNews_i2t dataset with 100000 samples
[2026-05-21 16:05:38,811] INFO [src.utils.basic_utils:19] 		Dataset#14 (dataset_parser=mmeb): VisualNews_i2t, num_rows=100000, prob=1.9230769230769231
[2026-05-21 16:05:38,880] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:38,891] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:38,900] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:38,953] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:38,980] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,054] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,064] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,152] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,159] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:39,240] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,246] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:39,327] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/WebQA?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:39,329] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,417] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:39,425] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:39,425] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MSCOCO_t2i?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:39,433] INFO [src.utils.basic_utils:19] Loaded mmeb/MSCOCO_t2i dataset with 100000 samples
[2026-05-21 16:05:39,433] INFO [src.utils.basic_utils:19] 		Dataset#15 (dataset_parser=mmeb): MSCOCO_t2i, num_rows=100000, prob=1.9230769230769231
[2026-05-21 16:05:39,503] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,517] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:39,525] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:39,576] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,604] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,672] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,675] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,766] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:39,773] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,867] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,869] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:39,954] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:39,956] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MSCOCO?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:40,040] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MSCOCO_i2t?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:40,049] INFO [src.utils.basic_utils:19] Loaded mmeb/MSCOCO_i2t dataset with 113287 samples
[2026-05-21 16:05:40,049] INFO [src.utils.basic_utils:19] 		Dataset#16 (dataset_parser=mmeb): MSCOCO_i2t, num_rows=113287, prob=1.9230769230769231
[2026-05-21 16:05:40,081] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/vidore/colpali_train_set/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:40,090] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/vidore/colpali_train_set/e13d3594064836f7fd69fad7e3d2b51065b335c7/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:40,133] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:40,141] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:40,172] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/vidore/colpali_train_set/resolve/e13d3594064836f7fd69fad7e3d2b51065b335c7/colpali_train_set.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:40,218] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:40,245] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/vidore/colpali_train_set/vidore/colpali_train_set.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:40,289] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:40,377] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/vidore/colpali_train_set/revision/e13d3594064836f7fd69fad7e3d2b51065b335c7 "HTTP/1.1 200 OK"
[2026-05-21 16:05:40,385] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:40,458] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/vidore/colpali_train_set/resolve/e13d3594064836f7fd69fad7e3d2b51065b335c7/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:40,480] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:40,551] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=vidore/colpali_train_set "HTTP/1.1 200 OK"
[2026-05-21 16:05:40,569] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:40,642] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/vidore/colpali_train_set/tree/e13d3594064836f7fd69fad7e3d2b51065b335c7/data?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:40,659] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/NIGHTS?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:40,666] INFO [src.utils.basic_utils:19] Loaded mmeb/NIGHTS dataset with 15941 samples
[2026-05-21 16:05:40,667] INFO [src.utils.basic_utils:19] 		Dataset#17 (dataset_parser=mmeb): NIGHTS, num_rows=15941, prob=0.9615384615384616
[2026-05-21 16:05:40,729] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/vidore/colpali_train_set/tree/e13d3594064836f7fd69fad7e3d2b51065b335c7?recursive=false&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:40,746] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:40,754] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:40,812] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/vidore/colpali_train_set/resolve/e13d3594064836f7fd69fad7e3d2b51065b335c7/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:40,839] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:40,911] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:40,978] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:40,986] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:41,007] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:41,101] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:41,128] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/resolve/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2/VisRAG-Ret-Train-In-domain-data.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:41,181] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:41,200] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/openbmb/VisRAG-Ret-Train-In-domain-data.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:41,268] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/WebQA?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:41,276] INFO [src.utils.basic_utils:19] Loaded mmeb/WebQA dataset with 17166 samples
[2026-05-21 16:05:41,276] INFO [src.utils.basic_utils:19] 		Dataset#18 (dataset_parser=mmeb): WebQA, num_rows=17166, prob=0.9615384615384616
[2026-05-21 16:05:41,277] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/revision/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2 "HTTP/1.1 200 OK"
[2026-05-21 16:05:41,359] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:41,362] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/resolve/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:41,367] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/TIGER-Lab/MMEB-train/76dd0a440b6d4c02776830a804443fffbb2d0bfa/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:41,447] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:41,453] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=openbmb/VisRAG-Ret-Train-In-domain-data "HTTP/1.1 200 OK"
[2026-05-21 16:05:41,519] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/TIGER-Lab/MMEB-train/TIGER-Lab/MMEB-train.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:41,547] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/tree/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2/data?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:41,615] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:41,630] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/tree/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2?recursive=false&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:41,714] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=TIGER-Lab/MMEB-train "HTTP/1.1 200 OK"
[2026-05-21 16:05:41,718] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/resolve/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:41,797] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/TIGER-Lab/MMEB-train/resolve/76dd0a440b6d4c02776830a804443fffbb2d0bfa/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:41,853] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/json/json.py "HTTP/1.1 200 OK"
[2026-05-21 16:05:41,882] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/TIGER-Lab/MMEB-train/tree/76dd0a440b6d4c02776830a804443fffbb2d0bfa/MSCOCO?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:41,890] INFO [src.utils.basic_utils:19] Loaded mmeb/MSCOCO dataset with 100000 samples
[2026-05-21 16:05:41,890] INFO [src.utils.basic_utils:19] 		Dataset#19 (dataset_parser=mmeb): MSCOCO, num_rows=100000, prob=1.9230769230769231
[2026-05-21 16:05:41,975] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/vidore/colpali_train_set/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:41,979] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/json/json.py "HTTP/1.1 200 OK"
[2026-05-21 16:05:41,983] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/vidore/colpali_train_set/e13d3594064836f7fd69fad7e3d2b51065b335c7/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:42,059] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/vidore/colpali_train_set/resolve/e13d3594064836f7fd69fad7e3d2b51065b335c7/colpali_train_set.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:42,063] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/json/json.py "HTTP/1.1 200 OK"
[2026-05-21 16:05:42,131] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/vidore/colpali_train_set/vidore/colpali_train_set.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:42,213] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/vidore/colpali_train_set/revision/e13d3594064836f7fd69fad7e3d2b51065b335c7 "HTTP/1.1 200 OK"
[2026-05-21 16:05:42,295] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/vidore/colpali_train_set/resolve/e13d3594064836f7fd69fad7e3d2b51065b335c7/.huggingface.yaml "HTTP/1.1 404 Not Found"
[transformers] The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'eos_token_id': 248046, 'pad_token_id': 248044}.
[2026-05-21 16:05:42,440] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=vidore/colpali_train_set "HTTP/1.1 200 OK"
[2026-05-21 16:05:42,542] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/vidore/colpali_train_set/tree/e13d3594064836f7fd69fad7e3d2b51065b335c7/data?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:42,629] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/vidore/colpali_train_set/tree/e13d3594064836f7fd69fad7e3d2b51065b335c7?recursive=false&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:42,714] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/vidore/colpali_train_set/resolve/e13d3594064836f7fd69fad7e3d2b51065b335c7/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:42,751] INFO [src.utils.basic_utils:19] 		Dataset#20 (dataset_parser=vidore): colpali_train_set, num_rows=118195, prob=19.230769230769234
[2026-05-21 16:05:42,862] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 16:05:42,870] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2/README.md "HTTP/1.1 200 OK"
[2026-05-21 16:05:42,951] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/resolve/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2/VisRAG-Ret-Train-In-domain-data.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:43,024] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/openbmb/VisRAG-Ret-Train-In-domain-data.py "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:43,105] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/revision/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2 "HTTP/1.1 200 OK"
[2026-05-21 16:05:43,191] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/resolve/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2/.huggingface.yaml "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:43,282] INFO [httpx:1025] HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=openbmb/VisRAG-Ret-Train-In-domain-data "HTTP/1.1 200 OK"
[2026-05-21 16:05:43,489] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/tree/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2/data?recursive=true&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:43,572] INFO [httpx:1025] HTTP Request: GET https://huggingface.co/api/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/tree/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2?recursive=false&expand=false "HTTP/1.1 200 OK"
[2026-05-21 16:05:43,655] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/datasets/openbmb/VisRAG-Ret-Train-In-domain-data/resolve/9c2ecd955e15151b57c6e4ae7a7bb13dc5ad09d2/dataset_infos.json "HTTP/1.1 404 Not Found"
[2026-05-21 16:05:43,679] INFO [src.utils.basic_utils:19] 		Dataset#21 (dataset_parser=visrag): visrag-indomain, num_rows=122752, prob=23.076923076923077
[2026-05-21 16:05:43,752] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/json/json.py "HTTP/1.1 200 OK"
[2026-05-21 16:05:43,762] INFO [src.utils.basic_utils:19] 		Dataset#22 (dataset_parser=llavahound_caption): video_caption_300k, num_rows=301751, prob=9.615384615384617
[2026-05-21 16:05:43,838] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/json/json.py "HTTP/1.1 200 OK"
[2026-05-21 16:05:43,847] INFO [src.utils.basic_utils:19] 		Dataset#23 (dataset_parser=llavahound_caption): video_caption_300k-video, num_rows=301751, prob=9.615384615384617
[2026-05-21 16:05:43,922] INFO [httpx:1025] HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/json/json.py "HTTP/1.1 200 OK"
[2026-05-21 16:05:43,930] INFO [src.utils.basic_utils:19] 		Dataset#24 (dataset_parser=llavahound_qa): video_qa_240k, num_rows=255000, prob=9.615384615384617
[2026-05-21 16:05:43,930] INFO [src.utils.basic_utils:19] WARNING: `interleave_batch_size` is deprecated. Please use `homogeneous_batch_size_per_device`.
[2026-05-21 16:05:43,930] INFO [src.utils.basic_utils:19] 
Initializing interleave datasets:
		world_size=2
		total num rows=2167622
		global batch size=384
		estimated num step per epoch=5644.848958333333
		homogeneous_batch_size_per_device=64.0
		interleave_batch_size (global)=128.0
[2026-05-21 16:05:43,932] INFO [src.utils.basic_utils:19] ==================================================
[2026-05-21 16:05:43,932] INFO [src.utils.basic_utils:19] Print the features of each dataset, make sure that all datasets have valid features.
[2026-05-21 16:05:43,933] INFO [src.utils.basic_utils:19] 		Dataset 0 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,933] INFO [src.utils.basic_utils:19] 		Dataset 1 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,933] INFO [src.utils.basic_utils:19] 		Dataset 2 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,933] INFO [src.utils.basic_utils:19] 		Dataset 3 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,933] INFO [src.utils.basic_utils:19] 		Dataset 4 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,933] INFO [src.utils.basic_utils:19] 		Dataset 5 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,934] INFO [src.utils.basic_utils:19] 		Dataset 6 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,934] INFO [src.utils.basic_utils:19] 		Dataset 7 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,934] INFO [src.utils.basic_utils:19] 		Dataset 8 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,934] INFO [src.utils.basic_utils:19] 		Dataset 9 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,934] INFO [src.utils.basic_utils:19] 		Dataset 10 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,934] INFO [src.utils.basic_utils:19] 		Dataset 11 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,934] INFO [src.utils.basic_utils:19] 		Dataset 12 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,935] INFO [src.utils.basic_utils:19] 		Dataset 13 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,935] INFO [src.utils.basic_utils:19] 		Dataset 14 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,935] INFO [src.utils.basic_utils:19] 		Dataset 15 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,935] INFO [src.utils.basic_utils:19] 		Dataset 16 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,935] INFO [src.utils.basic_utils:19] 		Dataset 17 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,935] INFO [src.utils.basic_utils:19] 		Dataset 18 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,935] INFO [src.utils.basic_utils:19] 		Dataset 19 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,935] INFO [src.utils.basic_utils:19] 		Dataset 20 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,936] INFO [src.utils.basic_utils:19] 		Dataset 21 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,936] INFO [src.utils.basic_utils:19] 		Dataset 22 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,936] INFO [src.utils.basic_utils:19] 		Dataset 23 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,936] INFO [src.utils.basic_utils:19] 		Dataset 24 features: ['query_text', 'query_image', 'pos_text', 'pos_image', 'neg_text', 'neg_image', 'global_dataset_name']
[2026-05-21 16:05:43,936] INFO [src.utils.basic_utils:19] ==================================================
[transformers] The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'eos_token_id': 248046, 'pad_token_id': 248044}.
[2026-05-21 16:05:44,542] INFO [src.trainer:320] ***** Running training *****
[2026-05-21 16:05:44,542] INFO [src.trainer:321]   Num examples = 1,920,000
[2026-05-21 16:05:44,542] INFO [src.trainer:322]   Num Epochs = 9,223,372,036,854,775,807
[2026-05-21 16:05:44,542] INFO [src.trainer:323]   Instantaneous batch size per device = 192
[2026-05-21 16:05:44,542] INFO [src.trainer:326]   Total train batch size (w. parallel, distributed & accumulation) = 384
[2026-05-21 16:05:44,542] INFO [src.trainer:327]   Gradient Accumulation steps = 1
[2026-05-21 16:05:44,542] INFO [src.trainer:328]   Total optimization steps = 5,000
[2026-05-21 16:05:44,543] INFO [src.trainer:329]   Number of trainable parameters = 7,526,400
[2026-05-21 16:05:44,545] INFO [src.trainer:320] ***** Running training *****
[2026-05-21 16:05:44,545] INFO [src.trainer:321]   Num examples = 1,920,000
[2026-05-21 16:05:44,545] INFO [src.trainer:322]   Num Epochs = 9,223,372,036,854,775,807
[2026-05-21 16:05:44,545] INFO [src.trainer:323]   Instantaneous batch size per device = 192
[2026-05-21 16:05:44,545] INFO [src.trainer:326]   Total train batch size (w. parallel, distributed & accumulation) = 384
[2026-05-21 16:05:44,545] INFO [src.trainer:327]   Gradient Accumulation steps = 1
[2026-05-21 16:05:44,545] INFO [src.trainer:328]   Total optimization steps = 5,000
[2026-05-21 16:05:44,547] INFO [src.trainer:329]   Number of trainable parameters = 7,526,400

  0%|          | 0/5000 [00:00<?, ?it/s]Too many dataloader workers: 4 (max is dataset.num_shards=2). Stopping 2 dataloader workers.
[2026-05-21 16:05:44,740] WARNING [datasets.iterable_dataset:2559] Too many dataloader workers: 4 (max is dataset.num_shards=2). Stopping 2 dataloader workers.
[transformers] Asked to sample `fps` frames per second but no video metadata was provided which is required when sampling with `fps`. Defaulting to `fps=24`. Please provide `video_metadata` for more accurate results.
[transformers] Asked to sample `fps` frames per second but no video metadata was provided which is required when sampling with `fps`. Defaulting to `fps=24`. Please provide `video_metadata` for more accurate results.
[transformers] Asked to sample `fps` frames per second but no video metadata was provided which is required when sampling with `fps`. Defaulting to `fps=24`. Please provide `video_metadata` for more accurate results.
[transformers] Asked to sample `fps` frames per second but no video metadata was provided which is required when sampling with `fps`. Defaulting to `fps=24`. Please provide `video_metadata` for more accurate results.
[2026-05-21 16:06:08,977] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])
[2026-05-21 16:06:09,402] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])
[2026-05-21 16:06:13,904] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])
[2026-05-21 16:06:13,917] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])
[2026-05-21 16:06:15,043] INFO [src.utils.basic_utils:10] rank0: dataset name: {'vidore/vidore/colpali_train_set', 'llavahound_caption/video_caption_300k'}
[2026-05-21 16:06:15,043] INFO [src.utils.basic_utils:10] rank0: 		dataset_name=llavahound_caption/video_caption_300k, count=128
[2026-05-21 16:06:15,043] INFO [src.utils.basic_utils:10] rank0: 		dataset_name=vidore/vidore/colpali_train_set, count=64
[2026-05-21 16:06:16,428] INFO [src.utils.basic_utils:10] rank1: dataset name: {'llavahound_caption/video_caption_300k', 'vidore/vidore/colpali_train_set'}
[2026-05-21 16:06:16,428] INFO [src.utils.basic_utils:10] rank1: 		dataset_name=llavahound_caption/video_caption_300k, count=128
[2026-05-21 16:06:16,428] INFO [src.utils.basic_utils:10] rank1: 		dataset_name=vidore/vidore/colpali_train_set, count=64
[2026-05-21 16:06:21,278] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 16:06:22,263] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 16:06:28,704] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])
[2026-05-21 16:06:29,969] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 16:06:37,485] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])
[2026-05-21 16:06:44,594] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 973])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 16:06:46,070] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 60])
[2026-05-21 16:06:47,080] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 969])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 16:06:48,635] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 16:06:53,368] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 58])

  0%|          | 1/5000 [01:35<132:48:59, 95.65s/it]
                                                    
{'loss': '0.7037', 'grad_norm': '19.87', 'learning_rate': '2e-07', 'epoch': '0.0002'}

  0%|          | 1/5000 [01:35<132:48:59, 95.65s/it][2026-05-21 16:07:20,324] INFO [src.utils.basic_utils:10] rank1: dataset name: {'llavahound_caption/video_caption_300k', 'vidore/vidore/colpali_train_set'}
[2026-05-21 16:07:20,324] INFO [src.utils.basic_utils:10] rank1: 		dataset_name=llavahound_caption/video_caption_300k, count=128
[2026-05-21 16:07:20,324] INFO [src.utils.basic_utils:10] rank1: 		dataset_name=vidore/vidore/colpali_train_set, count=64
[2026-05-21 16:07:20,327] INFO [src.utils.basic_utils:10] rank0: dataset name: {'vidore/vidore/colpali_train_set', 'llavahound_caption/video_caption_300k'}
[2026-05-21 16:07:20,327] INFO [src.utils.basic_utils:10] rank0: 		dataset_name=llavahound_caption/video_caption_300k, count=128
[2026-05-21 16:07:20,327] INFO [src.utils.basic_utils:10] rank0: 		dataset_name=vidore/vidore/colpali_train_set, count=64
[2026-05-21 16:07:23,287] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 82])
[2026-05-21 16:07:25,351] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 148])

  0%|          | 2/5000 [02:25<95:21:19, 68.68s/it] 
                                                   
{'loss': '0.4115', 'grad_norm': '17.66', 'learning_rate': '4e-07', 'epoch': '0.0004'}

  0%|          | 2/5000 [02:25<95:21:19, 68.68s/it][2026-05-21 16:08:10,102] INFO [src.utils.basic_utils:10] rank0: dataset name: {'llavahound_caption/video_caption_300k', 'vidore/vidore/colpali_train_set'}
[2026-05-21 16:08:10,102] INFO [src.utils.basic_utils:10] rank0: 		dataset_name=llavahound_caption/video_caption_300k, count=128
[2026-05-21 16:08:10,103] INFO [src.utils.basic_utils:10] rank0: 		dataset_name=vidore/vidore/colpali_train_set, count=64
[2026-05-21 16:08:10,131] INFO [src.utils.basic_utils:10] rank1: dataset name: {'vidore/vidore/colpali_train_set', 'llavahound_caption/video_caption_300k'}
[2026-05-21 16:08:10,131] INFO [src.utils.basic_utils:10] rank1: 		dataset_name=llavahound_caption/video_caption_300k, count=128
[2026-05-21 16:08:10,131] INFO [src.utils.basic_utils:10] rank1: 		dataset_name=vidore/vidore/colpali_train_set, count=64
[2026-05-21 16:08:12,567] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 406])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 245])
[2026-05-21 16:08:13,114] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 336])

  0%|          | 3/5000 [03:15<83:20:44, 60.04s/it]
                                                   
{'loss': '0.7318', 'grad_norm': '27.91', 'learning_rate': '6e-07', 'epoch': '0.0006'}

  0%|          | 3/5000 [03:15<83:20:44, 60.04s/it][2026-05-21 16:08:59,875] INFO [src.utils.basic_utils:10] rank1: dataset name: {'vidore/vidore/colpali_train_set', 'llavahound_caption/video_caption_300k'}
[2026-05-21 16:08:59,875] INFO [src.utils.basic_utils:10] rank1: 		dataset_name=llavahound_caption/video_caption_300k, count=128
[2026-05-21 16:08:59,875] INFO [src.utils.basic_utils:10] rank1: 		dataset_name=vidore/vidore/colpali_train_set, count=64
[2026-05-21 16:08:59,876] INFO [src.utils.basic_utils:10] rank0: dataset name: {'llavahound_caption/video_caption_300k', 'vidore/vidore/colpali_train_set'}
[2026-05-21 16:08:59,876] INFO [src.utils.basic_utils:10] rank0: 		dataset_name=llavahound_caption/video_caption_300k, count=128
[2026-05-21 16:08:59,876] INFO [src.utils.basic_utils:10] rank0: 		dataset_name=vidore/vidore/colpali_train_set, count=64
[2026-05-21 16:09:03,010] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 313])
[2026-05-21 16:09:03,014] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 281])

  0%|          | 4/5000 [04:07<78:50:04, 56.81s/it]
                                                   
{'loss': '0.4671', 'grad_norm': '23.79', 'learning_rate': '8e-07', 'epoch': '0.0008'}

  0%|          | 4/5000 [04:07<78:50:04, 56.81s/it][2026-05-21 16:09:51,708] INFO [src.utils.basic_utils:10] rank1: dataset name: {'mmeb/DocVQA', 'VisRAG-Indomain-data'}
[2026-05-21 16:09:51,708] INFO [src.utils.basic_utils:10] rank1: 		dataset_name=VisRAG-Indomain-data, count=128
[2026-05-21 16:09:51,708] INFO [src.utils.basic_utils:10] rank1: 		dataset_name=mmeb/DocVQA, count=64
[2026-05-21 16:09:51,709] INFO [src.utils.basic_utils:10] rank0: dataset name: {'mmeb/DocVQA', 'VisRAG-Indomain-data'}
[2026-05-21 16:09:51,709] INFO [src.utils.basic_utils:10] rank0: 		dataset_name=VisRAG-Indomain-data, count=128
[2026-05-21 16:09:51,710] INFO [src.utils.basic_utils:10] rank0: 		dataset_name=mmeb/DocVQA, count=64
[2026-05-21 16:09:54,086] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 497])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 307])
[2026-05-21 16:09:54,476] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 758])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 252])

  0%|          | 5/5000 [05:17<85:51:37, 61.88s/it]
                                                   
{'loss': '8.388', 'grad_norm': '736.3', 'learning_rate': '1e-06', 'epoch': '0.001'}

  0%|          | 5/5000 [05:17<85:51:37, 61.88s/it][2026-05-21 16:11:04,803] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 306])
[2026-05-21 16:11:04,846] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 256])

  0%|          | 6/5000 [06:26<89:05:12, 64.22s/it]
                                                   
{'loss': '7.923', 'grad_norm': '483', 'learning_rate': '1.2e-06', 'epoch': '0.0012'}

  0%|          | 6/5000 [06:26<89:05:12, 64.22s/it][2026-05-21 16:12:26,012] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 168])
[2026-05-21 16:12:27,814] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 178])

  0%|          | 7/5000 [07:13<81:06:53, 58.48s/it]
                                                   
{'loss': '0.5545', 'grad_norm': '26.42', 'learning_rate': '1.4e-06', 'epoch': '0.0014'}

  0%|          | 7/5000 [07:13<81:06:53, 58.48s/it][2026-05-21 16:13:13,798] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 233])
[2026-05-21 16:13:13,834] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 162])

  0%|          | 8/5000 [08:01<76:23:17, 55.09s/it]
                                                   
{'loss': '0.5839', 'grad_norm': '25.41', 'learning_rate': '1.6e-06', 'epoch': '0.0016'}

  0%|          | 8/5000 [08:01<76:23:17, 55.09s/it][2026-05-21 16:13:58,167] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 121])
[2026-05-21 16:13:59,273] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 104])

  0%|          | 9/5000 [08:31<65:34:54, 47.30s/it]
                                                   
{'loss': '0.9057', 'grad_norm': '23.16', 'learning_rate': '1.8e-06', 'epoch': '0.0018'}

  0%|          | 9/5000 [08:31<65:34:54, 47.30s/it][2026-05-21 16:14:27,159] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 372])
[2026-05-21 16:14:28,899] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 131])

  0%|          | 10/5000 [09:01<58:07:36, 41.94s/it]
                                                    
{'loss': '0.3041', 'grad_norm': '18.43', 'learning_rate': '2e-06', 'epoch': '0.002'}

  0%|          | 10/5000 [09:01<58:07:36, 41.94s/it][2026-05-21 16:14:52,528] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 245])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 16:14:54,131] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 241])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  0%|          | 11/5000 [09:40<56:45:21, 40.95s/it]
                                                    
{'loss': '0.8652', 'grad_norm': '15.4', 'learning_rate': '2.2e-06', 'epoch': '0.0022'}

  0%|          | 11/5000 [09:40<56:45:21, 40.95s/it][2026-05-21 16:15:30,226] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 267])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])
[2026-05-21 16:15:30,979] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 242])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

  0%|          | 12/5000 [10:09<51:47:42, 37.38s/it]
                                                    
{'loss': '0.7416', 'grad_norm': '11.99', 'learning_rate': '2.4e-06', 'epoch': '0.0024'}

  0%|          | 12/5000 [10:09<51:47:42, 37.38s/it][2026-05-21 16:15:59,936] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 16:16:00,322] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  0%|          | 13/5000 [11:03<58:43:34, 42.39s/it]
                                                    
{'loss': '1.23', 'grad_norm': '29.12', 'learning_rate': '2.6e-06', 'epoch': '0.0026'}

  0%|          | 13/5000 [11:03<58:43:34, 42.39s/it][2026-05-21 16:16:53,969] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 16:16:54,235] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1021])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  0%|          | 14/5000 [11:55<63:02:50, 45.52s/it]
                                                    
{'loss': '1.106', 'grad_norm': '27.74', 'learning_rate': '2.8e-06', 'epoch': '0.0028'}

  0%|          | 14/5000 [11:55<63:02:50, 45.52s/it][2026-05-21 16:17:50,018] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 95])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 16:17:52,437] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 90])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  0%|          | 15/5000 [12:40<62:50:33, 45.38s/it]
                                                    
{'loss': '0.6906', 'grad_norm': '26.51', 'learning_rate': '3e-06', 'epoch': '0.003'}

  0%|          | 15/5000 [12:40<62:50:33, 45.38s/it][2026-05-21 16:18:37,639] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 116])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 16:18:38,225] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 119])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  0%|          | 16/5000 [13:28<63:51:52, 46.13s/it]
                                                    
{'loss': '0.6708', 'grad_norm': '29.35', 'learning_rate': '3.2e-06', 'epoch': '0.0032'}

  0%|          | 16/5000 [13:28<63:51:52, 46.13s/it][2026-05-21 16:19:19,217] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 16:19:20,485] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  0%|          | 17/5000 [14:17<64:57:00, 46.92s/it]
                                                    
{'loss': '5.405', 'grad_norm': '262', 'learning_rate': '3.4e-06', 'epoch': '0.0034'}

  0%|          | 17/5000 [14:17<64:57:00, 46.92s/it][2026-05-21 16:20:07,731] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 16:20:08,308] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  0%|          | 18/5000 [15:06<65:51:50, 47.59s/it]
                                                    
{'loss': '5.417', 'grad_norm': '203.6', 'learning_rate': '3.6e-06', 'epoch': '0.0036'}

  0%|          | 18/5000 [15:06<65:51:50, 47.59s/it][2026-05-21 16:20:54,541] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 317])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 16:20:54,576] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 346])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])

  0%|          | 19/5000 [15:57<67:02:39, 48.46s/it]
                                                    
{'loss': '0.6377', 'grad_norm': '17.13', 'learning_rate': '3.8e-06', 'epoch': '0.0038'}

  0%|          | 19/5000 [15:57<67:02:39, 48.46s/it][2026-05-21 16:21:45,176] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 266])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 16:21:46,219] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 323])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  0%|          | 20/5000 [16:49<68:36:07, 49.59s/it]
                                                    
{'loss': '0.6639', 'grad_norm': '18.26', 'learning_rate': '4e-06', 'epoch': '0.004'}

  0%|          | 20/5000 [16:49<68:36:07, 49.59s/it][2026-05-21 16:22:37,864] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 16:22:38,959] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  0%|          | 21/5000 [17:33<66:29:15, 48.07s/it]
                                                    
{'loss': '2.499', 'grad_norm': '225', 'learning_rate': '4.2e-06', 'epoch': '0.0042'}

  0%|          | 21/5000 [17:34<66:29:15, 48.07s/it][2026-05-21 16:23:23,662] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-21 16:23:23,829] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  0%|          | 22/5000 [18:18<64:48:47, 46.87s/it]
                                                    
{'loss': '2.371', 'grad_norm': '186.4', 'learning_rate': '4.4e-06', 'epoch': '0.0044'}

  0%|          | 22/5000 [18:18<64:48:47, 46.87s/it][2026-05-21 16:24:16,994] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1045])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])
[2026-05-21 16:24:18,178] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1014])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 174])

  0%|          | 23/5000 [19:25<73:26:13, 53.12s/it]
                                                    
{'loss': '6.066', 'grad_norm': '301.3', 'learning_rate': '4.6e-06', 'epoch': '0.0046'}

  0%|          | 23/5000 [19:25<73:26:13, 53.12s/it][2026-05-21 16:25:22,803] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 190])
[2026-05-21 16:25:24,696] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 189])

  0%|          | 24/5000 [20:34<79:47:58, 57.73s/it]
                                                    
{'loss': '4.96', 'grad_norm': '124', 'learning_rate': '4.8e-06', 'epoch': '0.0048'}

  0%|          | 24/5000 [20:34<79:47:58, 57.73s/it][2026-05-21 16:26:21,342] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 416])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 16:26:21,785] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 307])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  0%|          | 25/5000 [21:18<74:20:59, 53.80s/it]
                                                    
{'loss': '3.232', 'grad_norm': '191.3', 'learning_rate': '5e-06', 'epoch': '0.005'}

  0%|          | 25/5000 [21:18<74:20:59, 53.80s/it][2026-05-21 16:27:06,315] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 303])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-21 16:27:06,898] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 504])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

  1%|          | 26/5000 [22:06<71:39:32, 51.86s/it]
                                                    
{'loss': '2.027', 'grad_norm': '102.7', 'learning_rate': '5.2e-06', 'epoch': '0.0052'}

  1%|          | 26/5000 [22:06<71:39:32, 51.86s/it][2026-05-21 16:27:53,210] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 288])
[2026-05-21 16:27:53,604] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 252])

  1%|          | 27/5000 [22:54<70:05:38, 50.74s/it]
                                                    
{'loss': '3.102', 'grad_norm': '185.4', 'learning_rate': '5.4e-06', 'epoch': '0.0054'}

  1%|          | 27/5000 [22:54<70:05:38, 50.74s/it][2026-05-21 16:28:41,681] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 302])
[2026-05-21 16:28:41,704] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 302])

  1%|          | 28/5000 [23:43<69:35:20, 50.39s/it]
                                                    
{'loss': '2.033', 'grad_norm': '115', 'learning_rate': '5.6e-06', 'epoch': '0.0056'}

  1%|          | 28/5000 [23:43<69:35:20, 50.39s/it][2026-05-21 16:29:34,489] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-21 16:29:34,530] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 188])

  1%|          | 29/5000 [24:36<70:23:32, 50.98s/it]
                                                    
{'loss': '0.6852', 'grad_norm': '19.5', 'learning_rate': '5.8e-06', 'epoch': '0.0058'}

  1%|          | 29/5000 [24:36<70:23:32, 50.98s/it][2026-05-21 16:30:26,692] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 138])
[2026-05-21 16:30:26,876] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 697])

  1%|          | 30/5000 [25:25<69:41:30, 50.48s/it]
                                                    
{'loss': '0.616', 'grad_norm': '16.31', 'learning_rate': '6e-06', 'epoch': '0.006'}

  1%|          | 30/5000 [25:25<69:41:30, 50.48s/it][2026-05-21 16:31:19,788] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 593])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 16:31:22,968] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

  1%|          | 31/5000 [25:55<60:58:52, 44.18s/it]
                                                    
{'loss': '0.8549', 'grad_norm': '34.96', 'learning_rate': '6.2e-06', 'epoch': '0.0062'}

  1%|          | 31/5000 [25:55<60:58:52, 44.18s/it][2026-05-21 16:31:49,481] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 16:31:51,412] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

  1%|          | 32/5000 [26:26<55:49:24, 40.45s/it]
                                                    
{'loss': '0.4396', 'grad_norm': '28.09', 'learning_rate': '6.4e-06', 'epoch': '0.0064'}

  1%|          | 32/5000 [26:26<55:49:24, 40.45s/it][2026-05-21 16:32:16,437] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])
[2026-05-21 16:32:17,185] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 264])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  1%|          | 33/5000 [26:56<51:29:16, 37.32s/it]
                                                    
{'loss': '0.682', 'grad_norm': '15.48', 'learning_rate': '6.6e-06', 'epoch': '0.0066'}

  1%|          | 33/5000 [26:56<51:29:16, 37.32s/it][2026-05-21 16:32:45,692] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 242])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 16:32:45,967] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 250])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  1%|          | 34/5000 [27:26<48:14:22, 34.97s/it]
                                                    
{'loss': '0.2641', 'grad_norm': '11.61', 'learning_rate': '6.8e-06', 'epoch': '0.0068'}

  1%|          | 34/5000 [27:26<48:14:22, 34.97s/it][2026-05-21 16:33:13,865] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 248])
[2026-05-21 16:33:15,287] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 278])

  1%|          | 35/5000 [28:13<53:06:20, 38.51s/it]
                                                    
{'loss': '0.6591', 'grad_norm': '16.29', 'learning_rate': '7e-06', 'epoch': '0.007'}

  1%|          | 35/5000 [28:13<53:06:20, 38.51s/it][2026-05-21 16:34:01,469] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 324])
[2026-05-21 16:34:01,750] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 508])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 253])

  1%|          | 36/5000 [29:10<60:48:23, 44.10s/it]
                                                    
{'loss': '0.5615', 'grad_norm': '16.29', 'learning_rate': '7.2e-06', 'epoch': '0.0072'}

  1%|          | 36/5000 [29:10<60:48:23, 44.10s/it][2026-05-21 16:35:01,168] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 226])
[2026-05-21 16:35:01,201] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 189])

  1%|          | 37/5000 [29:47<57:49:00, 41.94s/it]
                                                    
{'loss': '3.654', 'grad_norm': '9.54', 'learning_rate': '7.4e-06', 'epoch': '0.0074'}

  1%|          | 37/5000 [29:47<57:49:00, 41.94s/it][2026-05-21 16:35:37,892] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 98])
[2026-05-21 16:35:38,384] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 359])

  1%|          | 38/5000 [30:24<56:05:52, 40.70s/it]
                                                    
{'loss': '3.608', 'grad_norm': '8.213', 'learning_rate': '7.6e-06', 'epoch': '0.0076'}

  1%|          | 38/5000 [30:24<56:05:52, 40.70s/it][2026-05-21 16:36:42,558] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 249])
[2026-05-21 16:36:47,434] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 171])

  1%|          | 39/5000 [31:14<59:37:48, 43.27s/it]
                                                    
{'loss': '2.082', 'grad_norm': '110.3', 'learning_rate': '7.8e-06', 'epoch': '0.0078'}

  1%|          | 39/5000 [31:14<59:37:48, 43.27s/it][2026-05-21 16:37:37,919] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 206])
[2026-05-21 16:37:40,779] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 256])

  1%|          | 40/5000 [32:06<63:15:40, 45.92s/it]
                                                    
{'loss': '1.889', 'grad_norm': '54.91', 'learning_rate': '8e-06', 'epoch': '0.008'}

  1%|          | 40/5000 [32:06<63:15:40, 45.92s/it][2026-05-21 16:38:02,080] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-21 16:38:07,089] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])

  1%|          | 41/5000 [32:35<56:31:13, 41.03s/it]
                                                    
{'loss': '0.8239', 'grad_norm': '14.8', 'learning_rate': '8.2e-06', 'epoch': '0.0082'}

  1%|          | 41/5000 [32:35<56:31:13, 41.03s/it][2026-05-21 16:38:31,922] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 129])
[2026-05-21 16:38:33,331] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 215])

  1%|          | 42/5000 [33:07<52:37:57, 38.22s/it]
                                                    
{'loss': '0.2301', 'grad_norm': '11.74', 'learning_rate': '8.4e-06', 'epoch': '0.0084'}

  1%|          | 42/5000 [33:07<52:37:57, 38.22s/it][2026-05-21 16:38:56,561] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 69])
[2026-05-21 16:38:57,491] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1017])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 99])

  1%|          | 43/5000 [33:58<57:44:32, 41.94s/it]
                                                    
{'loss': '0.4243', 'grad_norm': '12.64', 'learning_rate': '8.6e-06', 'epoch': '0.0086'}

  1%|          | 43/5000 [33:58<57:44:32, 41.94s/it][2026-05-21 16:39:46,898] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 170])
[2026-05-21 16:39:47,006] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 102])

  1%|          | 44/5000 [34:47<60:51:24, 44.21s/it]
                                                    
{'loss': '0.5195', 'grad_norm': '14.88', 'learning_rate': '8.8e-06', 'epoch': '0.0088'}

  1%|          | 44/5000 [34:47<60:51:24, 44.21s/it][2026-05-21 16:40:45,076] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 215])
[2026-05-21 16:40:46,077] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

  1%|          | 45/5000 [35:36<62:43:30, 45.57s/it]
                                                    
{'loss': '0.7477', 'grad_norm': '16.37', 'learning_rate': '9e-06', 'epoch': '0.009'}

  1%|          | 45/5000 [35:36<62:43:30, 45.57s/it][2026-05-21 16:41:31,401] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-21 16:41:34,132] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 272])

  1%|          | 46/5000 [36:25<64:19:54, 46.75s/it]
                                                    
{'loss': '0.6566', 'grad_norm': '15.31', 'learning_rate': '9.2e-06', 'epoch': '0.0092'}

  1%|          | 46/5000 [36:25<64:19:54, 46.75s/it][2026-05-21 16:42:18,011] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 16:42:20,330] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  1%|          | 47/5000 [37:14<64:53:33, 47.17s/it]
                                                    
{'loss': '0.7285', 'grad_norm': '15.3', 'learning_rate': '9.4e-06', 'epoch': '0.0094'}

  1%|          | 47/5000 [37:14<64:53:33, 47.17s/it][2026-05-21 16:43:05,748] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-21 16:43:07,092] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])

  1%|          | 48/5000 [38:04<66:03:41, 48.03s/it]
                                                    
{'loss': '0.7699', 'grad_norm': '15.99', 'learning_rate': '9.6e-06', 'epoch': '0.0096'}

  1%|          | 48/5000 [38:04<66:03:41, 48.03s/it][2026-05-21 16:44:02,758] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 16:44:03,360] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  1%|          | 49/5000 [38:51<65:52:24, 47.90s/it]
                                                    
{'loss': '0.9521', 'grad_norm': '15.66', 'learning_rate': '9.8e-06', 'epoch': '0.0098'}

  1%|          | 49/5000 [38:51<65:52:24, 47.90s/it][2026-05-21 16:44:47,983] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 16:44:49,019] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  1%|          | 50/5000 [39:38<65:18:38, 47.50s/it]
                                                    
{'loss': '0.8631', 'grad_norm': '12.9', 'learning_rate': '1e-05', 'epoch': '0.01'}

  1%|          | 50/5000 [39:38<65:18:38, 47.50s/it][2026-05-21 16:45:37,337] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 16:45:44,056] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

  1%|          | 51/5000 [40:25<65:18:28, 47.51s/it]
                                                    
{'loss': '0.4236', 'grad_norm': '10.57', 'learning_rate': '1.02e-05', 'epoch': '0.0102'}

  1%|          | 51/5000 [40:25<65:18:28, 47.51s/it][2026-05-21 16:46:25,278] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 16:46:26,719] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  1%|          | 52/5000 [41:16<66:24:43, 48.32s/it]
                                                    
{'loss': '0.3695', 'grad_norm': '9.604', 'learning_rate': '1.04e-05', 'epoch': '0.0104'}

  1%|          | 52/5000 [41:16<66:24:43, 48.32s/it][2026-05-21 16:47:11,819] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 16:47:15,510] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

  1%|          | 53/5000 [42:29<76:35:47, 55.74s/it]
                                                    
{'loss': '1.834', 'grad_norm': '43.98', 'learning_rate': '1.06e-05', 'epoch': '0.0106'}

  1%|          | 53/5000 [42:29<76:35:47, 55.74s/it][2026-05-21 16:48:26,620] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 16:48:26,654] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

  1%|          | 54/5000 [43:38<82:24:19, 59.98s/it]
                                                    
{'loss': '1.318', 'grad_norm': '23.9', 'learning_rate': '1.08e-05', 'epoch': '0.0108'}

  1%|          | 54/5000 [43:38<82:24:19, 59.98s/it][2026-05-21 16:49:28,141] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 283])
[2026-05-21 16:49:28,606] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 228])

  1%|          | 55/5000 [44:31<79:18:46, 57.74s/it]
                                                    
{'loss': '0.5442', 'grad_norm': '13.77', 'learning_rate': '1.1e-05', 'epoch': '0.011'}

  1%|          | 55/5000 [44:31<79:18:46, 57.74s/it][2026-05-21 16:50:21,099] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 252])
[2026-05-21 16:50:21,229] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])

  1%|          | 56/5000 [45:21<76:19:29, 55.58s/it]
                                                    
{'loss': '0.1449', 'grad_norm': '6.416', 'learning_rate': '1.12e-05', 'epoch': '0.0112'}

  1%|          | 56/5000 [45:21<76:19:29, 55.58s/it][2026-05-21 16:51:19,938] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 163])
[2026-05-21 16:51:20,468] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])

  1%|          | 57/5000 [46:32<82:21:19, 59.98s/it]
                                                    
{'loss': '1.673', 'grad_norm': '19.59', 'learning_rate': '1.14e-05', 'epoch': '0.0114'}

  1%|          | 57/5000 [46:32<82:21:19, 59.98s/it][2026-05-21 16:52:30,403] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])
[2026-05-21 16:52:33,776] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 93])

  1%|          | 58/5000 [47:41<86:18:09, 62.87s/it]
                                                    
{'loss': '1.701', 'grad_norm': '29.3', 'learning_rate': '1.16e-05', 'epoch': '0.0116'}

  1%|          | 58/5000 [47:41<86:18:09, 62.87s/it][2026-05-21 16:53:30,332] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 16:53:33,485] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  1%|          | 59/5000 [48:33<81:52:04, 59.65s/it]
                                                    
{'loss': '0.6807', 'grad_norm': '10.34', 'learning_rate': '1.18e-05', 'epoch': '0.0118'}

  1%|          | 59/5000 [48:33<81:52:04, 59.65s/it][2026-05-21 16:54:23,825] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 16:54:24,212] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  1%|          | 60/5000 [49:24<78:12:10, 56.99s/it]
                                                    
{'loss': '0.2127', 'grad_norm': '9.932', 'learning_rate': '1.2e-05', 'epoch': '0.012'}

  1%|          | 60/5000 [49:24<78:12:10, 56.99s/it][2026-05-21 16:55:11,242] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 216])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 115])
[2026-05-21 16:55:11,328] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 792])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 164])

  1%|          | 61/5000 [50:10<73:44:22, 53.75s/it]
                                                    
{'loss': '0.5443', 'grad_norm': '14.37', 'learning_rate': '1.22e-05', 'epoch': '0.0122'}

  1%|          | 61/5000 [50:10<73:44:22, 53.75s/it][2026-05-21 16:55:56,893] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 125])
[2026-05-21 16:55:57,341] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 232])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 43])

  1%|          | 62/5000 [50:57<70:49:27, 51.63s/it]
                                                    
{'loss': '0.2055', 'grad_norm': '7.415', 'learning_rate': '1.24e-05', 'epoch': '0.0124'}

  1%|          | 62/5000 [50:57<70:49:27, 51.63s/it][2026-05-21 16:56:44,538] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 833])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 225])
[2026-05-21 16:56:44,671] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 224])

  1%|▏         | 63/5000 [51:44<68:40:53, 50.08s/it]
                                                    
{'loss': '0.9356', 'grad_norm': '22.96', 'learning_rate': '1.26e-05', 'epoch': '0.0126'}

  1%|▏         | 63/5000 [51:44<68:40:53, 50.08s/it][2026-05-21 16:57:31,267] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 275])
[2026-05-21 16:57:31,290] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 958])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 210])

  1%|▏         | 64/5000 [52:31<67:22:59, 49.15s/it]
                                                    
{'loss': '2.547', 'grad_norm': '30.32', 'learning_rate': '1.28e-05', 'epoch': '0.0128'}

  1%|▏         | 64/5000 [52:31<67:22:59, 49.15s/it][2026-05-21 16:58:20,585] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])
[2026-05-21 16:58:20,873] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 412])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  1%|▏         | 65/5000 [53:32<72:22:02, 52.79s/it]
                                                    
{'loss': '1.263', 'grad_norm': '18.96', 'learning_rate': '1.3e-05', 'epoch': '0.013'}

  1%|▏         | 65/5000 [53:32<72:22:02, 52.79s/it][2026-05-21 16:59:21,789] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-21 16:59:21,930] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  1%|▏         | 66/5000 [54:20<70:28:36, 51.42s/it]
                                                    
{'loss': '1.712', 'grad_norm': '19.18', 'learning_rate': '1.32e-05', 'epoch': '0.0132'}

  1%|▏         | 66/5000 [54:20<70:28:36, 51.42s/it][2026-05-21 17:00:12,059] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 17:00:12,659] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

  1%|▏         | 67/5000 [54:58<64:56:27, 47.39s/it]
                                                    
{'loss': '0.7561', 'grad_norm': '17.46', 'learning_rate': '1.34e-05', 'epoch': '0.0134'}

  1%|▏         | 67/5000 [54:58<64:56:27, 47.39s/it][2026-05-21 17:00:49,530] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 17:00:50,583] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

  1%|▏         | 68/5000 [55:23<55:49:49, 40.75s/it]
                                                    
{'loss': '0.7125', 'grad_norm': '17.05', 'learning_rate': '1.36e-05', 'epoch': '0.0136'}

  1%|▏         | 68/5000 [55:23<55:49:49, 40.75s/it][2026-05-21 17:01:14,707] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])
[2026-05-21 17:01:14,826] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])

  1%|▏         | 69/5000 [56:10<58:24:35, 42.64s/it]
                                                    
{'loss': '0.5682', 'grad_norm': '14.66', 'learning_rate': '1.38e-05', 'epoch': '0.0138'}

  1%|▏         | 69/5000 [56:10<58:24:35, 42.64s/it][2026-05-21 17:02:01,954] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1035])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 220])
[2026-05-21 17:02:02,833] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 232])

  1%|▏         | 70/5000 [56:58<60:29:28, 44.17s/it]
                                                    
{'loss': '0.399', 'grad_norm': '11.72', 'learning_rate': '1.4e-05', 'epoch': '0.014'}

  1%|▏         | 70/5000 [56:58<60:29:28, 44.17s/it][2026-05-21 17:02:55,764] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 972])
[2026-05-21 17:02:57,336] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  1%|▏         | 71/5000 [57:53<64:41:59, 47.26s/it]
                                                    
{'loss': '1.115', 'grad_norm': '17.88', 'learning_rate': '1.42e-05', 'epoch': '0.0142'}

  1%|▏         | 71/5000 [57:53<64:41:59, 47.26s/it][2026-05-21 17:03:49,916] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 17:03:50,140] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])

  1%|▏         | 72/5000 [58:46<67:02:06, 48.97s/it]
                                                    
{'loss': '1.142', 'grad_norm': '16.16', 'learning_rate': '1.44e-05', 'epoch': '0.0144'}

  1%|▏         | 72/5000 [58:46<67:02:06, 48.97s/it][2026-05-21 17:04:37,263] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 11])
[2026-05-21 17:04:37,828] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 13])

  1%|▏         | 73/5000 [59:41<69:42:18, 50.93s/it]
                                                    
{'loss': '1.118', 'grad_norm': '17.93', 'learning_rate': '1.46e-05', 'epoch': '0.0146'}

  1%|▏         | 73/5000 [59:41<69:42:18, 50.93s/it][2026-05-21 17:05:31,865] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 12])
[2026-05-21 17:05:32,447] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 27])

  1%|▏         | 74/5000 [1:00:35<70:46:04, 51.72s/it]
                                                      
{'loss': '1.063', 'grad_norm': '14.93', 'learning_rate': '1.48e-05', 'epoch': '0.0148'}

  1%|▏         | 74/5000 [1:00:35<70:46:04, 51.72s/it][2026-05-21 17:06:26,707] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 17:06:27,621] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  2%|▏         | 75/5000 [1:01:29<71:54:02, 52.56s/it]
                                                      
{'loss': '0.5312', 'grad_norm': '12.82', 'learning_rate': '1.5e-05', 'epoch': '0.015'}

  2%|▏         | 75/5000 [1:01:29<71:54:02, 52.56s/it][2026-05-21 17:07:21,548] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 17:07:21,820] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  2%|▏         | 76/5000 [1:02:26<73:37:28, 53.83s/it]
                                                      
{'loss': '0.4187', 'grad_norm': '14.57', 'learning_rate': '1.52e-05', 'epoch': '0.0152'}

  2%|▏         | 76/5000 [1:02:26<73:37:28, 53.83s/it][2026-05-21 17:08:37,384] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 17:08:42,069] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])

  2%|▏         | 77/5000 [1:03:30<77:58:06, 57.02s/it]
                                                      
{'loss': '1.776', 'grad_norm': '17.81', 'learning_rate': '1.54e-05', 'epoch': '0.0154'}

  2%|▏         | 77/5000 [1:03:30<77:58:06, 57.02s/it][2026-05-21 17:09:43,890] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 17:09:44,772] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 966])

  2%|▏         | 78/5000 [1:04:21<75:21:02, 55.11s/it]
                                                      
{'loss': '1.361', 'grad_norm': '16.61', 'learning_rate': '1.56e-05', 'epoch': '0.0156'}

  2%|▏         | 78/5000 [1:04:21<75:21:02, 55.11s/it][2026-05-21 17:10:10,702] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 17:10:11,482] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 493])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  2%|▏         | 79/5000 [1:05:11<73:04:42, 53.46s/it]
                                                      
{'loss': '1.077', 'grad_norm': '20.34', 'learning_rate': '1.58e-05', 'epoch': '0.0158'}

  2%|▏         | 79/5000 [1:05:11<73:04:42, 53.46s/it][2026-05-21 17:11:00,736] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])
[2026-05-21 17:11:02,034] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

  2%|▏         | 80/5000 [1:05:59<70:53:55, 51.88s/it]
                                                      
{'loss': '1.029', 'grad_norm': '18.53', 'learning_rate': '1.6e-05', 'epoch': '0.016'}

  2%|▏         | 80/5000 [1:05:59<70:53:55, 51.88s/it][2026-05-21 17:11:50,315] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 113])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 17:11:50,512] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 94])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  2%|▏         | 81/5000 [1:07:10<78:43:21, 57.61s/it]
                                                      
{'loss': '0.9726', 'grad_norm': '14.95', 'learning_rate': '1.62e-05', 'epoch': '0.0162'}

  2%|▏         | 81/5000 [1:07:10<78:43:21, 57.61s/it][2026-05-21 17:13:01,334] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 99])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 17:13:02,950] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 98])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

  2%|▏         | 82/5000 [1:08:21<84:16:24, 61.69s/it]
                                                      
{'loss': '1.294', 'grad_norm': '16.38', 'learning_rate': '1.64e-05', 'epoch': '0.0164'}

  2%|▏         | 82/5000 [1:08:21<84:16:24, 61.69s/it][2026-05-21 17:14:19,642] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1012])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 17:14:22,621] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 958])

  2%|▏         | 83/5000 [1:09:32<88:04:47, 64.49s/it]
                                                      
{'loss': '1.214', 'grad_norm': '20.6', 'learning_rate': '1.66e-05', 'epoch': '0.0166'}

  2%|▏         | 83/5000 [1:09:32<88:04:47, 64.49s/it][2026-05-21 17:15:32,700] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 17:15:35,990] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  2%|▏         | 84/5000 [1:10:48<92:34:30, 67.79s/it]
                                                      
{'loss': '1.436', 'grad_norm': '14.73', 'learning_rate': '1.68e-05', 'epoch': '0.0168'}

  2%|▏         | 84/5000 [1:10:48<92:34:30, 67.79s/it][2026-05-21 17:16:45,535] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 236])
[2026-05-21 17:16:45,739] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 129])

  2%|▏         | 85/5000 [1:11:40<86:16:06, 63.19s/it]
                                                      
{'loss': '0.8967', 'grad_norm': '13.78', 'learning_rate': '1.7e-05', 'epoch': '0.017'}

  2%|▏         | 85/5000 [1:11:40<86:16:06, 63.19s/it][2026-05-21 17:17:37,993] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 114])
[2026-05-21 17:17:38,589] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 258])

  2%|▏         | 86/5000 [1:12:32<81:45:34, 59.90s/it]
                                                      
{'loss': '0.7025', 'grad_norm': '8.466', 'learning_rate': '1.72e-05', 'epoch': '0.0172'}

  2%|▏         | 86/5000 [1:12:32<81:45:34, 59.90s/it][2026-05-21 17:18:29,565] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 176])
[2026-05-21 17:18:32,550] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 162])

  2%|▏         | 87/5000 [1:13:22<77:35:55, 56.86s/it]
                                                      
{'loss': '1.519', 'grad_norm': '17.62', 'learning_rate': '1.74e-05', 'epoch': '0.0174'}

  2%|▏         | 87/5000 [1:13:22<77:35:55, 56.86s/it][2026-05-21 17:19:21,523] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 246])
[2026-05-21 17:19:21,904] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

  2%|▏         | 88/5000 [1:14:12<74:43:59, 54.77s/it]
                                                      
{'loss': '1.721', 'grad_norm': '15.04', 'learning_rate': '1.76e-05', 'epoch': '0.0176'}

  2%|▏         | 88/5000 [1:14:12<74:43:59, 54.77s/it][2026-05-21 17:20:03,175] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-21 17:20:03,254] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

  2%|▏         | 89/5000 [1:15:22<81:06:37, 59.46s/it]
                                                      
{'loss': '0.5645', 'grad_norm': '15.65', 'learning_rate': '1.78e-05', 'epoch': '0.0178'}

  2%|▏         | 89/5000 [1:15:22<81:06:37, 59.46s/it][2026-05-21 17:21:13,458] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 178])
[2026-05-21 17:21:13,577] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1014])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])

  2%|▏         | 90/5000 [1:16:33<85:38:58, 62.80s/it]
                                                      
{'loss': '1.489', 'grad_norm': '23.13', 'learning_rate': '1.8e-05', 'epoch': '0.018'}

  2%|▏         | 90/5000 [1:16:33<85:38:58, 62.80s/it][2026-05-21 17:22:27,906] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 293])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 17:22:29,564] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 238])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  2%|▏         | 91/5000 [1:17:21<79:28:40, 58.28s/it]
                                                      
{'loss': '2.277', 'grad_norm': '15.35', 'learning_rate': '1.82e-05', 'epoch': '0.0182'}

  2%|▏         | 91/5000 [1:17:21<79:28:40, 58.28s/it][2026-05-21 17:23:17,406] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 281])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 17:23:19,015] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 262])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

  2%|▏         | 92/5000 [1:18:08<75:08:14, 55.11s/it]
                                                      
{'loss': '2.165', 'grad_norm': '12.85', 'learning_rate': '1.84e-05', 'epoch': '0.0184'}

  2%|▏         | 92/5000 [1:18:08<75:08:14, 55.11s/it][2026-05-21 17:23:56,539] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 20])
[2026-05-21 17:23:56,897] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 20])

  2%|▏         | 93/5000 [1:18:55<71:44:12, 52.63s/it]
                                                      
{'loss': '0.4077', 'grad_norm': '12.79', 'learning_rate': '1.86e-05', 'epoch': '0.0186'}

  2%|▏         | 93/5000 [1:18:55<71:44:12, 52.63s/it][2026-05-21 17:24:43,230] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 20])
[2026-05-21 17:24:43,508] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 21])

  2%|▏         | 94/5000 [1:19:43<69:34:06, 51.05s/it]
                                                      
{'loss': '0.339', 'grad_norm': '13.45', 'learning_rate': '1.88e-05', 'epoch': '0.0188'}

  2%|▏         | 94/5000 [1:19:43<69:34:06, 51.05s/it][2026-05-21 17:25:32,418] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])
[2026-05-21 17:25:32,561] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

  2%|▏         | 95/5000 [1:20:30<68:08:18, 50.01s/it]
                                                      
{'loss': '0.3524', 'grad_norm': '8.729', 'learning_rate': '1.9e-05', 'epoch': '0.019'}

  2%|▏         | 95/5000 [1:20:30<68:08:18, 50.01s/it][2026-05-21 17:26:19,890] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 93])
[2026-05-21 17:26:19,899] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 88])

  2%|▏         | 96/5000 [1:21:18<67:07:18, 49.27s/it]
                                                      
{'loss': '0.3603', 'grad_norm': '11.62', 'learning_rate': '1.92e-05', 'epoch': '0.0192'}

  2%|▏         | 96/5000 [1:21:18<67:07:18, 49.27s/it][2026-05-21 17:27:15,820] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 17:27:17,283] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  2%|▏         | 97/5000 [1:21:49<59:43:48, 43.86s/it]
                                                      
{'loss': '0.4324', 'grad_norm': '11.83', 'learning_rate': '1.94e-05', 'epoch': '0.0194'}

  2%|▏         | 97/5000 [1:21:49<59:43:48, 43.86s/it][2026-05-21 17:27:48,466] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 970])
[2026-05-21 17:27:49,550] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  2%|▏         | 98/5000 [1:22:19<54:17:34, 39.87s/it]
                                                      
{'loss': '0.1202', 'grad_norm': '6.112', 'learning_rate': '1.96e-05', 'epoch': '0.0196'}

  2%|▏         | 98/5000 [1:22:19<54:17:34, 39.87s/it][2026-05-21 17:28:10,558] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 17:28:11,873] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 194])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  2%|▏         | 99/5000 [1:23:06<56:59:01, 41.86s/it]
                                                      
{'loss': '2.618', 'grad_norm': '8.166', 'learning_rate': '1.98e-05', 'epoch': '0.0198'}

  2%|▏         | 99/5000 [1:23:06<56:59:01, 41.86s/it][2026-05-21 17:28:57,150] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 17:28:57,538] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])

  2%|▏         | 100/5000 [1:23:52<58:37:44, 43.07s/it]
                                                       
{'loss': '2.606', 'grad_norm': '6.198', 'learning_rate': '2e-05', 'epoch': '0.02'}

  2%|▏         | 100/5000 [1:23:52<58:37:44, 43.07s/it][2026-05-21 17:29:41,082] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 17:29:42,335] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  2%|▏         | 101/5000 [1:24:41<61:04:08, 44.88s/it]
                                                       
{'loss': '1.201', 'grad_norm': '23.17', 'learning_rate': '2e-05', 'epoch': '0.0202'}

  2%|▏         | 101/5000 [1:24:41<61:04:08, 44.88s/it][2026-05-21 17:30:30,285] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 17:30:30,340] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  2%|▏         | 102/5000 [1:25:27<61:41:49, 45.35s/it]
                                                       
{'loss': '2.272', 'grad_norm': '36.45', 'learning_rate': '1.999e-05', 'epoch': '0.0204'}

  2%|▏         | 102/5000 [1:25:27<61:41:49, 45.35s/it][2026-05-21 17:31:28,354] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 17:31:28,773] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  2%|▏         | 103/5000 [1:26:37<71:23:43, 52.49s/it]
                                                       
{'loss': '1.409', 'grad_norm': '14.7', 'learning_rate': '1.999e-05', 'epoch': '0.0206'}

  2%|▏         | 103/5000 [1:26:37<71:23:43, 52.49s/it][2026-05-21 17:32:37,453] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 17:32:37,993] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  2%|▏         | 104/5000 [1:27:46<78:06:09, 57.43s/it]
                                                       
{'loss': '1.671', 'grad_norm': '38.55', 'learning_rate': '1.998e-05', 'epoch': '0.0208'}

  2%|▏         | 104/5000 [1:27:46<78:06:09, 57.43s/it][2026-05-21 17:33:33,243] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 17:33:33,400] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 289])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  2%|▏         | 105/5000 [1:28:35<74:55:29, 55.10s/it]
                                                       
{'loss': '1.035', 'grad_norm': '19.84', 'learning_rate': '1.998e-05', 'epoch': '0.021'}

  2%|▏         | 105/5000 [1:28:35<74:55:29, 55.10s/it][2026-05-21 17:34:22,895] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 287])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 17:34:23,025] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 262])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  2%|▏         | 106/5000 [1:29:27<73:23:55, 53.99s/it]
                                                       
{'loss': '1.122', 'grad_norm': '15.68', 'learning_rate': '1.998e-05', 'epoch': '0.0212'}

  2%|▏         | 106/5000 [1:29:27<73:23:55, 53.99s/it][2026-05-21 17:35:16,739] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 26])
[2026-05-21 17:35:17,003] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 13])

  2%|▏         | 107/5000 [1:30:19<72:38:01, 53.44s/it]
                                                       
{'loss': '1.142', 'grad_norm': '16.82', 'learning_rate': '1.997e-05', 'epoch': '0.0214'}

  2%|▏         | 107/5000 [1:30:19<72:38:01, 53.44s/it][2026-05-21 17:36:07,997] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 970])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 22])
[2026-05-21 17:36:08,723] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 19])

  2%|▏         | 108/5000 [1:31:09<71:18:15, 52.47s/it]
                                                       
{'loss': '1.152', 'grad_norm': '20.71', 'learning_rate': '1.997e-05', 'epoch': '0.0216'}

  2%|▏         | 108/5000 [1:31:09<71:18:15, 52.47s/it][2026-05-21 17:36:58,624] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 265])
[2026-05-21 17:36:58,664] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])

  2%|▏         | 109/5000 [1:32:00<70:43:11, 52.05s/it]
                                                       
{'loss': '0.4236', 'grad_norm': '10.31', 'learning_rate': '1.996e-05', 'epoch': '0.0218'}

  2%|▏         | 109/5000 [1:32:00<70:43:11, 52.05s/it][2026-05-21 17:37:48,610] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 317])
[2026-05-21 17:37:50,226] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 308])

  2%|▏         | 110/5000 [1:32:52<70:44:32, 52.08s/it]
                                                       
{'loss': '0.3137', 'grad_norm': '10.24', 'learning_rate': '1.996e-05', 'epoch': '0.022'}

  2%|▏         | 110/5000 [1:32:52<70:44:32, 52.08s/it][2026-05-21 17:38:50,022] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 17:38:53,434] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])

  2%|▏         | 111/5000 [1:33:23<61:52:48, 45.57s/it]
                                                       
{'loss': '0.4825', 'grad_norm': '9.284', 'learning_rate': '1.996e-05', 'epoch': '0.0222'}

  2%|▏         | 111/5000 [1:33:23<61:52:48, 45.57s/it][2026-05-21 17:39:20,116] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 17:39:22,296] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

  2%|▏         | 112/5000 [1:33:53<55:32:46, 40.91s/it]
                                                       
{'loss': '0.198', 'grad_norm': '7.034', 'learning_rate': '1.995e-05', 'epoch': '0.0224'}

  2%|▏         | 112/5000 [1:33:53<55:32:46, 40.91s/it][2026-05-21 17:39:41,416] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 281])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 17:39:41,543] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 252])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 980])

  2%|▏         | 113/5000 [1:34:39<57:44:21, 42.53s/it]
                                                       
{'loss': '2.18', 'grad_norm': '16.57', 'learning_rate': '1.995e-05', 'epoch': '0.0226'}

  2%|▏         | 113/5000 [1:34:39<57:44:21, 42.53s/it][2026-05-21 17:40:26,792] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 212])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])
[2026-05-21 17:40:29,728] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 238])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  2%|▏         | 114/5000 [1:35:27<59:55:22, 44.15s/it]
                                                       
{'loss': '2.14', 'grad_norm': '13.53', 'learning_rate': '1.994e-05', 'epoch': '0.0228'}

  2%|▏         | 114/5000 [1:35:27<59:55:22, 44.15s/it][2026-05-21 17:41:16,246] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 329])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 980])
[2026-05-21 17:41:17,466] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  2%|▏         | 115/5000 [1:36:14<61:11:12, 45.09s/it]
                                                       
{'loss': '0.4225', 'grad_norm': '11.7', 'learning_rate': '1.994e-05', 'epoch': '0.023'}

  2%|▏         | 115/5000 [1:36:14<61:11:12, 45.09s/it][2026-05-21 17:42:02,841] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 303])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 975])
[2026-05-21 17:42:03,304] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 324])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  2%|▏         | 116/5000 [1:37:01<62:04:25, 45.75s/it]
                                                       
{'loss': '0.1292', 'grad_norm': '6.959', 'learning_rate': '1.993e-05', 'epoch': '0.0232'}

  2%|▏         | 116/5000 [1:37:01<62:04:25, 45.75s/it][2026-05-21 17:42:54,990] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 17:42:56,674] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  2%|▏         | 117/5000 [1:37:54<64:59:10, 47.91s/it]
                                                       
{'loss': '0.3672', 'grad_norm': '9.093', 'learning_rate': '1.993e-05', 'epoch': '0.0234'}

  2%|▏         | 117/5000 [1:37:54<64:59:10, 47.91s/it][2026-05-21 17:43:47,063] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 17:43:49,085] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 980])

  2%|▏         | 118/5000 [1:38:51<68:27:45, 50.48s/it]
                                                       
{'loss': '0.3814', 'grad_norm': '10.36', 'learning_rate': '1.993e-05', 'epoch': '0.0236'}

  2%|▏         | 118/5000 [1:38:51<68:27:45, 50.48s/it][2026-05-21 17:45:07,643] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 216])
[2026-05-21 17:45:11,774] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 223])

  2%|▏         | 119/5000 [1:39:42<68:33:27, 50.57s/it]
                                                       
{'loss': '1.194', 'grad_norm': '13.54', 'learning_rate': '1.992e-05', 'epoch': '0.0238'}

  2%|▏         | 119/5000 [1:39:42<68:33:27, 50.57s/it][2026-05-21 17:46:04,117] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-21 17:46:06,791] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 105])

  2%|▏         | 120/5000 [1:40:31<68:04:38, 50.22s/it]
                                                       
{'loss': '0.9087', 'grad_norm': '20.52', 'learning_rate': '1.992e-05', 'epoch': '0.024'}

  2%|▏         | 120/5000 [1:40:31<68:04:38, 50.22s/it][2026-05-21 17:46:32,542] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])
[2026-05-21 17:46:32,862] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  2%|▏         | 121/5000 [1:41:28<70:37:08, 52.11s/it]
                                                       
{'loss': '0.5107', 'grad_norm': '13.43', 'learning_rate': '1.991e-05', 'epoch': '0.0242'}

  2%|▏         | 121/5000 [1:41:28<70:37:08, 52.11s/it][2026-05-21 17:47:26,953] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 17:47:28,614] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  2%|▏         | 122/5000 [1:42:19<70:18:26, 51.89s/it]
                                                       
{'loss': '0.5617', 'grad_norm': '12.8', 'learning_rate': '1.991e-05', 'epoch': '0.0244'}

  2%|▏         | 122/5000 [1:42:19<70:18:26, 51.89s/it][2026-05-21 17:48:21,454] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 124])
[2026-05-21 17:48:23,881] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 130])

  2%|▏         | 123/5000 [1:43:37<80:57:55, 59.77s/it]
                                                       
{'loss': '0.6595', 'grad_norm': '20.74', 'learning_rate': '1.991e-05', 'epoch': '0.0246'}

  2%|▏         | 123/5000 [1:43:37<80:57:55, 59.77s/it][2026-05-21 17:49:40,862] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 342])
[2026-05-21 17:49:44,484] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 157])

  2%|▏         | 124/5000 [1:44:54<87:56:15, 64.93s/it]
                                                       
{'loss': '0.7842', 'grad_norm': '14.71', 'learning_rate': '1.99e-05', 'epoch': '0.0248'}

  2%|▏         | 124/5000 [1:44:54<87:56:15, 64.93s/it][2026-05-21 17:50:43,824] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 978])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 21])
[2026-05-21 17:50:44,350] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 981])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 12])

  2%|▎         | 125/5000 [1:45:47<83:07:44, 61.39s/it]
                                                       
{'loss': '0.6272', 'grad_norm': '12.54', 'learning_rate': '1.99e-05', 'epoch': '0.025'}

  2%|▎         | 125/5000 [1:45:47<83:07:44, 61.39s/it][2026-05-21 17:51:37,423] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 21])
[2026-05-21 17:51:39,122] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 16])

  3%|▎         | 126/5000 [1:46:40<79:50:06, 58.97s/it]
                                                       
{'loss': '0.4905', 'grad_norm': '10.93', 'learning_rate': '1.989e-05', 'epoch': '0.0252'}

  3%|▎         | 126/5000 [1:46:40<79:50:06, 58.97s/it][2026-05-21 17:52:27,896] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 399])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 85])
[2026-05-21 17:52:28,327] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 323])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 74])

  3%|▎         | 127/5000 [1:47:51<84:25:10, 62.37s/it]
                                                       
{'loss': '0.9779', 'grad_norm': '14.04', 'learning_rate': '1.989e-05', 'epoch': '0.0254'}

  3%|▎         | 127/5000 [1:47:51<84:25:10, 62.37s/it][2026-05-21 17:53:38,631] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 182])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 54])
[2026-05-21 17:53:39,705] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 62])

  3%|▎         | 128/5000 [1:49:01<87:34:31, 64.71s/it]
                                                       
{'loss': '0.8366', 'grad_norm': '13.47', 'learning_rate': '1.989e-05', 'epoch': '0.0256'}

  3%|▎         | 128/5000 [1:49:01<87:34:31, 64.71s/it][2026-05-21 17:54:50,843] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 231])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 17:54:52,029] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

  3%|▎         | 129/5000 [1:49:51<81:46:13, 60.43s/it]
                                                       
{'loss': '0.418', 'grad_norm': '10.47', 'learning_rate': '1.988e-05', 'epoch': '0.0258'}

  3%|▎         | 129/5000 [1:49:51<81:46:13, 60.43s/it][2026-05-21 17:55:39,978] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 164])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 17:55:41,011] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 441])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  3%|▎         | 130/5000 [1:50:42<77:54:57, 57.60s/it]
                                                       
{'loss': '0.3708', 'grad_norm': '11.51', 'learning_rate': '1.988e-05', 'epoch': '0.026'}

  3%|▎         | 130/5000 [1:50:42<77:54:57, 57.60s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-21 17:56:34,939] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])
[2026-05-21 17:56:36,398] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 226])

  3%|▎         | 131/5000 [1:51:27<72:46:50, 53.81s/it]
                                                       
{'loss': '1.329', 'grad_norm': '17.78', 'learning_rate': '1.987e-05', 'epoch': '0.0262'}

  3%|▎         | 131/5000 [1:51:27<72:46:50, 53.81s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-21 17:57:20,106] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 196])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 228])
[2026-05-21 17:57:20,107] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 497])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 194])

  3%|▎         | 132/5000 [1:52:14<69:44:44, 51.58s/it]
                                                       
{'loss': '0.4551', 'grad_norm': '12.83', 'learning_rate': '1.987e-05', 'epoch': '0.0264'}

  3%|▎         | 132/5000 [1:52:14<69:44:44, 51.58s/it][2026-05-21 17:58:01,305] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 279])
[2026-05-21 17:58:02,737] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 492])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])

  3%|▎         | 133/5000 [1:52:40<59:20:19, 43.89s/it]
                                                       
{'loss': '0.5744', 'grad_norm': '22.55', 'learning_rate': '1.987e-05', 'epoch': '0.0266'}

  3%|▎         | 133/5000 [1:52:40<59:20:19, 43.89s/it][2026-05-21 17:58:28,185] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])
[2026-05-21 17:58:28,345] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 287])

  3%|▎         | 134/5000 [1:53:08<53:02:40, 39.24s/it]
                                                       
{'loss': '0.2053', 'grad_norm': '12.31', 'learning_rate': '1.986e-05', 'epoch': '0.0268'}

  3%|▎         | 134/5000 [1:53:08<53:02:40, 39.24s/it][2026-05-21 17:58:58,144] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])
[2026-05-21 17:58:58,766] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 275])

  3%|▎         | 135/5000 [1:53:56<56:28:48, 41.79s/it]
                                                       
{'loss': '1.019', 'grad_norm': '17.77', 'learning_rate': '1.986e-05', 'epoch': '0.027'}

  3%|▎         | 135/5000 [1:53:56<56:28:48, 41.79s/it][2026-05-21 17:59:45,923] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 252])
[2026-05-21 17:59:46,237] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 257])

  3%|▎         | 136/5000 [1:54:46<59:41:29, 44.18s/it]
                                                       
{'loss': '0.7793', 'grad_norm': '14.87', 'learning_rate': '1.985e-05', 'epoch': '0.0272'}

  3%|▎         | 136/5000 [1:54:46<59:41:29, 44.18s/it][2026-05-21 18:00:41,280] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 275])
[2026-05-21 18:00:42,821] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])

  3%|▎         | 137/5000 [1:55:13<52:55:16, 39.18s/it]
                                                       
{'loss': '0.1622', 'grad_norm': '10.17', 'learning_rate': '1.985e-05', 'epoch': '0.0274'}

  3%|▎         | 137/5000 [1:55:13<52:55:16, 39.18s/it][2026-05-21 18:01:10,461] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 157])
[2026-05-21 18:01:13,548] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])

  3%|▎         | 138/5000 [1:55:42<48:43:35, 36.08s/it]
                                                       
{'loss': '0.09167', 'grad_norm': '7.145', 'learning_rate': '1.984e-05', 'epoch': '0.0276'}

  3%|▎         | 138/5000 [1:55:42<48:43:35, 36.08s/it][2026-05-21 18:01:34,193] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 978])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 18:01:35,029] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  3%|▎         | 139/5000 [1:56:13<46:44:19, 34.61s/it]
                                                       
{'loss': '0.5062', 'grad_norm': '12.63', 'learning_rate': '1.984e-05', 'epoch': '0.0278'}

  3%|▎         | 139/5000 [1:56:13<46:44:19, 34.61s/it][2026-05-21 18:02:05,216] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 18:02:18,692] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  3%|▎         | 140/5000 [1:56:44<45:20:17, 33.58s/it]
                                                       
{'loss': '0.162', 'grad_norm': '10.13', 'learning_rate': '1.984e-05', 'epoch': '0.028'}

  3%|▎         | 140/5000 [1:56:44<45:20:17, 33.58s/it][2026-05-21 18:02:35,853] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 260])
[2026-05-21 18:02:36,930] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  3%|▎         | 141/5000 [1:57:36<52:51:08, 39.16s/it]
                                                       
{'loss': '0.4325', 'grad_norm': '12.27', 'learning_rate': '1.983e-05', 'epoch': '0.0282'}

  3%|▎         | 141/5000 [1:57:36<52:51:08, 39.16s/it][2026-05-21 18:03:27,987] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 18:03:28,479] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  3%|▎         | 142/5000 [1:58:27<57:17:07, 42.45s/it]
                                                       
{'loss': '0.1702', 'grad_norm': '7.125', 'learning_rate': '1.983e-05', 'epoch': '0.0284'}

  3%|▎         | 142/5000 [1:58:27<57:17:07, 42.45s/it][2026-05-21 18:04:22,146] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 152])
[2026-05-21 18:04:25,554] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 126])

  3%|▎         | 143/5000 [1:59:14<59:27:45, 44.07s/it]
                                                       
{'loss': '1.619', 'grad_norm': '50.76', 'learning_rate': '1.982e-05', 'epoch': '0.0286'}

  3%|▎         | 143/5000 [1:59:14<59:27:45, 44.07s/it][2026-05-21 18:05:10,919] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])
[2026-05-21 18:05:11,560] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 160])

  3%|▎         | 144/5000 [2:00:02<60:47:20, 45.07s/it]
                                                       
{'loss': '2.544', 'grad_norm': '58.94', 'learning_rate': '1.982e-05', 'epoch': '0.0288'}

  3%|▎         | 144/5000 [2:00:02<60:47:20, 45.07s/it][2026-05-21 18:05:59,514] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])
[2026-05-21 18:06:03,891] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  3%|▎         | 145/5000 [2:00:54<63:28:16, 47.06s/it]
                                                       
{'loss': '0.2383', 'grad_norm': '11.41', 'learning_rate': '1.982e-05', 'epoch': '0.029'}

  3%|▎         | 145/5000 [2:00:54<63:28:16, 47.06s/it][2026-05-21 18:06:51,866] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 18:06:54,883] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  3%|▎         | 146/5000 [2:01:45<65:23:40, 48.50s/it]
                                                       
{'loss': '0.1159', 'grad_norm': '3.954', 'learning_rate': '1.981e-05', 'epoch': '0.0292'}

  3%|▎         | 146/5000 [2:01:45<65:23:40, 48.50s/it][2026-05-21 18:07:32,710] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 252])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 18:07:33,940] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 314])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  3%|▎         | 147/5000 [2:02:37<66:41:29, 49.47s/it]
                                                       
{'loss': '0.5023', 'grad_norm': '10.82', 'learning_rate': '1.981e-05', 'epoch': '0.0294'}

  3%|▎         | 147/5000 [2:02:37<66:41:29, 49.47s/it][2026-05-21 18:08:24,520] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 327])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-21 18:08:25,277] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  3%|▎         | 148/5000 [2:03:27<66:53:40, 49.63s/it]
                                                       
{'loss': '0.3815', 'grad_norm': '9.18', 'learning_rate': '1.98e-05', 'epoch': '0.0296'}

  3%|▎         | 148/5000 [2:03:27<66:53:40, 49.63s/it][2026-05-21 18:09:15,000] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 240])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 217])
[2026-05-21 18:09:15,729] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  3%|▎         | 149/5000 [2:04:12<65:05:01, 48.30s/it]
                                                       
{'loss': '0.5794', 'grad_norm': '16.27', 'learning_rate': '1.98e-05', 'epoch': '0.0298'}

  3%|▎         | 149/5000 [2:04:12<65:05:01, 48.30s/it][2026-05-21 18:10:00,807] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 18:10:00,852] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

  3%|▎         | 150/5000 [2:04:58<64:01:04, 47.52s/it]
                                                       
{'loss': '0.5813', 'grad_norm': '17.79', 'learning_rate': '1.98e-05', 'epoch': '0.03'}

  3%|▎         | 150/5000 [2:04:58<64:01:04, 47.52s/it][2026-05-21 18:10:47,359] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 984])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 254])
[2026-05-21 18:10:48,253] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 253])

  3%|▎         | 151/5000 [2:06:09<73:19:09, 54.43s/it]
                                                       
{'loss': '0.7077', 'grad_norm': '13.21', 'learning_rate': '1.979e-05', 'epoch': '0.0302'}

  3%|▎         | 151/5000 [2:06:09<73:19:09, 54.43s/it][2026-05-21 18:11:58,483] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])
[2026-05-21 18:11:59,144] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 319])

  3%|▎         | 152/5000 [2:07:19<79:37:41, 59.13s/it]
                                                       
{'loss': '0.7535', 'grad_norm': '16.95', 'learning_rate': '1.979e-05', 'epoch': '0.0304'}

  3%|▎         | 152/5000 [2:07:19<79:37:41, 59.13s/it][2026-05-21 18:13:16,204] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 18:13:17,023] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])

  3%|▎         | 153/5000 [2:07:48<67:40:55, 50.27s/it]
                                                       
{'loss': '0.5489', 'grad_norm': '13.65', 'learning_rate': '1.978e-05', 'epoch': '0.0306'}

  3%|▎         | 153/5000 [2:07:48<67:40:55, 50.27s/it][2026-05-21 18:13:48,899] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 18:13:51,081] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  3%|▎         | 154/5000 [2:08:24<61:48:26, 45.92s/it]
                                                       
{'loss': '0.3147', 'grad_norm': '12.46', 'learning_rate': '1.978e-05', 'epoch': '0.0308'}

  3%|▎         | 154/5000 [2:08:24<61:48:26, 45.92s/it][2026-05-21 18:14:17,644] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 114])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 18:14:18,094] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 120])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  3%|▎         | 155/5000 [2:08:58<57:04:51, 42.41s/it]
                                                       
{'loss': '0.5841', 'grad_norm': '15.66', 'learning_rate': '1.978e-05', 'epoch': '0.031'}

  3%|▎         | 155/5000 [2:08:58<57:04:51, 42.41s/it][2026-05-21 18:14:51,393] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 128])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 18:14:53,379] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 100])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  3%|▎         | 156/5000 [2:09:33<54:02:37, 40.16s/it]
                                                       
{'loss': '0.1664', 'grad_norm': '12.98', 'learning_rate': '1.977e-05', 'epoch': '0.0312'}

  3%|▎         | 156/5000 [2:09:33<54:02:37, 40.16s/it][2026-05-21 18:15:23,240] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-21 18:15:24,385] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

  3%|▎         | 157/5000 [2:10:21<57:10:38, 42.50s/it]
                                                       
{'loss': '0.4699', 'grad_norm': '10.7', 'learning_rate': '1.977e-05', 'epoch': '0.0314'}

  3%|▎         | 157/5000 [2:10:21<57:10:38, 42.50s/it][2026-05-21 18:16:09,460] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 426])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 18:16:11,221] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  3%|▎         | 158/5000 [2:11:09<59:08:04, 43.97s/it]
                                                       
{'loss': '0.1402', 'grad_norm': '8.013', 'learning_rate': '1.976e-05', 'epoch': '0.0316'}

  3%|▎         | 158/5000 [2:11:09<59:08:04, 43.97s/it][2026-05-21 18:17:09,308] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 97])
[2026-05-21 18:17:10,237] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 121])

  3%|▎         | 159/5000 [2:12:18<69:28:28, 51.66s/it]
                                                       
{'loss': '0.6971', 'grad_norm': '10.34', 'learning_rate': '1.976e-05', 'epoch': '0.0318'}

  3%|▎         | 159/5000 [2:12:18<69:28:28, 51.66s/it][2026-05-21 18:18:19,832] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 174])
[2026-05-21 18:18:21,593] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 200])

  3%|▎         | 160/5000 [2:13:29<77:00:50, 57.28s/it]
                                                       
{'loss': '0.6802', 'grad_norm': '18.75', 'learning_rate': '1.976e-05', 'epoch': '0.032'}

  3%|▎         | 160/5000 [2:13:29<77:00:50, 57.28s/it][2026-05-21 18:19:17,829] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 668])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 18:19:18,119] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 668])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  3%|▎         | 161/5000 [2:14:17<73:20:17, 54.56s/it]
                                                       
{'loss': '1.783', 'grad_norm': '14.11', 'learning_rate': '1.975e-05', 'epoch': '0.0322'}

  3%|▎         | 161/5000 [2:14:17<73:20:17, 54.56s/it][2026-05-21 18:20:05,233] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])
[2026-05-21 18:20:06,153] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 538])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  3%|▎         | 162/5000 [2:15:05<70:35:34, 52.53s/it]
                                                       
{'loss': '1.477', 'grad_norm': '15.42', 'learning_rate': '1.975e-05', 'epoch': '0.0324'}

  3%|▎         | 162/5000 [2:15:05<70:35:34, 52.53s/it][2026-05-21 18:21:23,384] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1025])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])
[2026-05-21 18:21:25,972] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

  3%|▎         | 163/5000 [2:15:56<70:08:43, 52.21s/it]
                                                       
{'loss': '1.052', 'grad_norm': '21.49', 'learning_rate': '1.974e-05', 'epoch': '0.0326'}

  3%|▎         | 163/5000 [2:15:56<70:08:43, 52.21s/it][2026-05-21 18:22:10,348] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 18:22:16,602] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

  3%|▎         | 164/5000 [2:16:47<69:30:03, 51.74s/it]
                                                       
{'loss': '1.083', 'grad_norm': '18.22', 'learning_rate': '1.974e-05', 'epoch': '0.0328'}

  3%|▎         | 164/5000 [2:16:47<69:30:03, 51.74s/it][2026-05-21 18:22:38,992] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])
[2026-05-21 18:22:39,065] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

  3%|▎         | 165/5000 [2:17:39<69:52:17, 52.02s/it]
                                                       
{'loss': '1.947', 'grad_norm': '10.75', 'learning_rate': '1.973e-05', 'epoch': '0.033'}

  3%|▎         | 165/5000 [2:17:39<69:52:17, 52.02s/it][2026-05-21 18:23:31,111] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 18:23:33,332] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

  3%|▎         | 166/5000 [2:18:36<71:37:28, 53.34s/it]
                                                       
{'loss': '1.81', 'grad_norm': '10.97', 'learning_rate': '1.973e-05', 'epoch': '0.0332'}

  3%|▎         | 166/5000 [2:18:36<71:37:28, 53.34s/it][2026-05-21 18:24:27,735] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 156])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 18:24:28,435] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 210])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  3%|▎         | 167/5000 [2:19:35<73:48:21, 54.98s/it]
                                                       
{'loss': '2.773', 'grad_norm': '10.94', 'learning_rate': '1.973e-05', 'epoch': '0.0334'}

  3%|▎         | 167/5000 [2:19:35<73:48:21, 54.98s/it][2026-05-21 18:25:25,802] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 160])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 18:25:26,757] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 151])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  3%|▎         | 168/5000 [2:20:32<74:48:33, 55.74s/it]
                                                       
{'loss': '2.415', 'grad_norm': '10.62', 'learning_rate': '1.972e-05', 'epoch': '0.0336'}

  3%|▎         | 168/5000 [2:20:32<74:48:33, 55.74s/it][2026-05-21 18:26:19,261] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 233])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 215])
[2026-05-21 18:26:19,765] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 220])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

  3%|▎         | 169/5000 [2:21:25<73:51:20, 55.04s/it]
                                                       
{'loss': '0.2919', 'grad_norm': '8.495', 'learning_rate': '1.972e-05', 'epoch': '0.0338'}

  3%|▎         | 169/5000 [2:21:25<73:51:20, 55.04s/it][2026-05-21 18:27:12,568] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 249])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-21 18:27:14,215] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

  3%|▎         | 170/5000 [2:22:17<72:31:26, 54.06s/it]
                                                       
{'loss': '0.3897', 'grad_norm': '9.634', 'learning_rate': '1.971e-05', 'epoch': '0.034'}

  3%|▎         | 170/5000 [2:22:17<72:31:26, 54.06s/it][2026-05-21 18:28:10,049] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 18:28:13,916] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  3%|▎         | 171/5000 [2:23:10<72:00:42, 53.68s/it]
                                                       
{'loss': '0.2935', 'grad_norm': '7.36', 'learning_rate': '1.971e-05', 'epoch': '0.0342'}

  3%|▎         | 171/5000 [2:23:10<72:00:42, 53.68s/it][2026-05-21 18:29:03,619] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 18:29:04,609] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  3%|▎         | 172/5000 [2:24:04<72:10:02, 53.81s/it]
                                                       
{'loss': '0.2511', 'grad_norm': '19.19', 'learning_rate': '1.971e-05', 'epoch': '0.0344'}

  3%|▎         | 172/5000 [2:24:04<72:10:02, 53.81s/it][2026-05-21 18:30:03,152] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 18:30:08,119] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  3%|▎         | 173/5000 [2:24:55<70:47:53, 52.80s/it]
                                                       
{'loss': '1.375', 'grad_norm': '11.44', 'learning_rate': '1.97e-05', 'epoch': '0.0346'}

  3%|▎         | 173/5000 [2:24:55<70:47:53, 52.80s/it][2026-05-21 18:30:52,714] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 18:30:53,815] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  3%|▎         | 174/5000 [2:25:44<69:23:50, 51.77s/it]
                                                       
{'loss': '1.112', 'grad_norm': '11.58', 'learning_rate': '1.97e-05', 'epoch': '0.0348'}

  3%|▎         | 174/5000 [2:25:44<69:23:50, 51.77s/it][2026-05-21 18:31:39,090] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 652])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 52])
[2026-05-21 18:31:43,076] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 659])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 63])

  4%|▎         | 175/5000 [2:26:10<58:51:20, 43.91s/it]
                                                       
{'loss': '0.2136', 'grad_norm': '10.15', 'learning_rate': '1.969e-05', 'epoch': '0.035'}

  4%|▎         | 175/5000 [2:26:10<58:51:20, 43.91s/it][2026-05-21 18:32:05,935] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 641])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 53])
[2026-05-21 18:32:07,968] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 669])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 69])

  4%|▎         | 176/5000 [2:26:39<53:04:18, 39.61s/it]
                                                       
{'loss': '0.1842', 'grad_norm': '9.03', 'learning_rate': '1.969e-05', 'epoch': '0.0352'}

  4%|▎         | 176/5000 [2:26:39<53:04:18, 39.61s/it][2026-05-21 18:32:29,209] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 18:32:29,988] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

  4%|▎         | 177/5000 [2:27:51<65:52:32, 49.17s/it]
                                                       
{'loss': '0.8582', 'grad_norm': '12.22', 'learning_rate': '1.969e-05', 'epoch': '0.0354'}

  4%|▎         | 177/5000 [2:27:51<65:52:32, 49.17s/it][2026-05-21 18:33:40,149] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 180])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])
[2026-05-21 18:33:40,264] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 431])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  4%|▎         | 178/5000 [2:29:02<74:38:25, 55.72s/it]
                                                       
{'loss': '0.7905', 'grad_norm': '12.31', 'learning_rate': '1.968e-05', 'epoch': '0.0356'}

  4%|▎         | 178/5000 [2:29:02<74:38:25, 55.72s/it][2026-05-21 18:34:49,618] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 399])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])
[2026-05-21 18:34:50,356] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 247])

  4%|▎         | 179/5000 [2:30:11<80:05:53, 59.81s/it]
                                                       
{'loss': '0.7992', 'grad_norm': '11.2', 'learning_rate': '1.968e-05', 'epoch': '0.0358'}

  4%|▎         | 179/5000 [2:30:11<80:05:53, 59.81s/it][2026-05-21 18:36:00,633] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 267])
[2026-05-21 18:36:00,919] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 273])

  4%|▎         | 180/5000 [2:31:19<83:33:28, 62.41s/it]
                                                       
{'loss': '1.206', 'grad_norm': '11.93', 'learning_rate': '1.967e-05', 'epoch': '0.036'}

  4%|▎         | 180/5000 [2:31:19<83:33:28, 62.41s/it][2026-05-21 18:37:07,761] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 386])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 975])
[2026-05-21 18:37:13,071] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 430])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  4%|▎         | 181/5000 [2:31:55<72:54:16, 54.46s/it]
                                                       
{'loss': '2.066', 'grad_norm': '11.32', 'learning_rate': '1.967e-05', 'epoch': '0.0362'}

  4%|▎         | 181/5000 [2:31:55<72:54:16, 54.46s/it][2026-05-21 18:37:44,160] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 384])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 18:37:44,819] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  4%|▎         | 182/5000 [2:32:31<65:30:34, 48.95s/it]
                                                       
{'loss': '1.953', 'grad_norm': '10.51', 'learning_rate': '1.967e-05', 'epoch': '0.0364'}

  4%|▎         | 182/5000 [2:32:31<65:30:34, 48.95s/it][2026-05-21 18:38:22,627] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 18:38:23,933] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 230])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  4%|▎         | 183/5000 [2:33:22<66:07:51, 49.42s/it]
                                                       
{'loss': '0.6188', 'grad_norm': '12.03', 'learning_rate': '1.966e-05', 'epoch': '0.0366'}

  4%|▎         | 183/5000 [2:33:22<66:07:51, 49.42s/it][2026-05-21 18:39:13,629] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 204])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])
[2026-05-21 18:39:14,399] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  4%|▎         | 184/5000 [2:34:12<66:25:44, 49.66s/it]
                                                       
{'loss': '0.8005', 'grad_norm': '10.17', 'learning_rate': '1.966e-05', 'epoch': '0.0368'}

  4%|▎         | 184/5000 [2:34:12<66:25:44, 49.66s/it][2026-05-21 18:40:01,056] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 18:40:01,201] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  4%|▎         | 185/5000 [2:34:41<58:12:09, 43.52s/it]
                                                       
{'loss': '0.5714', 'grad_norm': '13.3', 'learning_rate': '1.965e-05', 'epoch': '0.037'}

  4%|▎         | 185/5000 [2:34:43<58:12:09, 43.52s/it][2026-05-21 18:40:32,650] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 432])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])
[2026-05-21 18:40:32,812] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 175])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  4%|▎         | 186/5000 [2:35:15<54:19:21, 40.62s/it]
                                                       
{'loss': '0.2404', 'grad_norm': '11.38', 'learning_rate': '1.965e-05', 'epoch': '0.0372'}

  4%|▎         | 186/5000 [2:35:15<54:19:21, 40.62s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-21 18:41:15,420] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 219])
[2026-05-21 18:41:15,559] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 188])

  4%|▎         | 187/5000 [2:36:08<59:16:16, 44.33s/it]
                                                       
{'loss': '0.5146', 'grad_norm': '15.77', 'learning_rate': '1.964e-05', 'epoch': '0.0374'}

  4%|▎         | 187/5000 [2:36:08<59:16:16, 44.33s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-21 18:42:07,828] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 209])
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-21 18:42:09,191] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 217])

  4%|▍         | 188/5000 [2:37:00<62:02:45, 46.42s/it]
                                                       
{'loss': '1.491', 'grad_norm': '14.45', 'learning_rate': '1.964e-05', 'epoch': '0.0376'}

  4%|▍         | 188/5000 [2:37:00<62:02:45, 46.42s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-21 18:42:50,655] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 219])
[2026-05-21 18:42:51,046] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 202])

  4%|▍         | 189/5000 [2:37:49<63:16:28, 47.35s/it]
                                                       
{'loss': '0.6331', 'grad_norm': '12.07', 'learning_rate': '1.964e-05', 'epoch': '0.0378'}

  4%|▍         | 189/5000 [2:37:49<63:16:28, 47.35s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-21 18:43:40,136] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 190])
[2026-05-21 18:43:40,666] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 236])

  4%|▍         | 190/5000 [2:38:41<64:56:15, 48.60s/it]
                                                       
{'loss': '1.431', 'grad_norm': '12.87', 'learning_rate': '1.963e-05', 'epoch': '0.038'}

  4%|▍         | 190/5000 [2:38:41<64:56:15, 48.60s/it][2026-05-21 18:44:42,199] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 18:44:43,043] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 686])

  4%|▍         | 191/5000 [2:39:29<64:55:25, 48.60s/it]
                                                       
{'loss': '0.7268', 'grad_norm': '11.13', 'learning_rate': '1.963e-05', 'epoch': '0.0382'}

  4%|▍         | 191/5000 [2:39:29<64:55:25, 48.60s/it][2026-05-21 18:45:29,255] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1018])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 18:45:32,438] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  4%|▍         | 192/5000 [2:40:21<66:23:40, 49.71s/it]
                                                       
{'loss': '1.017', 'grad_norm': '10.41', 'learning_rate': '1.962e-05', 'epoch': '0.0384'}

  4%|▍         | 192/5000 [2:40:21<66:23:40, 49.71s/it][2026-05-21 18:46:19,813] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 102])
[2026-05-21 18:46:20,329] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 138])

  4%|▍         | 193/5000 [2:41:10<65:44:34, 49.24s/it]
                                                       
{'loss': '0.3492', 'grad_norm': '8.217', 'learning_rate': '1.962e-05', 'epoch': '0.0386'}

  4%|▍         | 193/5000 [2:41:10<65:44:34, 49.24s/it][2026-05-21 18:47:07,585] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 88])
[2026-05-21 18:47:08,494] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 95])

  4%|▍         | 194/5000 [2:41:58<65:17:02, 48.90s/it]
                                                       
{'loss': '0.3862', 'grad_norm': '9.945', 'learning_rate': '1.962e-05', 'epoch': '0.0388'}

  4%|▍         | 194/5000 [2:41:58<65:17:02, 48.90s/it][2026-05-21 18:47:45,549] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 296])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 18:47:45,771] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 315])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  4%|▍         | 195/5000 [2:42:45<64:42:00, 48.47s/it]
                                                       
{'loss': '0.2603', 'grad_norm': '6.348', 'learning_rate': '1.961e-05', 'epoch': '0.039'}

  4%|▍         | 195/5000 [2:42:45<64:42:00, 48.47s/it][2026-05-21 18:48:33,269] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 290])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-21 18:48:33,928] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 320])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  4%|▍         | 196/5000 [2:43:33<64:16:38, 48.17s/it]
                                                       
{'loss': '0.3817', 'grad_norm': '10.94', 'learning_rate': '1.961e-05', 'epoch': '0.0392'}

  4%|▍         | 196/5000 [2:43:33<64:16:38, 48.17s/it][2026-05-21 18:49:20,072] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 18:49:20,715] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 792])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  4%|▍         | 197/5000 [2:44:42<72:41:25, 54.48s/it]
                                                       
{'loss': '0.8594', 'grad_norm': '17.57', 'learning_rate': '1.96e-05', 'epoch': '0.0394'}

  4%|▍         | 197/5000 [2:44:42<72:41:25, 54.48s/it][2026-05-21 18:50:29,010] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 318])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 18:50:29,309] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 725])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  4%|▍         | 198/5000 [2:45:51<78:41:35, 59.00s/it]
                                                       
{'loss': '0.8227', 'grad_norm': '12.02', 'learning_rate': '1.96e-05', 'epoch': '0.0396'}

  4%|▍         | 198/5000 [2:45:51<78:41:35, 59.00s/it][2026-05-21 18:51:41,947] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 18:51:44,914] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])

  4%|▍         | 199/5000 [2:46:42<75:23:38, 56.53s/it]
                                                       
{'loss': '2.17', 'grad_norm': '14.97', 'learning_rate': '1.96e-05', 'epoch': '0.0398'}

  4%|▍         | 199/5000 [2:46:42<75:23:38, 56.53s/it][2026-05-21 18:52:33,950] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 18:52:34,942] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  4%|▍         | 200/5000 [2:47:31<72:23:22, 54.29s/it]
                                                       
{'loss': '2.115', 'grad_norm': '13.84', 'learning_rate': '1.959e-05', 'epoch': '0.04'}

  4%|▍         | 200/5000 [2:47:31<72:23:22, 54.29s/it][2026-05-21 18:53:22,581] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])
[2026-05-21 18:53:22,732] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  4%|▍         | 201/5000 [2:48:04<63:35:29, 47.70s/it]
                                                       
{'loss': '0.5948', 'grad_norm': '12.43', 'learning_rate': '1.959e-05', 'epoch': '0.0402'}

  4%|▍         | 201/5000 [2:48:04<63:35:29, 47.70s/it][2026-05-21 18:53:54,748] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 18:53:56,261] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

  4%|▍         | 202/5000 [2:48:36<57:30:59, 43.16s/it]
                                                       
{'loss': '0.09137', 'grad_norm': '6.405', 'learning_rate': '1.958e-05', 'epoch': '0.0404'}

  4%|▍         | 202/5000 [2:48:36<57:30:59, 43.16s/it][2026-05-21 18:54:24,716] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-21 18:54:25,920] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  4%|▍         | 203/5000 [2:49:19<57:12:01, 42.93s/it]
                                                       
{'loss': '0.9587', 'grad_norm': '20.06', 'learning_rate': '1.958e-05', 'epoch': '0.0406'}

  4%|▍         | 203/5000 [2:49:19<57:12:01, 42.93s/it][2026-05-21 18:55:08,298] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])
[2026-05-21 18:55:08,784] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 358])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  4%|▍         | 204/5000 [2:49:58<55:38:03, 41.76s/it]
                                                       
{'loss': '0.7526', 'grad_norm': '13.29', 'learning_rate': '1.958e-05', 'epoch': '0.0408'}

  4%|▍         | 204/5000 [2:49:58<55:38:03, 41.76s/it][2026-05-21 18:55:51,317] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])
[2026-05-21 18:55:52,294] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  4%|▍         | 205/5000 [2:51:08<66:53:36, 50.22s/it]
                                                       
{'loss': '0.9543', 'grad_norm': '18.1', 'learning_rate': '1.957e-05', 'epoch': '0.041'}

  4%|▍         | 205/5000 [2:51:08<66:53:36, 50.22s/it][2026-05-21 18:57:02,714] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 18:57:04,517] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  4%|▍         | 206/5000 [2:52:17<74:41:25, 56.09s/it]
                                                       
{'loss': '1.505', 'grad_norm': '18.91', 'learning_rate': '1.957e-05', 'epoch': '0.0412'}

  4%|▍         | 206/5000 [2:52:17<74:41:25, 56.09s/it][2026-05-21 18:58:06,032] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 295])
[2026-05-21 18:58:06,698] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 278])

  4%|▍         | 207/5000 [2:53:27<80:07:50, 60.19s/it]
                                                       
{'loss': '0.4392', 'grad_norm': '17.96', 'learning_rate': '1.956e-05', 'epoch': '0.0414'}

  4%|▍         | 207/5000 [2:53:27<80:07:50, 60.19s/it][2026-05-21 18:59:16,304] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 256])
[2026-05-21 18:59:16,811] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 286])

  4%|▍         | 208/5000 [2:54:37<83:55:25, 63.05s/it]
                                                       
{'loss': '0.8805', 'grad_norm': '16.98', 'learning_rate': '1.956e-05', 'epoch': '0.0416'}

  4%|▍         | 208/5000 [2:54:37<83:55:25, 63.05s/it][2026-05-21 19:00:25,593] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 59])
[2026-05-21 19:00:25,698] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])

  4%|▍         | 209/5000 [2:55:27<78:38:52, 59.10s/it]
                                                       
{'loss': '1.853', 'grad_norm': '25.48', 'learning_rate': '1.956e-05', 'epoch': '0.0418'}

  4%|▍         | 209/5000 [2:55:27<78:38:52, 59.10s/it][2026-05-21 19:01:16,408] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 62])
[2026-05-21 19:01:16,908] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 92])

  4%|▍         | 210/5000 [2:56:18<75:21:19, 56.63s/it]
                                                       
{'loss': '2.486', 'grad_norm': '27.99', 'learning_rate': '1.955e-05', 'epoch': '0.042'}

  4%|▍         | 210/5000 [2:56:18<75:21:19, 56.63s/it][2026-05-21 19:02:04,611] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 630])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])
[2026-05-21 19:02:05,129] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 632])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])

  4%|▍         | 211/5000 [2:57:06<72:14:30, 54.31s/it]
                                                       
{'loss': '0.72', 'grad_norm': '13.22', 'learning_rate': '1.955e-05', 'epoch': '0.0422'}

  4%|▍         | 211/5000 [2:57:06<72:14:30, 54.31s/it][2026-05-21 19:02:53,508] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 630])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])
[2026-05-21 19:02:54,071] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 637])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])

  4%|▍         | 212/5000 [2:57:54<69:44:01, 52.43s/it]
                                                       
{'loss': '0.4574', 'grad_norm': '7.855', 'learning_rate': '1.954e-05', 'epoch': '0.0424'}

  4%|▍         | 212/5000 [2:57:54<69:44:01, 52.43s/it][2026-05-21 19:03:43,794] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 681])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])
[2026-05-21 19:03:44,140] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 644])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  4%|▍         | 213/5000 [2:58:43<68:00:09, 51.14s/it]
                                                       
{'loss': '0.5513', 'grad_norm': '16.15', 'learning_rate': '1.954e-05', 'epoch': '0.0426'}

  4%|▍         | 213/5000 [2:58:43<68:00:09, 51.14s/it][2026-05-21 19:04:32,326] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 641])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 19:04:33,366] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 647])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

  4%|▍         | 214/5000 [2:59:32<67:12:02, 50.55s/it]
                                                       
{'loss': '0.42', 'grad_norm': '16.06', 'learning_rate': '1.953e-05', 'epoch': '0.0428'}

  4%|▍         | 214/5000 [2:59:32<67:12:02, 50.55s/it][2026-05-21 19:05:31,076] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 19:05:32,537] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  4%|▍         | 215/5000 [3:00:17<65:15:37, 49.10s/it]
                                                       
{'loss': '0.2929', 'grad_norm': '15.66', 'learning_rate': '1.953e-05', 'epoch': '0.043'}

  4%|▍         | 215/5000 [3:00:17<65:15:37, 49.10s/it][2026-05-21 19:06:17,497] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 19:06:18,277] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  4%|▍         | 216/5000 [3:01:04<64:09:31, 48.28s/it]
                                                       
{'loss': '0.378', 'grad_norm': '15.3', 'learning_rate': '1.953e-05', 'epoch': '0.0432'}

  4%|▍         | 216/5000 [3:01:04<64:09:31, 48.28s/it][2026-05-21 19:06:51,808] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 286])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 19:06:51,815] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 285])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

  4%|▍         | 217/5000 [3:01:37<57:56:50, 43.61s/it]
                                                       
{'loss': '1.015', 'grad_norm': '12.42', 'learning_rate': '1.952e-05', 'epoch': '0.0434'}

  4%|▍         | 217/5000 [3:01:37<57:56:50, 43.61s/it][2026-05-21 19:07:24,632] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 297])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 19:07:25,200] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 271])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  4%|▍         | 218/5000 [3:02:09<53:34:02, 40.33s/it]
                                                       
{'loss': '1.012', 'grad_norm': '12.52', 'learning_rate': '1.952e-05', 'epoch': '0.0436'}

  4%|▍         | 218/5000 [3:02:09<53:34:02, 40.33s/it][2026-05-21 19:08:09,444] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 180])
[2026-05-21 19:08:10,138] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1032])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 150])

  4%|▍         | 219/5000 [3:03:07<60:20:36, 45.44s/it]
                                                       
{'loss': '1.808', 'grad_norm': '13.15', 'learning_rate': '1.951e-05', 'epoch': '0.0438'}

  4%|▍         | 219/5000 [3:03:07<60:20:36, 45.44s/it][2026-05-21 19:09:06,871] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 105])
[2026-05-21 19:09:08,085] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 121])

  4%|▍         | 220/5000 [3:04:04<65:15:56, 49.15s/it]
                                                       
{'loss': '1.898', 'grad_norm': '14.74', 'learning_rate': '1.951e-05', 'epoch': '0.044'}

  4%|▍         | 220/5000 [3:04:04<65:15:56, 49.15s/it][2026-05-21 19:09:55,567] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 185])
[2026-05-21 19:09:56,068] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 363])

  4%|▍         | 221/5000 [3:04:59<67:20:08, 50.72s/it]
                                                       
{'loss': '0.4661', 'grad_norm': '12.52', 'learning_rate': '1.951e-05', 'epoch': '0.0442'}

  4%|▍         | 221/5000 [3:04:59<67:20:08, 50.72s/it][2026-05-21 19:10:49,984] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 206])
[2026-05-21 19:10:50,989] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])

  4%|▍         | 222/5000 [3:05:54<68:55:17, 51.93s/it]
                                                       
{'loss': '0.3652', 'grad_norm': '8.507', 'learning_rate': '1.95e-05', 'epoch': '0.0444'}

  4%|▍         | 222/5000 [3:05:54<68:55:17, 51.93s/it][2026-05-21 19:11:49,735] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 112])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 19:11:53,417] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 120])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  4%|▍         | 223/5000 [3:06:24<60:14:20, 45.40s/it]
                                                       
{'loss': '0.4142', 'grad_norm': '13.09', 'learning_rate': '1.95e-05', 'epoch': '0.0446'}

  4%|▍         | 223/5000 [3:06:24<60:14:20, 45.40s/it][2026-05-21 19:12:21,132] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 81])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 19:12:24,161] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 103])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  4%|▍         | 224/5000 [3:06:56<55:06:39, 41.54s/it]
                                                       
{'loss': '0.04351', 'grad_norm': '4.779', 'learning_rate': '1.949e-05', 'epoch': '0.0448'}

  4%|▍         | 224/5000 [3:06:56<55:06:39, 41.54s/it][2026-05-21 19:12:43,193] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 316])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 92])
[2026-05-21 19:12:43,719] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

  4%|▍         | 225/5000 [3:07:45<58:03:06, 43.77s/it]
                                                       
{'loss': '0.3411', 'grad_norm': '8.256', 'learning_rate': '1.949e-05', 'epoch': '0.045'}

  4%|▍         | 225/5000 [3:07:45<58:03:06, 43.77s/it][2026-05-21 19:13:32,741] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 200])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-21 19:13:33,140] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 82])

  5%|▍         | 226/5000 [3:08:32<59:10:45, 44.63s/it]
                                                       
{'loss': '0.3045', 'grad_norm': '8.004', 'learning_rate': '1.949e-05', 'epoch': '0.0452'}

  5%|▍         | 226/5000 [3:08:32<59:10:45, 44.63s/it][2026-05-21 19:14:18,806] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-21 19:14:19,273] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 568])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

  5%|▍         | 227/5000 [3:09:21<60:50:54, 45.89s/it]
                                                       
{'loss': '0.4171', 'grad_norm': '11.21', 'learning_rate': '1.948e-05', 'epoch': '0.0454'}

  5%|▍         | 227/5000 [3:09:21<60:50:54, 45.89s/it][2026-05-21 19:15:07,736] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 618])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-21 19:15:08,255] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 694])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])

  5%|▍         | 228/5000 [3:10:08<61:27:07, 46.36s/it]
                                                       
{'loss': '0.4787', 'grad_norm': '9.085', 'learning_rate': '1.948e-05', 'epoch': '0.0456'}

  5%|▍         | 228/5000 [3:10:08<61:27:07, 46.36s/it][2026-05-21 19:15:59,580] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 177])
[2026-05-21 19:15:59,785] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 154])

  5%|▍         | 229/5000 [3:10:56<61:55:38, 46.73s/it]
                                                       
{'loss': '1.325', 'grad_norm': '10.8', 'learning_rate': '1.947e-05', 'epoch': '0.0458'}

  5%|▍         | 229/5000 [3:10:56<61:55:38, 46.73s/it][2026-05-21 19:16:47,075] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 179])
[2026-05-21 19:16:47,228] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])

  5%|▍         | 230/5000 [3:11:45<62:48:40, 47.40s/it]
                                                       
{'loss': '0.5055', 'grad_norm': '13.96', 'learning_rate': '1.947e-05', 'epoch': '0.046'}

  5%|▍         | 230/5000 [3:11:45<62:48:40, 47.40s/it][2026-05-21 19:17:49,917] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 304])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-21 19:17:51,737] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

  5%|▍         | 231/5000 [3:12:11<54:26:27, 41.10s/it]
                                                       
{'loss': '0.2702', 'grad_norm': '10.12', 'learning_rate': '1.947e-05', 'epoch': '0.0462'}

  5%|▍         | 231/5000 [3:12:11<54:26:27, 41.10s/it][2026-05-21 19:18:18,326] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 262])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 19:18:21,729] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 504])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

  5%|▍         | 232/5000 [3:12:40<49:29:16, 37.36s/it]
                                                       
{'loss': '0.1383', 'grad_norm': '6.4', 'learning_rate': '1.946e-05', 'epoch': '0.0464'}

  5%|▍         | 232/5000 [3:12:40<49:29:16, 37.36s/it][2026-05-21 19:18:27,125] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 238])
[2026-05-21 19:18:28,499] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 453])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  5%|▍         | 233/5000 [3:13:13<47:53:16, 36.16s/it]
                                                       
{'loss': '1.868', 'grad_norm': '10.23', 'learning_rate': '1.946e-05', 'epoch': '0.0466'}

  5%|▍         | 233/5000 [3:13:13<47:53:16, 36.16s/it][2026-05-21 19:19:01,303] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 19:19:01,703] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  5%|▍         | 234/5000 [3:13:47<46:52:31, 35.41s/it]
                                                       
{'loss': '1.809', 'grad_norm': '10.24', 'learning_rate': '1.945e-05', 'epoch': '0.0468'}

  5%|▍         | 234/5000 [3:13:47<46:52:31, 35.41s/it][2026-05-21 19:19:34,513] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 263])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 19:19:34,766] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 264])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

  5%|▍         | 235/5000 [3:14:34<51:35:17, 38.98s/it]
                                                       
{'loss': '1.929', 'grad_norm': '8.798', 'learning_rate': '1.945e-05', 'epoch': '0.047'}

  5%|▍         | 235/5000 [3:14:34<51:35:17, 38.98s/it][2026-05-21 19:20:22,009] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 329])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 19:20:22,560] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 253])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  5%|▍         | 236/5000 [3:15:24<56:04:09, 42.37s/it]
                                                       
{'loss': '1.8', 'grad_norm': '10.3', 'learning_rate': '1.944e-05', 'epoch': '0.0472'}

  5%|▍         | 236/5000 [3:15:24<56:04:09, 42.37s/it][2026-05-21 19:21:14,518] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 245])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 19:21:14,644] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  5%|▍         | 237/5000 [3:15:56<51:55:51, 39.25s/it]
                                                       
{'loss': '0.25', 'grad_norm': '10.12', 'learning_rate': '1.944e-05', 'epoch': '0.0474'}

  5%|▍         | 237/5000 [3:15:56<51:55:51, 39.25s/it][2026-05-21 19:21:46,008] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 283])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 19:21:47,200] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 277])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])

  5%|▍         | 238/5000 [3:16:31<50:07:51, 37.90s/it]
                                                       
{'loss': '0.1038', 'grad_norm': '6.127', 'learning_rate': '1.944e-05', 'epoch': '0.0476'}

  5%|▍         | 238/5000 [3:16:31<50:07:51, 37.90s/it][2026-05-21 19:22:31,431] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])
[2026-05-21 19:22:32,462] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 308])

  5%|▍         | 239/5000 [3:17:05<48:31:23, 36.69s/it]
                                                       
{'loss': '0.5607', 'grad_norm': '12.84', 'learning_rate': '1.943e-05', 'epoch': '0.0478'}

  5%|▍         | 239/5000 [3:17:05<48:31:23, 36.69s/it][2026-05-21 19:23:05,481] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])
[2026-05-21 19:23:06,606] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 348])

  5%|▍         | 240/5000 [3:17:38<47:15:18, 35.74s/it]
                                                       
{'loss': '0.199', 'grad_norm': '13.04', 'learning_rate': '1.943e-05', 'epoch': '0.048'}

  5%|▍         | 240/5000 [3:17:38<47:15:18, 35.74s/it][2026-05-21 19:23:27,388] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 19:23:29,097] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  5%|▍         | 241/5000 [3:18:10<45:30:16, 34.42s/it]
                                                       
{'loss': '0.5036', 'grad_norm': '12.07', 'learning_rate': '1.942e-05', 'epoch': '0.0482'}

  5%|▍         | 241/5000 [3:18:10<45:30:16, 34.42s/it][2026-05-21 19:23:59,531] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-21 19:24:00,026] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  5%|▍         | 242/5000 [3:18:42<44:36:25, 33.75s/it]
                                                       
{'loss': '0.08242', 'grad_norm': '6.403', 'learning_rate': '1.942e-05', 'epoch': '0.0484'}

  5%|▍         | 242/5000 [3:18:42<44:36:25, 33.75s/it][2026-05-21 19:24:44,589] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 166])
[2026-05-21 19:24:44,967] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 215])

  5%|▍         | 243/5000 [3:19:36<52:37:12, 39.82s/it]
                                                       
{'loss': '0.988', 'grad_norm': '13.77', 'learning_rate': '1.942e-05', 'epoch': '0.0486'}

  5%|▍         | 243/5000 [3:19:36<52:37:12, 39.82s/it][2026-05-21 19:25:38,539] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 69])
[2026-05-21 19:25:42,573] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 101])

  5%|▍         | 244/5000 [3:20:33<59:27:11, 45.00s/it]
                                                       
{'loss': '0.2659', 'grad_norm': '9.5', 'learning_rate': '1.941e-05', 'epoch': '0.0488'}

  5%|▍         | 244/5000 [3:20:33<59:27:11, 45.00s/it][2026-05-21 19:26:22,558] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 251])
[2026-05-21 19:26:23,792] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 287])

  5%|▍         | 245/5000 [3:21:25<62:07:22, 47.03s/it]
                                                       
{'loss': '0.4616', 'grad_norm': '9.904', 'learning_rate': '1.941e-05', 'epoch': '0.049'}

  5%|▍         | 245/5000 [3:21:25<62:07:22, 47.03s/it][2026-05-21 19:27:14,319] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 256])
[2026-05-21 19:27:16,820] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])

  5%|▍         | 246/5000 [3:22:18<64:24:43, 48.78s/it]
                                                       
{'loss': '0.289', 'grad_norm': '8.802', 'learning_rate': '1.94e-05', 'epoch': '0.0492'}

  5%|▍         | 246/5000 [3:22:18<64:24:43, 48.78s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-21 19:28:04,383] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 259])
[2026-05-21 19:28:05,055] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])

  5%|▍         | 247/5000 [3:23:09<65:14:48, 49.42s/it]
                                                       
{'loss': '1.005', 'grad_norm': '12.5', 'learning_rate': '1.94e-05', 'epoch': '0.0494'}

  5%|▍         | 247/5000 [3:23:09<65:14:48, 49.42s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-21 19:28:55,265] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 168])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 240])
[2026-05-21 19:28:56,299] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 428])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 233])

  5%|▍         | 248/5000 [3:24:01<66:18:57, 50.24s/it]
                                                       
{'loss': '0.9116', 'grad_norm': '13.03', 'learning_rate': '1.94e-05', 'epoch': '0.0496'}

  5%|▍         | 248/5000 [3:24:01<66:18:57, 50.24s/it][2026-05-21 19:29:47,414] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 55])
[2026-05-21 19:29:47,895] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 64])

  5%|▍         | 249/5000 [3:24:50<66:03:46, 50.06s/it]
                                                       
{'loss': '0.3757', 'grad_norm': '10.34', 'learning_rate': '1.939e-05', 'epoch': '0.0498'}

  5%|▍         | 249/5000 [3:24:50<66:03:46, 50.06s/it][2026-05-21 19:30:37,586] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])
[2026-05-21 19:30:37,606] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 201])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])

  5%|▌         | 250/5000 [3:25:39<65:20:52, 49.53s/it]
                                                       
{'loss': '0.4014', 'grad_norm': '36.6', 'learning_rate': '1.939e-05', 'epoch': '0.05'}

  5%|▌         | 250/5000 [3:25:39<65:20:52, 49.53s/it][2026-05-21 19:31:23,812] INFO [src.utils.basic_utils:19] Saving model to /workspace/VLM2Vec/outputs/Qwen3.5-08b-continued-mixed-alltasks/checkpoint-250
[2026-05-21 19:31:24,339] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 19:31:24,368] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-21 19:31:26,504] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 476])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 19:31:27,102] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 562])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

  5%|▌         | 251/5000 [3:26:27<64:46:34, 49.10s/it]
                                                       
{'loss': '0.3882', 'grad_norm': '8.426', 'learning_rate': '1.938e-05', 'epoch': '0.0502'}

  5%|▌         | 251/5000 [3:26:27<64:46:34, 49.10s/it][2026-05-21 19:32:14,789] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 979])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 19:32:15,274] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  5%|▌         | 252/5000 [3:27:14<64:02:13, 48.55s/it]
                                                       
{'loss': '0.1359', 'grad_norm': '9.678', 'learning_rate': '1.938e-05', 'epoch': '0.0504'}

  5%|▌         | 252/5000 [3:27:14<64:02:13, 48.55s/it][2026-05-21 19:33:01,398] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 279])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 19:33:02,241] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 300])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

  5%|▌         | 253/5000 [3:27:39<54:48:31, 41.57s/it]
                                                       
{'loss': '0.06705', 'grad_norm': '6.683', 'learning_rate': '1.938e-05', 'epoch': '0.0506'}

  5%|▌         | 253/5000 [3:27:39<54:48:31, 41.57s/it][2026-05-21 19:33:26,862] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 266])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 19:33:27,508] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 289])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

  5%|▌         | 254/5000 [3:28:07<49:14:13, 37.35s/it]
                                                       
{'loss': '0.06951', 'grad_norm': '7.05', 'learning_rate': '1.937e-05', 'epoch': '0.0508'}

  5%|▌         | 254/5000 [3:28:07<49:14:13, 37.35s/it][2026-05-21 19:33:54,315] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 137])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 19:33:55,030] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 96])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  5%|▌         | 255/5000 [3:28:30<43:36:40, 33.09s/it]
                                                       
{'loss': '0.3775', 'grad_norm': '9.589', 'learning_rate': '1.937e-05', 'epoch': '0.051'}

  5%|▌         | 255/5000 [3:28:30<43:36:40, 33.09s/it][2026-05-21 19:34:18,709] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 84])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 980])
[2026-05-21 19:34:18,719] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 153])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

  5%|▌         | 256/5000 [3:28:57<41:12:21, 31.27s/it]
                                                       
{'loss': '0.3024', 'grad_norm': '10.1', 'learning_rate': '1.936e-05', 'epoch': '0.0512'}

  5%|▌         | 256/5000 [3:28:57<41:12:21, 31.27s/it][2026-05-21 19:34:46,510] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 305])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 19:34:46,725] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 288])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 967])

  5%|▌         | 257/5000 [3:29:31<42:08:54, 31.99s/it]
                                                       
{'loss': '1.305', 'grad_norm': '17.17', 'learning_rate': '1.936e-05', 'epoch': '0.0514'}

  5%|▌         | 257/5000 [3:29:31<42:08:54, 31.99s/it][2026-05-21 19:35:20,224] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 266])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 19:35:21,926] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 296])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  5%|▌         | 258/5000 [3:30:22<49:56:08, 37.91s/it]
                                                       
{'loss': '0.3765', 'grad_norm': '9.822', 'learning_rate': '1.936e-05', 'epoch': '0.0516'}

  5%|▌         | 258/5000 [3:30:22<49:56:08, 37.91s/it][2026-05-21 19:36:11,334] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 113])
[2026-05-21 19:36:11,996] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

  5%|▌         | 259/5000 [3:30:53<46:55:40, 35.63s/it]
                                                       
{'loss': '0.2349', 'grad_norm': '9.418', 'learning_rate': '1.935e-05', 'epoch': '0.0518'}

  5%|▌         | 259/5000 [3:30:53<46:55:40, 35.63s/it][2026-05-21 19:36:41,384] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 222])
[2026-05-21 19:36:42,028] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1016])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 208])

  5%|▌         | 260/5000 [3:31:23<44:39:03, 33.91s/it]
                                                       
{'loss': '0.2814', 'grad_norm': '11.25', 'learning_rate': '1.935e-05', 'epoch': '0.052'}

  5%|▌         | 260/5000 [3:31:23<44:39:03, 33.91s/it][2026-05-21 19:37:19,774] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])
[2026-05-21 19:37:20,847] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 281])

  5%|▌         | 261/5000 [3:32:09<49:27:51, 37.58s/it]
                                                       
{'loss': '0.7681', 'grad_norm': '17.1', 'learning_rate': '1.934e-05', 'epoch': '0.0522'}

  5%|▌         | 261/5000 [3:32:09<49:27:51, 37.58s/it][2026-05-21 19:38:06,732] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 237])
[2026-05-21 19:38:07,838] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])

  5%|▌         | 262/5000 [3:32:56<53:12:25, 40.43s/it]
                                                       
{'loss': '0.963', 'grad_norm': '16.15', 'learning_rate': '1.934e-05', 'epoch': '0.0524'}

  5%|▌         | 262/5000 [3:32:56<53:12:25, 40.43s/it][2026-05-21 19:38:44,510] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 19:38:44,581] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  5%|▌         | 263/5000 [3:33:42<55:32:09, 42.21s/it]
                                                       
{'loss': '0.7542', 'grad_norm': '14.77', 'learning_rate': '1.933e-05', 'epoch': '0.0526'}

  5%|▌         | 263/5000 [3:33:42<55:32:09, 42.21s/it][2026-05-21 19:39:30,894] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 199])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 19:39:31,816] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 502])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  5%|▌         | 264/5000 [3:34:29<57:31:13, 43.72s/it]
                                                       
{'loss': '0.526', 'grad_norm': '8.296', 'learning_rate': '1.933e-05', 'epoch': '0.0528'}

  5%|▌         | 264/5000 [3:34:29<57:31:13, 43.72s/it][2026-05-21 19:40:18,558] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 19:40:20,121] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])

  5%|▌         | 265/5000 [3:35:16<58:34:41, 44.54s/it]
                                                       
{'loss': '0.5631', 'grad_norm': '18.53', 'learning_rate': '1.933e-05', 'epoch': '0.053'}

  5%|▌         | 265/5000 [3:35:16<58:34:41, 44.54s/it][2026-05-21 19:41:06,580] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 19:41:07,532] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

  5%|▌         | 266/5000 [3:36:04<59:52:30, 45.53s/it]
                                                       
{'loss': '0.5957', 'grad_norm': '18.48', 'learning_rate': '1.932e-05', 'epoch': '0.0532'}

  5%|▌         | 266/5000 [3:36:04<59:52:30, 45.53s/it][2026-05-21 19:41:50,895] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 840])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-21 19:41:50,960] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

  5%|▌         | 267/5000 [3:36:50<60:14:19, 45.82s/it]
                                                       
{'loss': '0.5219', 'grad_norm': '10.63', 'learning_rate': '1.932e-05', 'epoch': '0.0534'}

  5%|▌         | 267/5000 [3:36:50<60:14:19, 45.82s/it][2026-05-21 19:42:37,509] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-21 19:42:37,563] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

  5%|▌         | 268/5000 [3:37:37<60:30:48, 46.04s/it]
                                                       
{'loss': '0.1606', 'grad_norm': '9.826', 'learning_rate': '1.931e-05', 'epoch': '0.0536'}

  5%|▌         | 268/5000 [3:37:37<60:30:48, 46.04s/it][2026-05-21 19:43:23,432] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 244])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-21 19:43:24,547] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

  5%|▌         | 269/5000 [3:38:25<61:23:40, 46.72s/it]
                                                       
{'loss': '0.3733', 'grad_norm': '10.82', 'learning_rate': '1.931e-05', 'epoch': '0.0538'}

  5%|▌         | 269/5000 [3:38:25<61:23:40, 46.72s/it][2026-05-21 19:44:11,725] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 279])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-21 19:44:11,748] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

  5%|▌         | 270/5000 [3:39:14<62:05:29, 47.26s/it]
                                                       
{'loss': '0.5659', 'grad_norm': '9.008', 'learning_rate': '1.931e-05', 'epoch': '0.054'}

  5%|▌         | 270/5000 [3:39:14<62:05:29, 47.26s/it][2026-05-21 19:45:07,840] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 579])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 19:45:10,498] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 617])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])

  5%|▌         | 271/5000 [3:40:22<70:24:06, 53.59s/it]
                                                       
{'loss': '1.148', 'grad_norm': '16.36', 'learning_rate': '1.93e-05', 'epoch': '0.0542'}

  5%|▌         | 271/5000 [3:40:22<70:24:06, 53.59s/it][2026-05-21 19:46:17,951] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 928])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-21 19:46:19,868] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 969])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])

  5%|▌         | 272/5000 [3:41:11<68:24:01, 52.08s/it]
                                                       
{'loss': '1.33', 'grad_norm': '15.97', 'learning_rate': '1.93e-05', 'epoch': '0.0544'}

  5%|▌         | 272/5000 [3:41:11<68:24:01, 52.08s/it][2026-05-21 19:46:59,530] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 774])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 19:47:02,406] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 523])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

  5%|▌         | 273/5000 [3:41:56<65:46:45, 50.10s/it]
                                                       
{'loss': '0.4303', 'grad_norm': '13.28', 'learning_rate': '1.929e-05', 'epoch': '0.0546'}

  5%|▌         | 273/5000 [3:41:56<65:46:45, 50.10s/it][2026-05-21 19:47:44,794] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 963])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 19:47:45,208] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 963])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  5%|▌         | 274/5000 [3:42:42<64:01:29, 48.77s/it]
                                                       
{'loss': '0.4647', 'grad_norm': '12.72', 'learning_rate': '1.929e-05', 'epoch': '0.0548'}

  5%|▌         | 274/5000 [3:42:42<64:01:29, 48.77s/it][2026-05-21 19:48:33,546] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 107])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 19:48:33,770] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 98])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  6%|▌         | 275/5000 [3:43:29<63:33:27, 48.42s/it]
                                                       
{'loss': '0.9537', 'grad_norm': '16.8', 'learning_rate': '1.929e-05', 'epoch': '0.055'}

  6%|▌         | 275/5000 [3:43:29<63:33:27, 48.42s/it][2026-05-21 19:49:22,178] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 79])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-21 19:49:22,559] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 105])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  6%|▌         | 276/5000 [3:43:57<55:28:26, 42.27s/it]
                                                       
{'loss': '0.777', 'grad_norm': '15.22', 'learning_rate': '1.928e-05', 'epoch': '0.0552'}

  6%|▌         | 276/5000 [3:43:57<55:28:26, 42.27s/it][2026-05-21 19:49:45,490] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 141])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])
[2026-05-21 19:49:47,308] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 173])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  6%|▌         | 277/5000 [3:44:52<60:16:14, 45.94s/it]
                                                       
{'loss': '1.351', 'grad_norm': '13.08', 'learning_rate': '1.928e-05', 'epoch': '0.0554'}

  6%|▌         | 277/5000 [3:44:52<60:16:14, 45.94s/it][2026-05-21 19:50:40,204] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 149])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 19:50:41,230] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 173])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  6%|▌         | 278/5000 [3:45:59<68:49:44, 52.47s/it]
                                                       
{'loss': '1.539', 'grad_norm': '16.28', 'learning_rate': '1.927e-05', 'epoch': '0.0556'}

  6%|▌         | 278/5000 [3:45:59<68:49:44, 52.47s/it][2026-05-21 19:51:46,827] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 273])
[2026-05-21 19:51:48,140] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 291])

  6%|▌         | 279/5000 [3:47:00<71:57:31, 54.87s/it]
                                                       
{'loss': '1.216', 'grad_norm': '17.33', 'learning_rate': '1.927e-05', 'epoch': '0.0558'}

  6%|▌         | 279/5000 [3:47:00<71:57:31, 54.87s/it][2026-05-21 19:52:47,841] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])
[2026-05-21 19:52:48,437] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])

  6%|▌         | 280/5000 [3:48:08<76:58:22, 58.71s/it]
                                                       
{'loss': '1.075', 'grad_norm': '13.04', 'learning_rate': '1.927e-05', 'epoch': '0.056'}

  6%|▌         | 280/5000 [3:48:08<76:58:22, 58.71s/it][2026-05-21 19:53:55,316] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])
[2026-05-21 19:53:55,586] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])

  6%|▌         | 281/5000 [3:48:55<72:23:19, 55.22s/it]
                                                       
{'loss': '1.654', 'grad_norm': '13.39', 'learning_rate': '1.926e-05', 'epoch': '0.0562'}

  6%|▌         | 281/5000 [3:48:55<72:23:19, 55.22s/it][2026-05-21 19:54:42,990] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 242])
[2026-05-21 19:54:43,294] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 265])

  6%|▌         | 282/5000 [3:49:42<69:21:36, 52.92s/it]
                                                       
{'loss': '1.682', 'grad_norm': '14.42', 'learning_rate': '1.926e-05', 'epoch': '0.0564'}

  6%|▌         | 282/5000 [3:49:42<69:21:36, 52.92s/it][2026-05-21 19:55:28,986] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 454])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 8])
[2026-05-21 19:55:31,002] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 11])

  6%|▌         | 283/5000 [3:50:28<66:29:17, 50.74s/it]
                                                       
{'loss': '0.7129', 'grad_norm': '20.16', 'learning_rate': '1.925e-05', 'epoch': '0.0566'}

  6%|▌         | 283/5000 [3:50:28<66:29:17, 50.74s/it][2026-05-21 19:56:15,137] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 959])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 14])
[2026-05-21 19:56:15,746] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 963])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 7])

  6%|▌         | 284/5000 [3:51:11<63:37:39, 48.57s/it]
                                                       
{'loss': '0.8881', 'grad_norm': '18.92', 'learning_rate': '1.925e-05', 'epoch': '0.0568'}

  6%|▌         | 284/5000 [3:51:11<63:37:39, 48.57s/it][2026-05-21 19:56:58,945] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 937])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 60])
[2026-05-21 19:56:59,114] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 730])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])

  6%|▌         | 285/5000 [3:51:41<56:02:44, 42.79s/it]
                                                       
{'loss': '0.7182', 'grad_norm': '13.71', 'learning_rate': '1.924e-05', 'epoch': '0.057'}

  6%|▌         | 285/5000 [3:51:41<56:02:44, 42.79s/it][2026-05-21 19:57:28,140] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 968])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 56])
[2026-05-21 19:57:29,822] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 83])

  6%|▌         | 286/5000 [3:52:10<50:42:35, 38.73s/it]
                                                       
{'loss': '0.08156', 'grad_norm': '6.406', 'learning_rate': '1.924e-05', 'epoch': '0.0572'}

  6%|▌         | 286/5000 [3:52:10<50:42:35, 38.73s/it][2026-05-21 19:58:10,735] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 158])
[2026-05-21 19:58:11,174] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 117])

  6%|▌         | 287/5000 [3:52:40<47:28:49, 36.27s/it]
                                                       
{'loss': '0.6097', 'grad_norm': '12.69', 'learning_rate': '1.924e-05', 'epoch': '0.0574'}

  6%|▌         | 287/5000 [3:52:40<47:28:49, 36.27s/it][2026-05-21 19:58:42,664] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 221])
[2026-05-21 19:58:43,346] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1018])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 292])

  6%|▌         | 288/5000 [3:53:14<46:36:14, 35.61s/it]
                                                       
{'loss': '0.09039', 'grad_norm': '5.447', 'learning_rate': '1.923e-05', 'epoch': '0.0576'}

  6%|▌         | 288/5000 [3:53:14<46:36:14, 35.61s/it][2026-05-21 19:59:14,974] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 19:59:15,012] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  6%|▌         | 289/5000 [3:54:01<50:51:45, 38.87s/it]
                                                       
{'loss': '1.105', 'grad_norm': '17.65', 'learning_rate': '1.923e-05', 'epoch': '0.0578'}

  6%|▌         | 289/5000 [3:54:01<50:51:45, 38.87s/it][2026-05-21 19:59:59,205] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-21 20:00:00,414] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  6%|▌         | 290/5000 [3:54:45<53:03:05, 40.55s/it]
                                                       
{'loss': '0.8334', 'grad_norm': '16.59', 'learning_rate': '1.922e-05', 'epoch': '0.058'}

  6%|▌         | 290/5000 [3:54:45<53:03:05, 40.55s/it][2026-05-21 20:00:36,999] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 416])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 20:00:37,301] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 233])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])

  6%|▌         | 291/5000 [3:55:29<54:11:08, 41.42s/it]
                                                       
{'loss': '0.4504', 'grad_norm': '11.61', 'learning_rate': '1.922e-05', 'epoch': '0.0582'}

  6%|▌         | 291/5000 [3:55:29<54:11:08, 41.42s/it][2026-05-21 20:01:19,235] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 169])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 975])
[2026-05-21 20:01:21,129] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  6%|▌         | 292/5000 [3:56:16<56:30:36, 43.21s/it]
                                                       
{'loss': '0.1459', 'grad_norm': '7.453', 'learning_rate': '1.922e-05', 'epoch': '0.0584'}

  6%|▌         | 292/5000 [3:56:16<56:30:36, 43.21s/it][2026-05-21 20:02:03,649] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 48])
[2026-05-21 20:02:04,480] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 333])

  6%|▌         | 293/5000 [3:57:05<58:37:08, 44.83s/it]
                                                       
{'loss': '0.5981', 'grad_norm': '12.48', 'learning_rate': '1.921e-05', 'epoch': '0.0586'}

  6%|▌         | 293/5000 [3:57:05<58:37:08, 44.83s/it][2026-05-21 20:02:52,216] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 358])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])
[2026-05-21 20:02:53,285] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 428])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 81])

  6%|▌         | 294/5000 [3:57:55<60:35:00, 46.35s/it]
                                                       
{'loss': '0.4836', 'grad_norm': '10.65', 'learning_rate': '1.921e-05', 'epoch': '0.0588'}

  6%|▌         | 294/5000 [3:57:55<60:35:00, 46.35s/it][2026-05-21 20:03:42,470] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 633])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])
[2026-05-21 20:03:42,975] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 647])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 84])

  6%|▌         | 295/5000 [3:59:04<69:30:26, 53.18s/it]
                                                       
{'loss': '1.291', 'grad_norm': '13.47', 'learning_rate': '1.92e-05', 'epoch': '0.059'}

  6%|▌         | 295/5000 [3:59:04<69:30:26, 53.18s/it][2026-05-21 20:04:52,033] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 703])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 58])
[2026-05-21 20:04:52,501] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 647])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 69])

  6%|▌         | 296/5000 [4:00:13<75:54:17, 58.09s/it]
                                                       
{'loss': '1.092', 'grad_norm': '11.88', 'learning_rate': '1.92e-05', 'epoch': '0.0592'}

  6%|▌         | 296/5000 [4:00:13<75:54:17, 58.09s/it][2026-05-21 20:06:03,341] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 654])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 20:06:03,452] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 637])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  6%|▌         | 297/5000 [4:01:04<73:01:17, 55.90s/it]
                                                       
{'loss': '0.7046', 'grad_norm': '11.24', 'learning_rate': '1.92e-05', 'epoch': '0.0594'}

  6%|▌         | 297/5000 [4:01:04<73:01:17, 55.90s/it][2026-05-21 20:06:53,770] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 692])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-21 20:06:54,022] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 655])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  6%|▌         | 298/5000 [4:01:54<70:37:51, 54.08s/it]
                                                       
{'loss': '0.7068', 'grad_norm': '9.966', 'learning_rate': '1.919e-05', 'epoch': '0.0596'}

  6%|▌         | 298/5000 [4:01:54<70:37:51, 54.08s/it][2026-05-21 20:07:43,297] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])
[2026-05-21 20:07:51,138] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  6%|▌         | 299/5000 [4:02:25<61:43:24, 47.27s/it]
                                                       
{'loss': '1.234', 'grad_norm': '24.73', 'learning_rate': '1.919e-05', 'epoch': '0.0598'}

  6%|▌         | 299/5000 [4:02:25<61:43:24, 47.27s/it][2026-05-21 20:08:19,320] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 20:08:20,670] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  6%|▌         | 300/5000 [4:02:56<55:10:23, 42.26s/it]
                                                       
{'loss': '2.09', 'grad_norm': '32.23', 'learning_rate': '1.918e-05', 'epoch': '0.06'}

  6%|▌         | 300/5000 [4:02:56<55:10:23, 42.26s/it][2026-05-21 20:08:44,830] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 10])
[2026-05-21 20:08:45,551] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 19])

  6%|▌         | 301/5000 [4:03:30<51:46:06, 39.66s/it]
                                                       
{'loss': '0.873', 'grad_norm': '11.02', 'learning_rate': '1.918e-05', 'epoch': '0.0602'}

  6%|▌         | 301/5000 [4:03:30<51:46:06, 39.66s/it][2026-05-21 20:09:17,969] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 13])
[2026-05-21 20:09:18,593] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 13])

  6%|▌         | 302/5000 [4:04:04<49:38:28, 38.04s/it]
                                                       
{'loss': '0.8497', 'grad_norm': '9.471', 'learning_rate': '1.918e-05', 'epoch': '0.0604'}

  6%|▌         | 302/5000 [4:04:04<49:38:28, 38.04s/it][2026-05-21 20:10:01,678] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 110])
[2026-05-21 20:10:04,754] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 112])

  6%|▌         | 303/5000 [4:05:01<57:07:30, 43.78s/it]
                                                       
{'loss': '1.586', 'grad_norm': '13.43', 'learning_rate': '1.917e-05', 'epoch': '0.0606'}

  6%|▌         | 303/5000 [4:05:01<57:07:30, 43.78s/it][2026-05-21 20:11:02,418] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 97])
[2026-05-21 20:11:04,244] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 183])

  6%|▌         | 304/5000 [4:05:59<62:34:38, 47.97s/it]
                                                       
{'loss': '2.072', 'grad_norm': '13.76', 'learning_rate': '1.917e-05', 'epoch': '0.0608'}

  6%|▌         | 304/5000 [4:05:59<62:34:38, 47.97s/it][2026-05-21 20:11:47,457] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])
[2026-05-21 20:11:50,207] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 230])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])

  6%|▌         | 305/5000 [4:07:09<71:24:00, 54.75s/it]
                                                       
{'loss': '0.7796', 'grad_norm': '15.91', 'learning_rate': '1.916e-05', 'epoch': '0.061'}

  6%|▌         | 305/5000 [4:07:09<71:24:00, 54.75s/it][2026-05-21 20:12:58,814] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 172])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 20:12:59,231] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 435])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  6%|▌         | 306/5000 [4:08:19<77:18:43, 59.29s/it]
                                                       
{'loss': '1.463', 'grad_norm': '17.95', 'learning_rate': '1.916e-05', 'epoch': '0.0612'}

  6%|▌         | 306/5000 [4:08:19<77:18:43, 59.29s/it][2026-05-21 20:14:12,241] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 275])
[2026-05-21 20:14:15,653] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 303])

  6%|▌         | 307/5000 [4:09:06<72:23:31, 55.53s/it]
                                                       
{'loss': '1.028', 'grad_norm': '20.62', 'learning_rate': '1.916e-05', 'epoch': '0.0614'}

  6%|▌         | 307/5000 [4:09:06<72:23:31, 55.53s/it][2026-05-21 20:15:00,743] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 505])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 324])
[2026-05-21 20:15:00,844] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 260])

  6%|▌         | 308/5000 [4:09:51<68:16:59, 52.39s/it]
                                                       
{'loss': '1.072', 'grad_norm': '19.8', 'learning_rate': '1.915e-05', 'epoch': '0.0616'}

  6%|▌         | 308/5000 [4:09:51<68:16:59, 52.39s/it][2026-05-21 20:15:42,689] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 299])
[2026-05-21 20:15:43,028] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])

  6%|▌         | 309/5000 [4:10:41<67:25:27, 51.74s/it]
                                                       
{'loss': '0.306', 'grad_norm': '9.767', 'learning_rate': '1.915e-05', 'epoch': '0.0618'}

  6%|▌         | 309/5000 [4:10:41<67:25:27, 51.74s/it][2026-05-21 20:16:32,995] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 272])
[2026-05-21 20:16:33,709] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 234])

  6%|▌         | 310/5000 [4:11:33<67:12:18, 51.59s/it]
                                                       
{'loss': '0.3348', 'grad_norm': '8.755', 'learning_rate': '1.914e-05', 'epoch': '0.062'}

  6%|▌         | 310/5000 [4:11:33<67:12:18, 51.59s/it][2026-05-21 20:17:38,327] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 301])
[2026-05-21 20:17:39,213] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 433])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])

  6%|▌         | 311/5000 [4:12:19<65:14:00, 50.08s/it]
                                                       
{'loss': '0.6996', 'grad_norm': '10.78', 'learning_rate': '1.914e-05', 'epoch': '0.0622'}

  6%|▌         | 311/5000 [4:12:19<65:14:00, 50.08s/it][2026-05-21 20:18:31,021] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 286])
[2026-05-21 20:18:35,155] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 608])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 287])

  6%|▌         | 312/5000 [4:13:09<65:20:43, 50.18s/it]
                                                       
{'loss': '1.088', 'grad_norm': '10.15', 'learning_rate': '1.913e-05', 'epoch': '0.0624'}

  6%|▌         | 312/5000 [4:13:10<65:20:43, 50.18s/it][2026-05-21 20:19:00,016] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 28])
[2026-05-21 20:19:01,120] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 19])

  6%|▋         | 313/5000 [4:13:39<57:25:51, 44.11s/it]
                                                       
{'loss': '0.2267', 'grad_norm': '9.344', 'learning_rate': '1.913e-05', 'epoch': '0.0626'}

  6%|▋         | 313/5000 [4:13:39<57:25:51, 44.11s/it][2026-05-21 20:19:29,348] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1014])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 28])
[2026-05-21 20:19:30,478] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 28])

  6%|▋         | 314/5000 [4:14:14<53:47:00, 41.32s/it]
                                                       
{'loss': '0.1014', 'grad_norm': '7.403', 'learning_rate': '1.913e-05', 'epoch': '0.0628'}

  6%|▋         | 314/5000 [4:14:14<53:47:00, 41.32s/it][2026-05-21 20:20:01,585] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 248])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 247])
[2026-05-21 20:20:02,311] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

  6%|▋         | 315/5000 [4:15:05<57:21:16, 44.07s/it]
                                                       
{'loss': '0.4693', 'grad_norm': '10.97', 'learning_rate': '1.912e-05', 'epoch': '0.063'}

  6%|▋         | 315/5000 [4:15:05<57:21:16, 44.07s/it][2026-05-21 20:20:52,908] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 20:20:52,974] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  6%|▋         | 316/5000 [4:15:55<59:39:21, 45.85s/it]
                                                       
{'loss': '0.2442', 'grad_norm': '11.42', 'learning_rate': '1.912e-05', 'epoch': '0.0632'}

  6%|▋         | 316/5000 [4:15:55<59:39:21, 45.85s/it][2026-05-21 20:21:41,949] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 226])
[2026-05-21 20:21:42,237] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 340])

  6%|▋         | 317/5000 [4:16:24<53:14:02, 40.92s/it]
                                                       
{'loss': '0.611', 'grad_norm': '13.28', 'learning_rate': '1.911e-05', 'epoch': '0.0634'}

  6%|▋         | 317/5000 [4:16:24<53:14:02, 40.92s/it][2026-05-21 20:22:11,650] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 288])
[2026-05-21 20:22:11,733] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 493])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 256])

  6%|▋         | 318/5000 [4:17:09<54:49:09, 42.15s/it]
                                                       
{'loss': '0.1394', 'grad_norm': '7.46', 'learning_rate': '1.911e-05', 'epoch': '0.0636'}

  6%|▋         | 318/5000 [4:17:09<54:49:09, 42.15s/it][2026-05-21 20:23:00,186] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 100])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 20:23:00,414] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 104])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])

  6%|▋         | 319/5000 [4:18:00<58:16:41, 44.82s/it]
                                                       
{'loss': '0.6616', 'grad_norm': '19.78', 'learning_rate': '1.911e-05', 'epoch': '0.0638'}

  6%|▋         | 319/5000 [4:18:00<58:16:41, 44.82s/it][2026-05-21 20:23:51,876] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 109])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 20:23:54,609] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 76])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  6%|▋         | 320/5000 [4:18:51<60:23:38, 46.46s/it]
                                                       
{'loss': '0.7815', 'grad_norm': '19.99', 'learning_rate': '1.91e-05', 'epoch': '0.064'}

  6%|▋         | 320/5000 [4:18:51<60:23:38, 46.46s/it][2026-05-21 20:24:40,789] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 318])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 941])
[2026-05-21 20:24:41,671] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 271])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  6%|▋         | 321/5000 [4:19:22<54:30:58, 41.94s/it]
                                                       
{'loss': '0.427', 'grad_norm': '11.66', 'learning_rate': '1.91e-05', 'epoch': '0.0642'}

  6%|▋         | 321/5000 [4:19:22<54:30:58, 41.94s/it][2026-05-21 20:25:13,619] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 339])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-21 20:25:14,169] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 358])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  6%|▋         | 322/5000 [4:19:55<51:10:40, 39.38s/it]
                                                       
{'loss': '0.05897', 'grad_norm': '5.731', 'learning_rate': '1.909e-05', 'epoch': '0.0644'}

  6%|▋         | 322/5000 [4:19:55<51:10:40, 39.38s/it][2026-05-21 20:25:46,888] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 20:25:54,878] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1018])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  6%|▋         | 323/5000 [4:20:37<52:10:20, 40.16s/it]
                                                       
{'loss': '0.4509', 'grad_norm': '10.51', 'learning_rate': '1.909e-05', 'epoch': '0.0646'}

  6%|▋         | 323/5000 [4:20:37<52:10:20, 40.16s/it][2026-05-21 20:26:32,612] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 993])
[2026-05-21 20:26:35,047] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])

  6%|▋         | 324/5000 [4:21:16<51:30:41, 39.66s/it]
                                                       
{'loss': '0.2659', 'grad_norm': '7.71', 'learning_rate': '1.909e-05', 'epoch': '0.0648'}

  6%|▋         | 324/5000 [4:21:16<51:30:41, 39.66s/it][2026-05-21 20:27:17,967] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 201])
[2026-05-21 20:27:19,600] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 160])

  6%|▋         | 325/5000 [4:22:11<57:33:33, 44.32s/it]
                                                       
{'loss': '0.9753', 'grad_norm': '14.84', 'learning_rate': '1.908e-05', 'epoch': '0.065'}

  6%|▋         | 325/5000 [4:22:11<57:33:33, 44.32s/it][2026-05-21 20:28:12,662] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 349])
[2026-05-21 20:28:13,840] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])

  7%|▋         | 326/5000 [4:23:05<61:16:29, 47.19s/it]
                                                       
{'loss': '0.8888', 'grad_norm': '17.26', 'learning_rate': '1.908e-05', 'epoch': '0.0652'}

  7%|▋         | 326/5000 [4:23:05<61:16:29, 47.19s/it][2026-05-21 20:29:15,456] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 20:29:15,565] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])

  7%|▋         | 327/5000 [4:23:53<61:32:44, 47.41s/it]
                                                       
{'loss': '0.4046', 'grad_norm': '12.68', 'learning_rate': '1.907e-05', 'epoch': '0.0654'}

  7%|▋         | 327/5000 [4:23:53<61:32:44, 47.41s/it][2026-05-21 20:30:01,196] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 20:30:05,994] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  7%|▋         | 328/5000 [4:24:44<63:02:23, 48.58s/it]
                                                       
{'loss': '0.4725', 'grad_norm': '8.176', 'learning_rate': '1.907e-05', 'epoch': '0.0656'}

  7%|▋         | 328/5000 [4:24:44<63:02:23, 48.58s/it][2026-05-21 20:30:33,836] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 261])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 20:30:35,879] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 297])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])

  7%|▋         | 329/5000 [4:25:55<71:49:39, 55.36s/it]
                                                       
{'loss': '1.116', 'grad_norm': '13.33', 'learning_rate': '1.907e-05', 'epoch': '0.0658'}

  7%|▋         | 329/5000 [4:25:55<71:49:39, 55.36s/it][2026-05-21 20:31:44,620] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 266])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 20:31:46,420] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 287])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  7%|▋         | 330/5000 [4:27:06<77:58:29, 60.11s/it]
                                                       
{'loss': '0.8101', 'grad_norm': '13.09', 'learning_rate': '1.906e-05', 'epoch': '0.066'}

  7%|▋         | 330/5000 [4:27:06<77:58:29, 60.11s/it][2026-05-21 20:33:06,987] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 211])
[2026-05-21 20:33:07,773] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 176])

  7%|▋         | 331/5000 [4:27:58<74:32:29, 57.47s/it]
                                                       
{'loss': '0.5084', 'grad_norm': '11.87', 'learning_rate': '1.906e-05', 'epoch': '0.0662'}

  7%|▋         | 331/5000 [4:27:58<74:32:29, 57.47s/it][2026-05-21 20:33:58,290] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 230])
[2026-05-21 20:33:59,100] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 220])

  7%|▋         | 332/5000 [4:28:50<72:18:08, 55.76s/it]
                                                       
{'loss': '0.4743', 'grad_norm': '15.81', 'learning_rate': '1.905e-05', 'epoch': '0.0664'}

  7%|▋         | 332/5000 [4:28:50<72:18:08, 55.76s/it][2026-05-21 20:34:35,939] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 124])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 25])
[2026-05-21 20:34:36,478] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 156])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 25])

  7%|▋         | 333/5000 [4:29:58<77:22:54, 59.69s/it]
                                                       
{'loss': '0.6351', 'grad_norm': '9.702', 'learning_rate': '1.905e-05', 'epoch': '0.0666'}

  7%|▋         | 333/5000 [4:29:58<77:22:54, 59.69s/it][2026-05-21 20:35:44,807] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 180])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 22])
[2026-05-21 20:35:45,370] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 180])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 26])

  7%|▋         | 334/5000 [4:31:08<81:14:11, 62.68s/it]
                                                       
{'loss': '0.4942', 'grad_norm': '9.147', 'learning_rate': '1.904e-05', 'epoch': '0.0668'}

  7%|▋         | 334/5000 [4:31:08<81:14:11, 62.68s/it][2026-05-21 20:36:55,967] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 220])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 291])
[2026-05-21 20:36:56,061] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  7%|▋         | 335/5000 [4:31:55<74:59:43, 57.87s/it]
                                                       
{'loss': '0.8341', 'grad_norm': '11.4', 'learning_rate': '1.904e-05', 'epoch': '0.067'}

  7%|▋         | 335/5000 [4:31:55<74:59:43, 57.87s/it][2026-05-21 20:37:42,838] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 20:37:43,589] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  7%|▋         | 336/5000 [4:32:42<70:49:42, 54.67s/it]
                                                       
{'loss': '0.6185', 'grad_norm': '7.516', 'learning_rate': '1.904e-05', 'epoch': '0.0672'}

  7%|▋         | 336/5000 [4:32:42<70:49:42, 54.67s/it][2026-05-21 20:38:33,303] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 975])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 185])
[2026-05-21 20:38:34,819] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

  7%|▋         | 337/5000 [4:33:33<69:35:58, 53.73s/it]
                                                       
{'loss': '0.4182', 'grad_norm': '12.32', 'learning_rate': '1.903e-05', 'epoch': '0.0674'}

  7%|▋         | 337/5000 [4:33:33<69:35:58, 53.73s/it][2026-05-21 20:39:25,293] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 20:39:29,176] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1014])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  7%|▋         | 338/5000 [4:34:25<68:49:35, 53.15s/it]
                                                       
{'loss': '0.3457', 'grad_norm': '9.788', 'learning_rate': '1.903e-05', 'epoch': '0.0676'}

  7%|▋         | 338/5000 [4:34:25<68:49:35, 53.15s/it][2026-05-21 20:40:15,197] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-21 20:40:15,828] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])

  7%|▋         | 339/5000 [4:34:49<57:31:04, 44.42s/it]
                                                       
{'loss': '0.5032', 'grad_norm': '11.66', 'learning_rate': '1.902e-05', 'epoch': '0.0678'}

  7%|▋         | 339/5000 [4:34:49<57:31:04, 44.42s/it][2026-05-21 20:40:38,662] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 20:40:40,105] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  7%|▋         | 340/5000 [4:35:14<49:39:56, 38.37s/it]
                                                       
{'loss': '0.4897', 'grad_norm': '10.9', 'learning_rate': '1.902e-05', 'epoch': '0.068'}

  7%|▋         | 340/5000 [4:35:14<49:39:56, 38.37s/it][2026-05-21 20:41:03,248] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 299])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 20:41:03,919] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 230])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  7%|▋         | 341/5000 [4:35:47<47:37:12, 36.80s/it]
                                                       
{'loss': '0.5528', 'grad_norm': '13.7', 'learning_rate': '1.902e-05', 'epoch': '0.0682'}

  7%|▋         | 341/5000 [4:35:47<47:37:12, 36.80s/it][2026-05-21 20:41:36,508] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 255])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])
[2026-05-21 20:41:38,177] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 249])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  7%|▋         | 342/5000 [4:36:19<45:59:35, 35.55s/it]
                                                       
{'loss': '0.04468', 'grad_norm': '4.497', 'learning_rate': '1.901e-05', 'epoch': '0.0684'}

  7%|▋         | 342/5000 [4:36:19<45:59:35, 35.55s/it][2026-05-21 20:42:09,122] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1020])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 20:42:09,302] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

  7%|▋         | 343/5000 [4:37:11<52:06:54, 40.29s/it]
                                                       
{'loss': '0.4236', 'grad_norm': '8.856', 'learning_rate': '1.901e-05', 'epoch': '0.0686'}

  7%|▋         | 343/5000 [4:37:11<52:06:54, 40.29s/it][2026-05-21 20:43:00,477] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-21 20:43:01,170] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  7%|▋         | 344/5000 [4:38:04<57:07:12, 44.17s/it]
                                                       
{'loss': '0.1453', 'grad_norm': '18.09', 'learning_rate': '1.9e-05', 'epoch': '0.0688'}

  7%|▋         | 344/5000 [4:38:04<57:07:12, 44.17s/it][2026-05-21 20:44:01,741] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1019])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 94])
[2026-05-21 20:44:02,857] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 73])

  7%|▋         | 345/5000 [4:39:12<66:13:39, 51.22s/it]
                                                       
{'loss': '2.942', 'grad_norm': '15.02', 'learning_rate': '1.9e-05', 'epoch': '0.069'}

  7%|▋         | 345/5000 [4:39:12<66:13:39, 51.22s/it][2026-05-21 20:45:08,064] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 65])
[2026-05-21 20:45:08,776] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 99])

  7%|▋         | 346/5000 [4:40:18<72:05:26, 55.76s/it]
                                                       
{'loss': '2.957', 'grad_norm': '12.55', 'learning_rate': '1.9e-05', 'epoch': '0.0692'}

  7%|▋         | 346/5000 [4:40:18<72:05:26, 55.76s/it][2026-05-21 20:46:05,163] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 235])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-21 20:46:05,678] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

  7%|▋         | 347/5000 [4:41:04<68:26:45, 52.96s/it]
                                                       
{'loss': '0.6088', 'grad_norm': '10.69', 'learning_rate': '1.899e-05', 'epoch': '0.0694'}

  7%|▋         | 347/5000 [4:41:04<68:26:45, 52.96s/it][2026-05-21 20:46:51,489] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 222])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-21 20:46:53,304] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 531])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

  7%|▋         | 348/5000 [4:41:49<65:01:07, 50.32s/it]
                                                       
{'loss': '0.3566', 'grad_norm': '9.105', 'learning_rate': '1.899e-05', 'epoch': '0.0696'}

  7%|▋         | 348/5000 [4:41:49<65:01:07, 50.32s/it][2026-05-21 20:47:35,839] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 180])
[2026-05-21 20:47:36,249] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 398])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

  7%|▋         | 349/5000 [4:42:39<65:02:39, 50.35s/it]
                                                       
{'loss': '0.4677', 'grad_norm': '11.22', 'learning_rate': '1.898e-05', 'epoch': '0.0698'}

  7%|▋         | 349/5000 [4:42:39<65:02:39, 50.35s/it][2026-05-21 20:48:26,150] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 129])
[2026-05-21 20:48:26,753] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

  7%|▋         | 350/5000 [4:43:28<64:36:38, 50.02s/it]
                                                       
{'loss': '0.06434', 'grad_norm': '6.606', 'learning_rate': '1.898e-05', 'epoch': '0.07'}

  7%|▋         | 350/5000 [4:43:28<64:36:38, 50.02s/it][2026-05-21 20:49:15,634] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 803])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 20:49:15,750] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 239])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

  7%|▋         | 351/5000 [4:44:15<63:12:56, 48.95s/it]
                                                       
{'loss': '0.8892', 'grad_norm': '16.83', 'learning_rate': '1.898e-05', 'epoch': '0.0702'}

  7%|▋         | 351/5000 [4:44:15<63:12:56, 48.95s/it][2026-05-21 20:50:02,244] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 950])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 20:50:02,771] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 972])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

  7%|▋         | 352/5000 [4:45:00<61:47:53, 47.86s/it]
                                                       
{'loss': '0.8438', 'grad_norm': '13.23', 'learning_rate': '1.897e-05', 'epoch': '0.0704'}

  7%|▋         | 352/5000 [4:45:00<61:47:53, 47.86s/it][2026-05-21 20:50:47,868] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 648])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 11])
[2026-05-21 20:50:49,039] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 925])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 6])

  7%|▋         | 353/5000 [4:45:25<52:55:43, 41.00s/it]
                                                       
{'loss': '0.1826', 'grad_norm': '8.297', 'learning_rate': '1.897e-05', 'epoch': '0.0706'}

  7%|▋         | 353/5000 [4:45:25<52:55:43, 41.00s/it][2026-05-21 20:51:13,355] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 981])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 8])
[2026-05-21 20:51:13,996] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 981])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 7])

  7%|▋         | 354/5000 [4:45:56<49:00:10, 37.97s/it]
                                                       
{'loss': '0.2024', 'grad_norm': '7.721', 'learning_rate': '1.896e-05', 'epoch': '0.0708'}

  7%|▋         | 354/5000 [4:45:56<49:00:10, 37.97s/it][2026-05-21 20:51:42,597] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 178])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 111])
[2026-05-21 20:51:42,626] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 220])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])

  7%|▋         | 355/5000 [4:46:22<44:35:56, 34.57s/it]
                                                       
{'loss': '1.281', 'grad_norm': '19.02', 'learning_rate': '1.896e-05', 'epoch': '0.071'}

  7%|▋         | 355/5000 [4:46:22<44:35:56, 34.57s/it][2026-05-21 20:52:09,204] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 202])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-21 20:52:09,217] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 202])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

  7%|▋         | 356/5000 [4:46:49<41:31:30, 32.19s/it]
                                                       
{'loss': '2.001', 'grad_norm': '29.97', 'learning_rate': '1.896e-05', 'epoch': '0.0712'}

  7%|▋         | 356/5000 [4:46:49<41:31:30, 32.19s/it][2026-05-21 20:52:35,306] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 202])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-21 20:52:35,327] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 202])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

  7%|▋         | 357/5000 [4:47:29<44:38:44, 34.62s/it]
                                                       
{'loss': '0.4491', 'grad_norm': '12.12', 'learning_rate': '1.895e-05', 'epoch': '0.0714'}

  7%|▋         | 357/5000 [4:47:29<44:38:44, 34.62s/it][2026-05-21 20:53:15,602] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 206])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-21 20:53:15,654] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 190])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

  7%|▋         | 358/5000 [4:48:17<49:40:57, 38.53s/it]
                                                       
{'loss': '0.345', 'grad_norm': '11.9', 'learning_rate': '1.895e-05', 'epoch': '0.0716'}

  7%|▋         | 358/5000 [4:48:17<49:40:57, 38.53s/it][2026-05-21 20:54:08,877] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 100])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 20:54:10,672] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 113])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  7%|▋         | 359/5000 [4:49:02<52:12:06, 40.49s/it]
                                                       
{'loss': '2.382', 'grad_norm': '18.72', 'learning_rate': '1.894e-05', 'epoch': '0.0718'}

  7%|▋         | 359/5000 [4:49:02<52:12:06, 40.49s/it][2026-05-21 20:54:54,154] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 98])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 20:54:58,602] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 97])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  7%|▋         | 360/5000 [4:49:49<54:38:38, 42.40s/it]
                                                       
{'loss': '2.221', 'grad_norm': '16.93', 'learning_rate': '1.894e-05', 'epoch': '0.072'}

  7%|▋         | 360/5000 [4:49:49<54:38:38, 42.40s/it][2026-05-21 20:55:44,000] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 20:55:44,801] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  7%|▋         | 361/5000 [4:50:20<50:08:10, 38.91s/it]
                                                       
{'loss': '2.396', 'grad_norm': '10.56', 'learning_rate': '1.893e-05', 'epoch': '0.0722'}

  7%|▋         | 361/5000 [4:50:20<50:08:10, 38.91s/it][2026-05-21 20:56:15,286] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 20:56:17,644] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  7%|▋         | 362/5000 [4:50:47<45:47:23, 35.54s/it]
                                                       
{'loss': '2.433', 'grad_norm': '11.44', 'learning_rate': '1.893e-05', 'epoch': '0.0724'}

  7%|▋         | 362/5000 [4:50:47<45:47:23, 35.54s/it]
  7%|▋         | 363/5000 [4:51:11<40:57:47, 31.80s/it]
                                                       
{'loss': '2.563', 'grad_norm': '12.73', 'learning_rate': '1.893e-05', 'epoch': '0.0726'}

  7%|▋         | 363/5000 [4:51:11<40:57:47, 31.80s/it][2026-05-21 20:56:58,793] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 20:56:58,933] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 20:57:23,780] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  7%|▋         | 364/5000 [4:51:40<40:10:59, 31.20s/it]
                                                       
{'loss': '2.467', 'grad_norm': '10.23', 'learning_rate': '1.892e-05', 'epoch': '0.0728'}

  7%|▋         | 364/5000 [4:51:40<40:10:59, 31.20s/it][2026-05-21 20:57:28,344] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 20:57:38,218] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 285])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 20:57:42,447] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 265])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  7%|▋         | 365/5000 [4:52:34<48:40:39, 37.81s/it]
                                                       
{'loss': '1.523', 'grad_norm': '13.98', 'learning_rate': '1.892e-05', 'epoch': '0.073'}

  7%|▋         | 365/5000 [4:52:34<48:40:39, 37.81s/it][2026-05-21 20:58:35,739] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 265])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 961])
[2026-05-21 20:58:42,590] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  7%|▋         | 366/5000 [4:53:26<54:16:46, 42.17s/it]
                                                       
{'loss': '0.5801', 'grad_norm': '12.58', 'learning_rate': '1.891e-05', 'epoch': '0.0732'}

  7%|▋         | 366/5000 [4:53:26<54:16:46, 42.17s/it][2026-05-21 20:59:13,748] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 20:59:14,287] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  7%|▋         | 367/5000 [4:54:37<65:30:25, 50.90s/it]
                                                       
{'loss': '0.4367', 'grad_norm': '12.56', 'learning_rate': '1.891e-05', 'epoch': '0.0734'}

  7%|▋         | 367/5000 [4:54:37<65:30:25, 50.90s/it][2026-05-21 21:00:25,566] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 21:00:25,737] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 618])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

  7%|▋         | 368/5000 [4:55:49<73:40:05, 57.26s/it]
                                                       
{'loss': '0.6726', 'grad_norm': '13.06', 'learning_rate': '1.891e-05', 'epoch': '0.0736'}

  7%|▋         | 368/5000 [4:55:49<73:40:05, 57.26s/it][2026-05-21 21:01:35,961] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-21 21:01:35,996] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

  7%|▋         | 369/5000 [4:57:00<78:44:44, 61.21s/it]
                                                       
{'loss': '0.5567', 'grad_norm': '12.2', 'learning_rate': '1.89e-05', 'epoch': '0.0738'}

  7%|▋         | 369/5000 [4:57:00<78:44:44, 61.21s/it][2026-05-21 21:02:46,447] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-21 21:02:46,483] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])

  7%|▋         | 370/5000 [4:58:09<81:52:01, 63.65s/it]
                                                       
{'loss': '0.7214', 'grad_norm': '10.06', 'learning_rate': '1.89e-05', 'epoch': '0.074'}

  7%|▋         | 370/5000 [4:58:09<81:52:01, 63.65s/it][2026-05-21 21:03:56,082] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 148])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 22])
[2026-05-21 21:03:56,113] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 172])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 39])

  7%|▋         | 371/5000 [4:58:56<75:23:44, 58.64s/it]
                                                       
{'loss': '0.5325', 'grad_norm': '10.36', 'learning_rate': '1.889e-05', 'epoch': '0.0742'}

  7%|▋         | 371/5000 [4:58:56<75:23:44, 58.64s/it][2026-05-21 21:04:42,911] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 204])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 18])
[2026-05-21 21:04:42,929] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 188])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 25])

  7%|▋         | 372/5000 [4:59:43<70:48:06, 55.07s/it]
                                                       
{'loss': '0.321', 'grad_norm': '8.086', 'learning_rate': '1.889e-05', 'epoch': '0.0744'}

  7%|▋         | 372/5000 [4:59:43<70:48:06, 55.07s/it][2026-05-21 21:05:33,086] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 140])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])
[2026-05-21 21:05:33,395] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 140])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

  7%|▋         | 373/5000 [5:00:21<64:28:25, 50.16s/it]
                                                       
{'loss': '1.968', 'grad_norm': '17.16', 'learning_rate': '1.889e-05', 'epoch': '0.0746'}

  7%|▋         | 373/5000 [5:00:21<64:28:25, 50.16s/it][2026-05-21 21:06:12,498] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 148])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 21:06:12,750] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 148])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  7%|▋         | 374/5000 [5:01:02<60:47:38, 47.31s/it]
                                                       
{'loss': '1.63', 'grad_norm': '14.5', 'learning_rate': '1.888e-05', 'epoch': '0.0748'}

  7%|▋         | 374/5000 [5:01:02<60:47:38, 47.31s/it][2026-05-21 21:07:18,901] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

  8%|▊         | 375/5000 [5:01:35<55:15:32, 43.01s/it]
                                                       
{'loss': '1.947', 'grad_norm': '16.88', 'learning_rate': '1.888e-05', 'epoch': '0.075'}

  8%|▊         | 375/5000 [5:01:35<55:15:32, 43.01s/it][2026-05-21 21:07:22,121] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 180])

  8%|▊         | 376/5000 [5:02:12<52:48:11, 41.11s/it]
                                                       
{'loss': '1.772', 'grad_norm': '13.41', 'learning_rate': '1.887e-05', 'epoch': '0.0752'}

  8%|▊         | 376/5000 [5:02:12<52:48:11, 41.11s/it][2026-05-21 21:07:59,818] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 198])
[2026-05-21 21:08:01,518] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])
[2026-05-21 21:08:03,689] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 89])
[2026-05-21 21:08:03,719] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 179])

  8%|▊         | 377/5000 [5:02:42<48:35:57, 37.84s/it]
                                                       
{'loss': '0.6085', 'grad_norm': '14.19', 'learning_rate': '1.887e-05', 'epoch': '0.0754'}

  8%|▊         | 377/5000 [5:02:42<48:35:57, 37.84s/it][2026-05-21 21:08:33,283] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 138])
[2026-05-21 21:08:33,477] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 88])

  8%|▊         | 378/5000 [5:03:07<43:46:57, 34.10s/it]
                                                       
{'loss': '0.6783', 'grad_norm': '14.29', 'learning_rate': '1.887e-05', 'epoch': '0.0756'}

  8%|▊         | 378/5000 [5:03:07<43:46:57, 34.10s/it][2026-05-21 21:09:08,718] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])
[2026-05-21 21:09:10,182] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  8%|▊         | 379/5000 [5:03:58<50:12:50, 39.12s/it]
                                                       
{'loss': '1.418', 'grad_norm': '15.38', 'learning_rate': '1.886e-05', 'epoch': '0.0758'}

  8%|▊         | 379/5000 [5:03:58<50:12:50, 39.12s/it][2026-05-21 21:10:00,331] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 21:10:00,421] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 981])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  8%|▊         | 380/5000 [5:04:48<54:09:06, 42.20s/it]
                                                       
{'loss': '0.7443', 'grad_norm': '9.811', 'learning_rate': '1.886e-05', 'epoch': '0.076'}

  8%|▊         | 380/5000 [5:04:48<54:09:06, 42.20s/it][2026-05-21 21:10:36,706] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 21:10:37,865] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  8%|▊         | 381/5000 [5:05:37<56:48:52, 44.28s/it]
                                                       
{'loss': '1.377', 'grad_norm': '13.02', 'learning_rate': '1.885e-05', 'epoch': '0.0762'}

  8%|▊         | 381/5000 [5:05:37<56:48:52, 44.28s/it][2026-05-21 21:11:26,762] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-21 21:11:28,636] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  8%|▊         | 382/5000 [5:06:27<59:12:40, 46.16s/it]
                                                       
{'loss': '1.293', 'grad_norm': '12.55', 'learning_rate': '1.885e-05', 'epoch': '0.0764'}

  8%|▊         | 382/5000 [5:06:27<59:12:40, 46.16s/it][2026-05-21 21:12:14,323] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 109])
[2026-05-21 21:12:14,382] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])

  8%|▊         | 383/5000 [5:07:14<59:21:04, 46.28s/it]
                                                       
{'loss': '1.224', 'grad_norm': '11.39', 'learning_rate': '1.884e-05', 'epoch': '0.0766'}

  8%|▊         | 383/5000 [5:07:14<59:21:04, 46.28s/it][2026-05-21 21:13:00,636] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 220])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 56])
[2026-05-21 21:13:01,370] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 432])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])

  8%|▊         | 384/5000 [5:08:01<59:41:09, 46.55s/it]
                                                       
{'loss': '1.351', 'grad_norm': '10.15', 'learning_rate': '1.884e-05', 'epoch': '0.0768'}

  8%|▊         | 384/5000 [5:08:01<59:41:09, 46.55s/it][2026-05-21 21:13:52,207] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])
[2026-05-21 21:13:52,477] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 54])

  8%|▊         | 385/5000 [5:09:11<68:45:47, 53.64s/it]
                                                       
{'loss': '1.263', 'grad_norm': '11.18', 'learning_rate': '1.884e-05', 'epoch': '0.077'}

  8%|▊         | 385/5000 [5:09:11<68:45:47, 53.64s/it][2026-05-21 21:15:03,735] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 52])
[2026-05-21 21:15:04,362] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 60])

  8%|▊         | 386/5000 [5:10:22<75:25:02, 58.84s/it]
                                                       
{'loss': '1.079', 'grad_norm': '9.305', 'learning_rate': '1.883e-05', 'epoch': '0.0772'}

  8%|▊         | 386/5000 [5:10:22<75:25:02, 58.84s/it][2026-05-21 21:16:10,102] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 278])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 21:16:10,947] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 311])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  8%|▊         | 387/5000 [5:11:13<72:31:40, 56.60s/it]
                                                       
{'loss': '0.2935', 'grad_norm': '9.565', 'learning_rate': '1.883e-05', 'epoch': '0.0774'}

  8%|▊         | 387/5000 [5:11:14<72:31:40, 56.60s/it][2026-05-21 21:17:02,233] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 297])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 21:17:02,301] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 296])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

  8%|▊         | 388/5000 [5:12:04<70:17:57, 54.87s/it]
                                                       
{'loss': '0.06144', 'grad_norm': '4.921', 'learning_rate': '1.882e-05', 'epoch': '0.0776'}

  8%|▊         | 388/5000 [5:12:04<70:17:57, 54.87s/it][2026-05-21 21:17:51,751] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 236])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 21:17:52,023] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 351])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  8%|▊         | 389/5000 [5:12:27<57:55:46, 45.23s/it]
                                                       
{'loss': '2.588', 'grad_norm': '14.9', 'learning_rate': '1.882e-05', 'epoch': '0.0778'}

  8%|▊         | 389/5000 [5:12:27<57:55:46, 45.23s/it][2026-05-21 21:18:14,486] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 266])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 21:18:15,900] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 438])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  8%|▊         | 390/5000 [5:12:53<50:40:34, 39.57s/it]
                                                       
{'loss': '2.263', 'grad_norm': '10.41', 'learning_rate': '1.882e-05', 'epoch': '0.078'}

  8%|▊         | 390/5000 [5:12:53<50:40:34, 39.57s/it][2026-05-21 21:18:41,342] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 280])
[2026-05-21 21:18:42,042] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 334])

  8%|▊         | 391/5000 [5:13:40<53:19:06, 41.65s/it]
                                                       
{'loss': '0.1759', 'grad_norm': '13.04', 'learning_rate': '1.881e-05', 'epoch': '0.0782'}

  8%|▊         | 391/5000 [5:13:40<53:19:06, 41.65s/it][2026-05-21 21:19:28,719] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 287])
[2026-05-21 21:19:28,748] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 242])

  8%|▊         | 392/5000 [5:14:27<55:13:49, 43.15s/it]
                                                       
{'loss': '0.2539', 'grad_norm': '3.537', 'learning_rate': '1.881e-05', 'epoch': '0.0784'}

  8%|▊         | 392/5000 [5:14:27<55:13:49, 43.15s/it][2026-05-21 21:20:16,241] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 493])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-21 21:20:20,298] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 212])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

  8%|▊         | 393/5000 [5:14:59<51:00:58, 39.87s/it]
                                                       
{'loss': '0.4398', 'grad_norm': '11.43', 'learning_rate': '1.88e-05', 'epoch': '0.0786'}

  8%|▊         | 393/5000 [5:14:59<51:00:58, 39.87s/it][2026-05-21 21:20:47,942] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 21:20:49,697] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 471])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  8%|▊         | 394/5000 [5:15:32<48:22:23, 37.81s/it]
                                                       
{'loss': '0.05268', 'grad_norm': '6.777', 'learning_rate': '1.88e-05', 'epoch': '0.0788'}

  8%|▊         | 394/5000 [5:15:32<48:22:23, 37.81s/it][2026-05-21 21:21:21,110] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 272])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])
[2026-05-21 21:21:22,129] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 787])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 312])

  8%|▊         | 395/5000 [5:16:00<44:49:15, 35.04s/it]
                                                       
{'loss': '0.2427', 'grad_norm': '9.814', 'learning_rate': '1.88e-05', 'epoch': '0.079'}

  8%|▊         | 395/5000 [5:16:00<44:49:15, 35.04s/it][2026-05-21 21:21:49,229] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 297])
[2026-05-21 21:21:49,824] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 344])

  8%|▊         | 396/5000 [5:16:33<43:44:50, 34.21s/it]
                                                       
{'loss': '0.0853', 'grad_norm': '5.817', 'learning_rate': '1.879e-05', 'epoch': '0.0792'}

  8%|▊         | 396/5000 [5:16:33<43:44:50, 34.21s/it][2026-05-21 21:22:21,369] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 21:22:21,744] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  8%|▊         | 397/5000 [5:17:05<42:56:24, 33.58s/it]
                                                       
{'loss': '0.4803', 'grad_norm': '11.6', 'learning_rate': '1.879e-05', 'epoch': '0.0794'}

  8%|▊         | 397/5000 [5:17:05<42:56:24, 33.58s/it][2026-05-21 21:22:53,038] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 243])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 21:22:56,838] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 277])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 993])

  8%|▊         | 398/5000 [5:17:36<42:07:37, 32.95s/it]
                                                       
{'loss': '0.05827', 'grad_norm': '4.812', 'learning_rate': '1.878e-05', 'epoch': '0.0796'}

  8%|▊         | 398/5000 [5:17:36<42:07:37, 32.95s/it][2026-05-21 21:23:28,400] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 270])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 21:23:28,667] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 255])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  8%|▊         | 399/5000 [5:18:28<49:23:56, 38.65s/it]
                                                       
{'loss': '1.281', 'grad_norm': '13.11', 'learning_rate': '1.878e-05', 'epoch': '0.0798'}

  8%|▊         | 399/5000 [5:18:28<49:23:56, 38.65s/it][2026-05-21 21:24:19,971] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 256])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 21:24:24,240] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 300])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  8%|▊         | 400/5000 [5:19:22<55:04:18, 43.10s/it]
                                                       
{'loss': '0.6738', 'grad_norm': '10.04', 'learning_rate': '1.878e-05', 'epoch': '0.08'}

  8%|▊         | 400/5000 [5:19:22<55:04:18, 43.10s/it][2026-05-21 21:25:13,008] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1019])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 21:25:13,288] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

  8%|▊         | 401/5000 [5:20:02<54:03:58, 42.32s/it]
                                                       
{'loss': '0.5589', 'grad_norm': '11.57', 'learning_rate': '1.877e-05', 'epoch': '0.0802'}

  8%|▊         | 401/5000 [5:20:02<54:03:58, 42.32s/it][2026-05-21 21:25:53,816] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 21:25:53,876] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  8%|▊         | 402/5000 [5:20:33<49:41:15, 38.90s/it]
                                                       
{'loss': '0.2496', 'grad_norm': '9.705', 'learning_rate': '1.877e-05', 'epoch': '0.0804'}

  8%|▊         | 402/5000 [5:20:33<49:41:15, 38.90s/it][2026-05-21 21:26:41,678] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 21:26:44,659] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  8%|▊         | 403/5000 [5:21:35<58:30:27, 45.82s/it]
                                                       
{'loss': '1.44', 'grad_norm': '13.07', 'learning_rate': '1.876e-05', 'epoch': '0.0806'}

  8%|▊         | 403/5000 [5:21:36<58:30:27, 45.82s/it][2026-05-21 21:27:45,819] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 21:27:48,078] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  8%|▊         | 404/5000 [5:22:29<61:28:02, 48.15s/it]
                                                       
{'loss': '0.67', 'grad_norm': '17.34', 'learning_rate': '1.876e-05', 'epoch': '0.0808'}

  8%|▊         | 404/5000 [5:22:29<61:28:02, 48.15s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-21 21:28:17,529] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 98])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 21:28:18,637] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 79])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  8%|▊         | 405/5000 [5:23:15<60:44:42, 47.59s/it]
                                                       
{'loss': '1.288', 'grad_norm': '10.57', 'learning_rate': '1.876e-05', 'epoch': '0.081'}

  8%|▊         | 405/5000 [5:23:15<60:44:42, 47.59s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-21 21:29:04,190] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 91])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 21:29:06,991] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 71])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  8%|▊         | 406/5000 [5:24:05<61:48:44, 48.44s/it]
                                                       
{'loss': '0.6035', 'grad_norm': '12.36', 'learning_rate': '1.875e-05', 'epoch': '0.0812'}

  8%|▊         | 406/5000 [5:24:05<61:48:44, 48.44s/it][2026-05-21 21:29:54,789] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 64])
[2026-05-21 21:29:55,960] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 100])

  8%|▊         | 407/5000 [5:24:59<63:52:30, 50.07s/it]
                                                       
{'loss': '0.4683', 'grad_norm': '13.09', 'learning_rate': '1.875e-05', 'epoch': '0.0814'}

  8%|▊         | 407/5000 [5:24:59<63:52:30, 50.07s/it][2026-05-21 21:30:48,649] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 115])
[2026-05-21 21:30:49,187] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 162])

  8%|▊         | 408/5000 [5:25:51<64:25:33, 50.51s/it]
                                                       
{'loss': '0.3452', 'grad_norm': '9.821', 'learning_rate': '1.874e-05', 'epoch': '0.0816'}

  8%|▊         | 408/5000 [5:25:51<64:25:33, 50.51s/it][2026-05-21 21:31:49,514] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 223])
[2026-05-21 21:31:51,727] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 176])

  8%|▊         | 409/5000 [5:27:04<73:10:12, 57.38s/it]
                                                       
{'loss': '0.6895', 'grad_norm': '10.18', 'learning_rate': '1.874e-05', 'epoch': '0.0818'}

  8%|▊         | 409/5000 [5:27:04<73:10:12, 57.38s/it][2026-05-21 21:33:04,911] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 153])
[2026-05-21 21:33:06,225] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 139])

  8%|▊         | 410/5000 [5:28:18<79:34:30, 62.41s/it]
                                                       
{'loss': '0.3557', 'grad_norm': '12.02', 'learning_rate': '1.873e-05', 'epoch': '0.082'}

  8%|▊         | 410/5000 [5:28:18<79:34:30, 62.41s/it][2026-05-21 21:34:14,801] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 155])
[2026-05-21 21:34:16,969] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 137])

  8%|▊         | 411/5000 [5:29:05<73:22:06, 57.56s/it]
                                                       
{'loss': '0.3858', 'grad_norm': '14.25', 'learning_rate': '1.873e-05', 'epoch': '0.0822'}

  8%|▊         | 411/5000 [5:29:05<73:22:06, 57.56s/it][2026-05-21 21:35:00,007] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 95])
[2026-05-21 21:35:01,436] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 92])

  8%|▊         | 412/5000 [5:29:51<68:57:36, 54.11s/it]
                                                       
{'loss': '0.4511', 'grad_norm': '8.994', 'learning_rate': '1.873e-05', 'epoch': '0.0824'}

  8%|▊         | 412/5000 [5:29:51<68:57:36, 54.11s/it][2026-05-21 21:35:51,223] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 182])
[2026-05-21 21:35:54,052] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 174])

  8%|▊         | 413/5000 [5:30:39<66:52:59, 52.49s/it]
                                                       
{'loss': '0.9559', 'grad_norm': '18.09', 'learning_rate': '1.872e-05', 'epoch': '0.0826'}

  8%|▊         | 413/5000 [5:30:39<66:52:59, 52.49s/it][2026-05-21 21:36:40,328] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 137])
[2026-05-21 21:36:41,432] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 149])

  8%|▊         | 414/5000 [5:31:29<65:51:12, 51.69s/it]
                                                       
{'loss': '0.8183', 'grad_norm': '17.24', 'learning_rate': '1.872e-05', 'epoch': '0.0828'}

  8%|▊         | 414/5000 [5:31:29<65:51:12, 51.69s/it][2026-05-21 21:37:50,067] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 182])
[2026-05-21 21:37:51,107] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1012])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 89])

  8%|▊         | 415/5000 [5:32:27<68:03:58, 53.44s/it]
                                                       
{'loss': '0.5216', 'grad_norm': '14.02', 'learning_rate': '1.871e-05', 'epoch': '0.083'}

  8%|▊         | 415/5000 [5:32:27<68:03:58, 53.44s/it][2026-05-21 21:38:47,693] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 117])
[2026-05-21 21:38:57,709] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 95])

  8%|▊         | 416/5000 [5:33:20<68:06:34, 53.49s/it]
                                                       
{'loss': '0.4745', 'grad_norm': '12.24', 'learning_rate': '1.871e-05', 'epoch': '0.0832'}

  8%|▊         | 416/5000 [5:33:20<68:06:34, 53.49s/it][2026-05-21 21:39:21,069] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 21:39:26,387] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

  8%|▊         | 417/5000 [5:34:11<67:11:11, 52.78s/it]
                                                       
{'loss': '0.3703', 'grad_norm': '10.74', 'learning_rate': '1.871e-05', 'epoch': '0.0834'}

  8%|▊         | 417/5000 [5:34:11<67:11:11, 52.78s/it][2026-05-21 21:40:10,036] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 21:40:13,827] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])

  8%|▊         | 418/5000 [5:34:59<65:09:20, 51.19s/it]
                                                       
{'loss': '0.4834', 'grad_norm': '12.27', 'learning_rate': '1.87e-05', 'epoch': '0.0836'}

  8%|▊         | 418/5000 [5:34:59<65:09:20, 51.19s/it][2026-05-21 21:41:01,289] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 165])
[2026-05-21 21:41:04,425] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 172])

  8%|▊         | 419/5000 [5:35:52<65:56:13, 51.82s/it]
                                                       
{'loss': '0.5087', 'grad_norm': '10.39', 'learning_rate': '1.87e-05', 'epoch': '0.0838'}

  8%|▊         | 419/5000 [5:35:52<65:56:13, 51.82s/it][2026-05-21 21:41:53,220] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 194])
[2026-05-21 21:41:55,871] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 109])

  8%|▊         | 420/5000 [5:36:42<65:14:29, 51.28s/it]
                                                       
{'loss': '0.4956', 'grad_norm': '9.571', 'learning_rate': '1.869e-05', 'epoch': '0.084'}

  8%|▊         | 420/5000 [5:36:42<65:14:29, 51.28s/it][2026-05-21 21:42:29,703] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 79])
[2026-05-21 21:42:30,028] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 109])

  8%|▊         | 421/5000 [5:37:31<64:18:19, 50.56s/it]
                                                       
{'loss': '0.3881', 'grad_norm': '12.73', 'learning_rate': '1.869e-05', 'epoch': '0.0842'}

  8%|▊         | 421/5000 [5:37:31<64:18:19, 50.56s/it][2026-05-21 21:43:18,238] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 171])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 81])
[2026-05-21 21:43:19,992] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 502])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 67])

  8%|▊         | 422/5000 [5:38:20<63:38:44, 50.05s/it]
                                                       
{'loss': '0.5981', 'grad_norm': '10.54', 'learning_rate': '1.869e-05', 'epoch': '0.0844'}

  8%|▊         | 422/5000 [5:38:20<63:38:44, 50.05s/it][2026-05-21 21:44:21,950] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 21:44:22,598] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  8%|▊         | 423/5000 [5:39:32<71:52:32, 56.53s/it]
                                                       
{'loss': '1.242', 'grad_norm': '12.71', 'learning_rate': '1.868e-05', 'epoch': '0.0846'}

  8%|▊         | 423/5000 [5:39:32<71:52:32, 56.53s/it][2026-05-21 21:45:32,546] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 21:45:42,722] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  8%|▊         | 424/5000 [5:40:43<77:38:47, 61.09s/it]
                                                       
{'loss': '0.8561', 'grad_norm': '8.671', 'learning_rate': '1.868e-05', 'epoch': '0.0848'}

  8%|▊         | 424/5000 [5:40:43<77:38:47, 61.09s/it][2026-05-21 21:46:42,876] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 21:46:46,676] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  8%|▊         | 425/5000 [5:41:34<73:37:19, 57.93s/it]
                                                       
{'loss': '0.256', 'grad_norm': '7.738', 'learning_rate': '1.867e-05', 'epoch': '0.085'}

  8%|▊         | 425/5000 [5:41:34<73:37:19, 57.93s/it][2026-05-21 21:47:32,759] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])
[2026-05-21 21:47:39,664] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  9%|▊         | 426/5000 [5:42:24<70:42:56, 55.66s/it]
                                                       
{'loss': '0.2406', 'grad_norm': '12.79', 'learning_rate': '1.867e-05', 'epoch': '0.0852'}

  9%|▊         | 426/5000 [5:42:24<70:42:56, 55.66s/it][2026-05-21 21:48:15,242] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 290])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])
[2026-05-21 21:48:15,373] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 344])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 980])

  9%|▊         | 427/5000 [5:42:52<60:00:13, 47.24s/it]
                                                       
{'loss': '0.2133', 'grad_norm': '10.76', 'learning_rate': '1.867e-05', 'epoch': '0.0854'}

  9%|▊         | 427/5000 [5:42:52<60:00:13, 47.24s/it][2026-05-21 21:48:41,825] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 240])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 21:48:42,528] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 297])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  9%|▊         | 428/5000 [5:43:22<53:32:47, 42.16s/it]
                                                       
{'loss': '0.1298', 'grad_norm': '8.157', 'learning_rate': '1.866e-05', 'epoch': '0.0856'}

  9%|▊         | 428/5000 [5:43:22<53:32:47, 42.16s/it][2026-05-21 21:49:12,255] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 21:49:13,612] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 962])

  9%|▊         | 429/5000 [5:44:33<64:35:33, 50.87s/it]
                                                       
{'loss': '0.8067', 'grad_norm': '13.82', 'learning_rate': '1.866e-05', 'epoch': '0.0858'}

  9%|▊         | 429/5000 [5:44:33<64:35:33, 50.87s/it][2026-05-21 21:50:23,015] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 21:50:23,942] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  9%|▊         | 430/5000 [5:45:45<72:18:46, 56.96s/it]
                                                       
{'loss': '0.8554', 'grad_norm': '10.61', 'learning_rate': '1.865e-05', 'epoch': '0.086'}

  9%|▊         | 430/5000 [5:45:45<72:18:46, 56.96s/it][2026-05-21 21:51:35,848] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 109])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 21:51:38,765] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 102])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  9%|▊         | 431/5000 [5:46:57<78:10:17, 61.59s/it]
                                                       
{'loss': '0.6035', 'grad_norm': '14.27', 'learning_rate': '1.865e-05', 'epoch': '0.0862'}

  9%|▊         | 431/5000 [5:46:57<78:10:17, 61.59s/it][2026-05-21 21:52:48,117] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 92])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 21:52:52,407] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 98])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])

  9%|▊         | 432/5000 [5:48:08<81:56:21, 64.58s/it]
                                                       
{'loss': '0.8313', 'grad_norm': '12.59', 'learning_rate': '1.864e-05', 'epoch': '0.0864'}

  9%|▊         | 432/5000 [5:48:08<81:56:21, 64.58s/it][2026-05-21 21:53:57,737] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 21:53:57,930] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  9%|▊         | 433/5000 [5:48:57<75:45:03, 59.71s/it]
                                                       
{'loss': '0.7163', 'grad_norm': '10.95', 'learning_rate': '1.864e-05', 'epoch': '0.0866'}

  9%|▊         | 433/5000 [5:48:57<75:45:03, 59.71s/it][2026-05-21 21:54:46,188] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 21:54:47,525] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  9%|▊         | 434/5000 [5:49:45<71:11:17, 56.13s/it]
                                                       
{'loss': '0.2089', 'grad_norm': '10.06', 'learning_rate': '1.864e-05', 'epoch': '0.0868'}

  9%|▊         | 434/5000 [5:49:45<71:11:17, 56.13s/it][2026-05-21 21:55:31,561] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])
[2026-05-21 21:55:32,771] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 280])

  9%|▊         | 435/5000 [5:50:35<68:56:15, 54.36s/it]
                                                       
{'loss': '0.8918', 'grad_norm': '12.51', 'learning_rate': '1.863e-05', 'epoch': '0.087'}

  9%|▊         | 435/5000 [5:50:35<68:56:15, 54.36s/it][2026-05-21 21:56:22,386] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 252])
[2026-05-21 21:56:22,884] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 310])

  9%|▊         | 436/5000 [5:51:26<67:40:26, 53.38s/it]
                                                       
{'loss': '0.1687', 'grad_norm': '8.853', 'learning_rate': '1.863e-05', 'epoch': '0.0872'}

  9%|▊         | 436/5000 [5:51:26<67:40:26, 53.38s/it][2026-05-21 21:57:15,588] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 133])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 21:57:16,855] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 141])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])

  9%|▊         | 437/5000 [5:52:14<65:41:26, 51.83s/it]
                                                       
{'loss': '1.546', 'grad_norm': '11.32', 'learning_rate': '1.862e-05', 'epoch': '0.0874'}

  9%|▊         | 437/5000 [5:52:14<65:41:26, 51.83s/it][2026-05-21 21:58:03,326] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 165])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 21:58:05,390] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 125])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])

  9%|▉         | 438/5000 [5:53:03<64:45:09, 51.10s/it]
                                                       
{'loss': '0.8918', 'grad_norm': '19.86', 'learning_rate': '1.862e-05', 'epoch': '0.0876'}

  9%|▉         | 438/5000 [5:53:03<64:45:09, 51.10s/it][2026-05-21 21:58:51,200] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 264])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 21:58:53,677] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 984])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

  9%|▉         | 439/5000 [5:53:53<64:17:45, 50.75s/it]
                                                       
{'loss': '2.142', 'grad_norm': '40.1', 'learning_rate': '1.862e-05', 'epoch': '0.0878'}

  9%|▉         | 439/5000 [5:53:53<64:17:45, 50.75s/it][2026-05-21 21:59:42,250] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-21 21:59:43,363] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 694])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

  9%|▉         | 440/5000 [5:54:43<63:45:06, 50.33s/it]
                                                       
{'loss': '2.05', 'grad_norm': '38.15', 'learning_rate': '1.861e-05', 'epoch': '0.088'}

  9%|▉         | 440/5000 [5:54:43<63:45:06, 50.33s/it][2026-05-21 22:00:31,858] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 257])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 22:00:33,575] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

  9%|▉         | 441/5000 [5:55:14<56:25:04, 44.55s/it]
                                                       
{'loss': '0.4535', 'grad_norm': '14.24', 'learning_rate': '1.861e-05', 'epoch': '0.0882'}

  9%|▉         | 441/5000 [5:55:14<56:25:04, 44.55s/it][2026-05-21 22:01:05,295] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 243])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 22:01:05,459] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

  9%|▉         | 442/5000 [5:55:45<51:14:13, 40.47s/it]
                                                       
{'loss': '0.1675', 'grad_norm': '9.968', 'learning_rate': '1.86e-05', 'epoch': '0.0884'}

  9%|▉         | 442/5000 [5:55:45<51:14:13, 40.47s/it][2026-05-21 22:01:37,670] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 987])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 226])
[2026-05-21 22:01:38,644] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])

  9%|▉         | 443/5000 [5:56:35<54:50:49, 43.33s/it]
                                                       
{'loss': '0.9672', 'grad_norm': '15.8', 'learning_rate': '1.86e-05', 'epoch': '0.0886'}

  9%|▉         | 443/5000 [5:56:35<54:50:49, 43.33s/it][2026-05-21 22:02:28,628] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1014])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 203])
[2026-05-21 22:02:29,308] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 986])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 290])

  9%|▉         | 444/5000 [5:57:25<57:26:51, 45.39s/it]
                                                       
{'loss': '1.068', 'grad_norm': '10.98', 'learning_rate': '1.86e-05', 'epoch': '0.0888'}

  9%|▉         | 444/5000 [5:57:25<57:26:51, 45.39s/it][2026-05-21 22:03:12,664] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])
[2026-05-21 22:03:12,929] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 267])

  9%|▉         | 445/5000 [5:58:15<59:12:54, 46.80s/it]
                                                       
{'loss': '0.4741', 'grad_norm': '11.84', 'learning_rate': '1.859e-05', 'epoch': '0.089'}

  9%|▉         | 445/5000 [5:58:15<59:12:54, 46.80s/it][2026-05-21 22:04:02,778] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])
[2026-05-21 22:04:03,198] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 279])

  9%|▉         | 446/5000 [5:58:54<56:03:39, 44.32s/it]
                                                       
{'loss': '0.2974', 'grad_norm': '10.55', 'learning_rate': '1.859e-05', 'epoch': '0.0892'}

  9%|▉         | 446/5000 [5:58:54<56:03:39, 44.32s/it][2026-05-21 22:04:40,729] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 417])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 72])
[2026-05-21 22:04:41,388] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])

  9%|▉         | 447/5000 [5:59:46<58:56:49, 46.61s/it]
                                                       
{'loss': '1.207', 'grad_norm': '12.47', 'learning_rate': '1.858e-05', 'epoch': '0.0894'}

  9%|▉         | 447/5000 [5:59:46<58:56:49, 46.61s/it][2026-05-21 22:05:33,838] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 462])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 90])
[2026-05-21 22:05:33,874] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 58])

  9%|▉         | 448/5000 [6:00:47<64:43:51, 51.19s/it]
                                                       
{'loss': '0.9328', 'grad_norm': '13.96', 'learning_rate': '1.858e-05', 'epoch': '0.0896'}

  9%|▉         | 448/5000 [6:00:47<64:43:51, 51.19s/it][2026-05-21 22:06:34,633] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 416])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 62])
[2026-05-21 22:06:35,224] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 424])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 69])

  9%|▉         | 449/5000 [6:01:35<63:26:26, 50.18s/it]
                                                       
{'loss': '0.2773', 'grad_norm': '18.27', 'learning_rate': '1.858e-05', 'epoch': '0.0898'}

  9%|▉         | 449/5000 [6:01:35<63:26:26, 50.18s/it][2026-05-21 22:07:22,474] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 64])
[2026-05-21 22:07:23,995] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 519])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 82])

  9%|▉         | 450/5000 [6:02:24<62:47:49, 49.69s/it]
                                                       
{'loss': '0.2517', 'grad_norm': '15.66', 'learning_rate': '1.857e-05', 'epoch': '0.09'}

  9%|▉         | 450/5000 [6:02:24<62:47:49, 49.69s/it][2026-05-21 22:08:11,944] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 211])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 22:08:13,096] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 426])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

  9%|▉         | 451/5000 [6:02:54<55:33:35, 43.97s/it]
                                                       
{'loss': '0.618', 'grad_norm': '16.58', 'learning_rate': '1.857e-05', 'epoch': '0.0902'}

  9%|▉         | 451/5000 [6:02:54<55:33:35, 43.97s/it][2026-05-21 22:08:43,628] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 22:08:44,340] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 486])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  9%|▉         | 452/5000 [6:03:25<50:23:29, 39.89s/it]
                                                       
{'loss': '0.4909', 'grad_norm': '19.03', 'learning_rate': '1.856e-05', 'epoch': '0.0904'}

  9%|▉         | 452/5000 [6:03:25<50:23:29, 39.89s/it][2026-05-21 22:09:14,599] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 311])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 22:09:14,755] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

  9%|▉         | 453/5000 [6:03:52<45:23:23, 35.94s/it]
                                                       
{'loss': '0.3317', 'grad_norm': '12.31', 'learning_rate': '1.856e-05', 'epoch': '0.0906'}

  9%|▉         | 453/5000 [6:03:52<45:23:23, 35.94s/it][2026-05-21 22:09:41,828] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 466])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 22:09:45,532] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

  9%|▉         | 454/5000 [6:04:37<49:01:08, 38.82s/it]
                                                       
{'loss': '0.3581', 'grad_norm': '11.59', 'learning_rate': '1.856e-05', 'epoch': '0.0908'}

  9%|▉         | 454/5000 [6:04:37<49:01:08, 38.82s/it][2026-05-21 22:10:24,460] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 230])
[2026-05-21 22:10:25,609] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 511])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 307])

  9%|▉         | 455/5000 [6:05:03<44:09:52, 34.98s/it]
                                                       
{'loss': '0.3682', 'grad_norm': '10.66', 'learning_rate': '1.855e-05', 'epoch': '0.091'}

  9%|▉         | 455/5000 [6:05:03<44:09:52, 34.98s/it][2026-05-21 22:10:51,510] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 297])
[2026-05-21 22:10:51,594] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 302])

  9%|▉         | 456/5000 [6:05:39<44:36:13, 35.34s/it]
                                                       
{'loss': '0.3037', 'grad_norm': '10.37', 'learning_rate': '1.855e-05', 'epoch': '0.0912'}

  9%|▉         | 456/5000 [6:05:39<44:36:13, 35.34s/it][2026-05-21 22:11:26,080] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-21 22:11:26,592] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 231])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

  9%|▉         | 457/5000 [6:06:26<48:53:39, 38.75s/it]
                                                       
{'loss': '1.052', 'grad_norm': '12.69', 'learning_rate': '1.854e-05', 'epoch': '0.0914'}

  9%|▉         | 457/5000 [6:06:26<48:53:39, 38.75s/it][2026-05-21 22:12:12,705] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 168])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-21 22:12:13,841] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 480])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

  9%|▉         | 458/5000 [6:07:17<53:22:25, 42.30s/it]
                                                       
{'loss': '0.3166', 'grad_norm': '8.622', 'learning_rate': '1.854e-05', 'epoch': '0.0916'}

  9%|▉         | 458/5000 [6:07:17<53:22:25, 42.30s/it][2026-05-21 22:13:12,135] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 894])
[2026-05-21 22:13:16,812] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 987])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  9%|▉         | 459/5000 [6:08:09<57:06:47, 45.28s/it]
                                                       
{'loss': '0.9035', 'grad_norm': '13.18', 'learning_rate': '1.853e-05', 'epoch': '0.0918'}

  9%|▉         | 459/5000 [6:08:09<57:06:47, 45.28s/it][2026-05-21 22:14:07,176] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 22:14:08,951] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])

  9%|▉         | 460/5000 [6:09:04<61:00:24, 48.38s/it]
                                                       
{'loss': '0.5635', 'grad_norm': '8.487', 'learning_rate': '1.853e-05', 'epoch': '0.092'}

  9%|▉         | 460/5000 [6:09:04<61:00:24, 48.38s/it][2026-05-21 22:14:56,389] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 981])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 22:14:56,411] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])

  9%|▉         | 461/5000 [6:09:36<54:45:54, 43.44s/it]
                                                       
{'loss': '0.4246', 'grad_norm': '10.7', 'learning_rate': '1.853e-05', 'epoch': '0.0922'}

  9%|▉         | 461/5000 [6:09:36<54:45:54, 43.44s/it][2026-05-21 22:15:28,449] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 987])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-21 22:15:30,348] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 972])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  9%|▉         | 462/5000 [6:10:12<51:50:38, 41.13s/it]
                                                       
{'loss': '0.06', 'grad_norm': '5.564', 'learning_rate': '1.852e-05', 'epoch': '0.0924'}

  9%|▉         | 462/5000 [6:10:12<51:50:38, 41.13s/it][2026-05-21 22:16:00,945] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 176])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 22:16:02,181] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 249])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  9%|▉         | 463/5000 [6:10:37<45:51:25, 36.39s/it]
                                                       
{'loss': '0.1941', 'grad_norm': '5.579', 'learning_rate': '1.852e-05', 'epoch': '0.0926'}

  9%|▉         | 463/5000 [6:10:37<45:51:25, 36.39s/it][2026-05-21 22:16:26,332] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 180])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 22:16:26,620] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 176])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  9%|▉         | 464/5000 [6:11:08<43:30:54, 34.54s/it]
                                                       
{'loss': '0.1788', 'grad_norm': '5.318', 'learning_rate': '1.851e-05', 'epoch': '0.0928'}

  9%|▉         | 464/5000 [6:11:08<43:30:54, 34.54s/it][2026-05-21 22:16:58,747] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 100])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 22:16:59,875] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 100])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  9%|▉         | 465/5000 [6:12:22<58:29:59, 46.44s/it]
                                                       
{'loss': '0.6869', 'grad_norm': '15.31', 'learning_rate': '1.851e-05', 'epoch': '0.093'}

  9%|▉         | 465/5000 [6:12:22<58:29:59, 46.44s/it][2026-05-21 22:18:12,169] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 114])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 22:18:14,005] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 94])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

  9%|▉         | 466/5000 [6:13:33<67:42:56, 53.77s/it]
                                                       
{'loss': '0.8494', 'grad_norm': '11.06', 'learning_rate': '1.851e-05', 'epoch': '0.0932'}

  9%|▉         | 466/5000 [6:13:33<67:42:56, 53.77s/it][2026-05-21 22:19:34,976] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 22:19:38,699] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 993])

  9%|▉         | 467/5000 [6:14:43<74:02:09, 58.80s/it]
                                                       
{'loss': '0.7862', 'grad_norm': '16.34', 'learning_rate': '1.85e-05', 'epoch': '0.0934'}

  9%|▉         | 467/5000 [6:14:43<74:02:09, 58.80s/it][2026-05-21 22:20:43,602] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 22:20:46,667] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 958])

  9%|▉         | 468/5000 [6:15:54<78:24:31, 62.28s/it]
                                                       
{'loss': '0.9438', 'grad_norm': '10.2', 'learning_rate': '1.85e-05', 'epoch': '0.0936'}

  9%|▉         | 468/5000 [6:15:54<78:24:31, 62.28s/it][2026-05-21 22:21:43,824] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 510])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 22:21:44,662] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  9%|▉         | 469/5000 [6:16:41<72:37:35, 57.70s/it]
                                                       
{'loss': '0.4503', 'grad_norm': '12.14', 'learning_rate': '1.849e-05', 'epoch': '0.0938'}

  9%|▉         | 469/5000 [6:16:41<72:37:35, 57.70s/it][2026-05-21 22:22:31,075] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 22:22:31,466] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 430])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  9%|▉         | 470/5000 [6:17:27<68:29:11, 54.43s/it]
                                                       
{'loss': '0.6347', 'grad_norm': '10.99', 'learning_rate': '1.849e-05', 'epoch': '0.094'}

  9%|▉         | 470/5000 [6:17:27<68:29:11, 54.43s/it][2026-05-21 22:23:15,528] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-21 22:23:16,822] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  9%|▉         | 471/5000 [6:18:16<66:06:31, 52.55s/it]
                                                       
{'loss': '1.112', 'grad_norm': '12.84', 'learning_rate': '1.849e-05', 'epoch': '0.0942'}

  9%|▉         | 471/5000 [6:18:16<66:06:31, 52.55s/it][2026-05-21 22:24:04,319] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 430])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 22:24:04,784] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

  9%|▉         | 472/5000 [6:19:03<64:19:13, 51.14s/it]
                                                       
{'loss': '1.146', 'grad_norm': '14.83', 'learning_rate': '1.848e-05', 'epoch': '0.0944'}

  9%|▉         | 472/5000 [6:19:03<64:19:13, 51.14s/it][2026-05-21 22:24:55,063] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 124])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 22:24:56,581] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 87])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

  9%|▉         | 473/5000 [6:20:16<72:18:21, 57.50s/it]
                                                       
{'loss': '0.534', 'grad_norm': '11.72', 'learning_rate': '1.848e-05', 'epoch': '0.0946'}

  9%|▉         | 473/5000 [6:20:16<72:18:21, 57.50s/it][2026-05-21 22:26:06,068] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 93])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 22:26:08,250] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 101])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

  9%|▉         | 474/5000 [6:21:25<76:39:02, 60.97s/it]
                                                       
{'loss': '0.5523', 'grad_norm': '9.717', 'learning_rate': '1.847e-05', 'epoch': '0.0948'}

  9%|▉         | 474/5000 [6:21:25<76:39:02, 60.97s/it][2026-05-21 22:27:12,808] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 960])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 81])
[2026-05-21 22:27:13,400] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 530])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 63])

 10%|▉         | 475/5000 [6:22:15<72:33:55, 57.73s/it]
                                                       
{'loss': '0.6342', 'grad_norm': '11.5', 'learning_rate': '1.847e-05', 'epoch': '0.095'}

 10%|▉         | 475/5000 [6:22:15<72:33:55, 57.73s/it][2026-05-21 22:28:03,998] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 927])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 65])
[2026-05-21 22:28:04,852] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])

 10%|▉         | 476/5000 [6:23:06<69:53:24, 55.62s/it]
                                                       
{'loss': '0.1706', 'grad_norm': '7.197', 'learning_rate': '1.847e-05', 'epoch': '0.0952'}

 10%|▉         | 476/5000 [6:23:06<69:53:24, 55.62s/it][2026-05-21 22:28:52,685] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 203])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 11])
[2026-05-21 22:28:54,211] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 9])

 10%|▉         | 477/5000 [6:23:56<67:51:07, 54.01s/it]
                                                       
{'loss': '0.7495', 'grad_norm': '11.88', 'learning_rate': '1.846e-05', 'epoch': '0.0954'}

 10%|▉         | 477/5000 [6:23:56<67:51:07, 54.01s/it][2026-05-21 22:29:44,559] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 8])
[2026-05-21 22:29:44,565] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 9])

 10%|▉         | 478/5000 [6:24:48<66:59:19, 53.33s/it]
                                                       
{'loss': '0.2241', 'grad_norm': '7.024', 'learning_rate': '1.846e-05', 'epoch': '0.0956'}

 10%|▉         | 478/5000 [6:24:48<66:59:19, 53.33s/it][2026-05-21 22:30:39,956] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 981])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-21 22:30:40,546] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 10%|▉         | 479/5000 [6:25:39<66:04:13, 52.61s/it]
                                                       
{'loss': '1.149', 'grad_norm': '13.53', 'learning_rate': '1.845e-05', 'epoch': '0.0958'}

 10%|▉         | 479/5000 [6:25:39<66:04:13, 52.61s/it][2026-05-21 22:31:30,776] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-21 22:31:32,735] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 10%|▉         | 480/5000 [6:26:31<66:03:34, 52.61s/it]
                                                       
{'loss': '1.295', 'grad_norm': '14.66', 'learning_rate': '1.845e-05', 'epoch': '0.096'}

 10%|▉         | 480/5000 [6:26:31<66:03:34, 52.61s/it][2026-05-21 22:32:29,340] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])
[2026-05-21 22:32:29,511] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 10%|▉         | 481/5000 [6:27:14<62:26:34, 49.74s/it]
                                                       
{'loss': '0.4175', 'grad_norm': '10.79', 'learning_rate': '1.844e-05', 'epoch': '0.0962'}

 10%|▉         | 481/5000 [6:27:14<62:26:34, 49.74s/it][2026-05-21 22:33:11,591] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 22:33:12,907] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 10%|▉         | 482/5000 [6:28:01<61:20:01, 48.87s/it]
                                                       
{'loss': '0.1588', 'grad_norm': '6.942', 'learning_rate': '1.844e-05', 'epoch': '0.0964'}

 10%|▉         | 482/5000 [6:28:01<61:20:01, 48.87s/it][2026-05-21 22:33:58,440] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 22:33:59,772] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 10%|▉         | 483/5000 [6:28:46<59:49:24, 47.68s/it]
                                                       
{'loss': '1.405', 'grad_norm': '19.15', 'learning_rate': '1.844e-05', 'epoch': '0.0966'}

 10%|▉         | 483/5000 [6:28:46<59:49:24, 47.68s/it][2026-05-21 22:34:45,081] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 22:34:45,760] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 473])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 10%|▉         | 484/5000 [6:29:31<58:54:51, 46.96s/it]
                                                       
{'loss': '0.5855', 'grad_norm': '11.85', 'learning_rate': '1.843e-05', 'epoch': '0.0968'}

 10%|▉         | 484/5000 [6:29:31<58:54:51, 46.96s/it][2026-05-21 22:35:23,822] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 22:35:25,359] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 10%|▉         | 485/5000 [6:30:43<68:15:57, 54.43s/it]
                                                       
{'loss': '0.3992', 'grad_norm': '15.21', 'learning_rate': '1.843e-05', 'epoch': '0.097'}

 10%|▉         | 485/5000 [6:30:43<68:15:57, 54.43s/it][2026-05-21 22:36:36,530] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])
[2026-05-21 22:36:36,725] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 10%|▉         | 486/5000 [6:31:55<74:37:58, 59.52s/it]
                                                       
{'loss': '0.363', 'grad_norm': '11.38', 'learning_rate': '1.842e-05', 'epoch': '0.0972'}

 10%|▉         | 486/5000 [6:31:55<74:37:58, 59.52s/it][2026-05-21 22:37:49,815] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 287])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 215])
[2026-05-21 22:37:51,983] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 412])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 10%|▉         | 487/5000 [6:33:04<78:27:16, 62.58s/it]
                                                       
{'loss': '0.5041', 'grad_norm': '8.519', 'learning_rate': '1.842e-05', 'epoch': '0.0974'}

 10%|▉         | 487/5000 [6:33:04<78:27:16, 62.58s/it][2026-05-21 22:39:01,703] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 415])
[2026-05-21 22:39:03,338] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 10%|▉         | 488/5000 [6:34:12<80:21:27, 64.12s/it]
                                                       
{'loss': '0.8636', 'grad_norm': '10.49', 'learning_rate': '1.842e-05', 'epoch': '0.0976'}

 10%|▉         | 488/5000 [6:34:12<80:21:27, 64.12s/it][2026-05-21 22:40:00,041] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])
[2026-05-21 22:40:00,241] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 257])

 10%|▉         | 489/5000 [6:35:02<75:04:38, 59.92s/it]
                                                       
{'loss': '0.6951', 'grad_norm': '11.27', 'learning_rate': '1.841e-05', 'epoch': '0.0978'}

 10%|▉         | 489/5000 [6:35:02<75:04:38, 59.92s/it][2026-05-21 22:40:50,393] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 308])
[2026-05-21 22:40:50,498] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 260])

 10%|▉         | 490/5000 [6:35:51<70:47:39, 56.51s/it]
                                                       
{'loss': '0.4449', 'grad_norm': '6.71', 'learning_rate': '1.841e-05', 'epoch': '0.098'}

 10%|▉         | 490/5000 [6:35:51<70:47:39, 56.51s/it][2026-05-21 22:41:51,895] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 108])
[2026-05-21 22:41:54,316] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])

 10%|▉         | 491/5000 [6:36:59<75:09:27, 60.01s/it]
                                                       
{'loss': '0.8156', 'grad_norm': '11.67', 'learning_rate': '1.84e-05', 'epoch': '0.0982'}

 10%|▉         | 491/5000 [6:36:59<75:09:27, 60.01s/it][2026-05-21 22:43:03,371] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1028])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 114])
[2026-05-21 22:43:03,873] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 174])

 10%|▉         | 492/5000 [6:38:12<80:08:50, 64.00s/it]
                                                       
{'loss': '0.5245', 'grad_norm': '7.989', 'learning_rate': '1.84e-05', 'epoch': '0.0984'}

 10%|▉         | 492/5000 [6:38:12<80:08:50, 64.00s/it]
 10%|▉         | 493/5000 [6:38:44<67:55:04, 54.25s/it]
                                                       
{'loss': '0.4223', 'grad_norm': '10.86', 'learning_rate': '1.84e-05', 'epoch': '0.0986'}

 10%|▉         | 493/5000 [6:38:44<67:55:04, 54.25s/it][2026-05-21 22:44:28,975] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 22:44:29,159] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 22:44:59,190] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 22:45:01,306] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 10%|▉         | 494/5000 [6:39:16<59:46:59, 47.76s/it]
                                                       
{'loss': '0.07409', 'grad_norm': '7.797', 'learning_rate': '1.839e-05', 'epoch': '0.0988'}

 10%|▉         | 494/5000 [6:39:16<59:46:59, 47.76s/it][2026-05-21 22:45:03,776] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 22:45:03,895] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 269])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 10%|▉         | 495/5000 [6:39:47<53:14:01, 42.54s/it]
                                                       
{'loss': '2.273', 'grad_norm': '39.92', 'learning_rate': '1.839e-05', 'epoch': '0.099'}

 10%|▉         | 495/5000 [6:39:47<53:14:01, 42.54s/it][2026-05-21 22:45:33,997] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 22:45:34,632] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 426])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 10%|▉         | 496/5000 [6:40:16<48:11:14, 38.52s/it]
                                                       
{'loss': '1.625', 'grad_norm': '33.89', 'learning_rate': '1.838e-05', 'epoch': '0.0992'}

 10%|▉         | 496/5000 [6:40:16<48:11:14, 38.52s/it][2026-05-21 22:46:08,896] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 129])
[2026-05-21 22:46:09,753] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 145])

 10%|▉         | 497/5000 [6:41:09<53:46:27, 42.99s/it]
                                                       
{'loss': '0.4811', 'grad_norm': '13.7', 'learning_rate': '1.838e-05', 'epoch': '0.0994'}

 10%|▉         | 497/5000 [6:41:09<53:46:27, 42.99s/it][2026-05-21 22:47:01,856] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 84])
[2026-05-21 22:47:03,475] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 107])

 10%|▉         | 498/5000 [6:42:03<57:46:05, 46.19s/it]
                                                       
{'loss': '0.4547', 'grad_norm': '14.77', 'learning_rate': '1.838e-05', 'epoch': '0.0996'}

 10%|▉         | 498/5000 [6:42:03<57:46:05, 46.19s/it][2026-05-21 22:48:09,015] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 251])
[2026-05-21 22:48:10,543] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 244])

 10%|▉         | 499/5000 [6:42:53<59:20:02, 47.46s/it]
                                                       
{'loss': '0.2738', 'grad_norm': '10.27', 'learning_rate': '1.837e-05', 'epoch': '0.0998'}

 10%|▉         | 499/5000 [6:42:53<59:20:02, 47.46s/it][2026-05-21 22:48:58,132] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 275])
[2026-05-21 22:49:01,006] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 288])

 10%|█         | 500/5000 [6:43:44<60:34:20, 48.46s/it]
                                                       
{'loss': '0.3003', 'grad_norm': '12.92', 'learning_rate': '1.837e-05', 'epoch': '0.1'}

 10%|█         | 500/5000 [6:43:44<60:34:20, 48.46s/it][2026-05-21 22:49:29,210] INFO [src.utils.basic_utils:19] Saving model to /workspace/VLM2Vec/outputs/Qwen3.5-08b-continued-mixed-alltasks/checkpoint-500
[2026-05-21 22:49:30,327] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-21 22:49:30,339] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-21 22:49:33,943] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 22:49:34,502] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 407])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 10%|█         | 501/5000 [6:44:17<54:53:12, 43.92s/it]
                                                       
{'loss': '0.2291', 'grad_norm': '10.55', 'learning_rate': '1.836e-05', 'epoch': '0.1002'}

 10%|█         | 501/5000 [6:44:17<54:53:12, 43.92s/it][2026-05-21 22:50:08,308] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 447])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 22:50:09,068] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

 10%|█         | 502/5000 [6:44:52<51:23:46, 41.14s/it]
                                                       
{'loss': '0.241', 'grad_norm': '10.61', 'learning_rate': '1.836e-05', 'epoch': '0.1004'}

 10%|█         | 502/5000 [6:44:52<51:23:46, 41.14s/it][2026-05-21 22:50:40,447] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 219])
[2026-05-21 22:50:40,754] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 417])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 315])

 10%|█         | 503/5000 [6:45:39<53:42:14, 42.99s/it]
                                                       
{'loss': '0.244', 'grad_norm': '8.937', 'learning_rate': '1.836e-05', 'epoch': '0.1006'}

 10%|█         | 503/5000 [6:45:39<53:42:14, 42.99s/it][2026-05-21 22:51:28,372] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 343])
[2026-05-21 22:51:28,932] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 242])

 10%|█         | 504/5000 [6:46:27<55:21:47, 44.33s/it]
                                                       
{'loss': '0.2869', 'grad_norm': '8.052', 'learning_rate': '1.835e-05', 'epoch': '0.1008'}

 10%|█         | 504/5000 [6:46:27<55:21:47, 44.33s/it][2026-05-21 22:52:18,631] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 321])
[2026-05-21 22:52:19,389] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])

 10%|█         | 505/5000 [6:47:18<57:54:59, 46.38s/it]
                                                       
{'loss': '0.3659', 'grad_norm': '14.8', 'learning_rate': '1.835e-05', 'epoch': '0.101'}

 10%|█         | 505/5000 [6:47:18<57:54:59, 46.38s/it][2026-05-21 22:53:10,074] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 254])
[2026-05-21 22:53:10,513] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 228])

 10%|█         | 506/5000 [6:48:09<59:33:14, 47.71s/it]
                                                       
{'loss': '0.1664', 'grad_norm': '9.041', 'learning_rate': '1.834e-05', 'epoch': '0.1012'}

 10%|█         | 506/5000 [6:48:09<59:33:14, 47.71s/it][2026-05-21 22:54:14,147] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 212])
[2026-05-21 22:54:18,752] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 298])

 10%|█         | 507/5000 [6:49:00<60:59:06, 48.86s/it]
                                                       
{'loss': '0.6271', 'grad_norm': '9.572', 'learning_rate': '1.834e-05', 'epoch': '0.1014'}

 10%|█         | 507/5000 [6:49:00<60:59:06, 48.86s/it][2026-05-21 22:55:05,469] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 206])
[2026-05-21 22:55:09,115] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 109])

 10%|█         | 508/5000 [6:49:52<62:11:52, 49.85s/it]
                                                       
{'loss': '0.8489', 'grad_norm': '12.33', 'learning_rate': '1.833e-05', 'epoch': '0.1016'}

 10%|█         | 508/5000 [6:49:52<62:11:52, 49.85s/it][2026-05-21 22:55:40,508] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 238])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 274])
[2026-05-21 22:55:40,872] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 416])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 286])

 10%|█         | 509/5000 [6:50:20<53:57:53, 43.26s/it]
                                                       
{'loss': '0.2484', 'grad_norm': '9.421', 'learning_rate': '1.833e-05', 'epoch': '0.1018'}

 10%|█         | 509/5000 [6:50:20<53:57:53, 43.26s/it][2026-05-21 22:56:08,838] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])
[2026-05-21 22:56:09,362] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 502])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 245])

 10%|█         | 510/5000 [6:50:51<49:19:03, 39.54s/it]
                                                       
{'loss': '0.1212', 'grad_norm': '7.826', 'learning_rate': '1.833e-05', 'epoch': '0.102'}

 10%|█         | 510/5000 [6:50:51<49:19:03, 39.54s/it][2026-05-21 22:56:38,373] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 539])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 86])
[2026-05-21 22:56:38,862] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 923])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 60])

 10%|█         | 511/5000 [6:51:40<52:47:04, 42.33s/it]
                                                       
{'loss': '0.5488', 'grad_norm': '9.709', 'learning_rate': '1.832e-05', 'epoch': '0.1022'}

 10%|█         | 511/5000 [6:51:40<52:47:04, 42.33s/it][2026-05-21 22:57:27,564] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 52])
[2026-05-21 22:57:28,819] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 97])

 10%|█         | 512/5000 [6:52:28<54:54:23, 44.04s/it]
                                                       
{'loss': '0.4018', 'grad_norm': '9.48', 'learning_rate': '1.832e-05', 'epoch': '0.1024'}

 10%|█         | 512/5000 [6:52:28<54:54:23, 44.04s/it][2026-05-21 22:58:15,707] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 544])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 22:58:16,636] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 10%|█         | 513/5000 [6:53:16<56:24:22, 45.26s/it]
                                                       
{'loss': '0.3485', 'grad_norm': '8.605', 'learning_rate': '1.831e-05', 'epoch': '0.1026'}

 10%|█         | 513/5000 [6:53:16<56:24:22, 45.26s/it][2026-05-21 22:59:04,773] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 968])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 22:59:04,860] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 10%|█         | 514/5000 [6:54:04<57:26:14, 46.09s/it]
                                                       
{'loss': '0.2806', 'grad_norm': '12.49', 'learning_rate': '1.831e-05', 'epoch': '0.1028'}

 10%|█         | 514/5000 [6:54:04<57:26:14, 46.09s/it][2026-05-21 22:59:55,704] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 601])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 23:00:02,327] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 547])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 10%|█         | 515/5000 [6:54:35<51:41:30, 41.49s/it]
                                                       
{'loss': '0.3823', 'grad_norm': '9.807', 'learning_rate': '1.831e-05', 'epoch': '0.103'}

 10%|█         | 515/5000 [6:54:35<51:41:30, 41.49s/it][2026-05-21 23:00:25,891] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 967])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 23:00:26,016] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 802])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 10%|█         | 516/5000 [6:55:06<47:35:17, 38.21s/it]
                                                       
{'loss': '0.05692', 'grad_norm': '4.705', 'learning_rate': '1.83e-05', 'epoch': '0.1032'}

 10%|█         | 516/5000 [6:55:06<47:35:17, 38.21s/it][2026-05-21 23:00:55,265] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 544])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 23:00:55,330] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 10%|█         | 517/5000 [6:55:49<49:35:14, 39.82s/it]
                                                       
{'loss': '0.4564', 'grad_norm': '20.42', 'learning_rate': '1.83e-05', 'epoch': '0.1034'}

 10%|█         | 517/5000 [6:55:49<49:35:14, 39.82s/it][2026-05-21 23:01:40,065] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-21 23:01:40,955] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 10%|█         | 518/5000 [6:56:37<52:32:48, 42.21s/it]
                                                       
{'loss': '0.1862', 'grad_norm': '8.044', 'learning_rate': '1.829e-05', 'epoch': '0.1036'}

 10%|█         | 518/5000 [6:56:37<52:32:48, 42.21s/it][2026-05-21 23:02:27,731] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 23:02:28,576] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 965])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 10%|█         | 519/5000 [6:57:27<55:38:52, 44.71s/it]
                                                       
{'loss': '0.5271', 'grad_norm': '11.38', 'learning_rate': '1.829e-05', 'epoch': '0.1038'}

 10%|█         | 519/5000 [6:57:27<55:38:52, 44.71s/it][2026-05-21 23:03:19,276] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 23:03:19,493] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 958])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 10%|█         | 520/5000 [6:58:21<59:01:45, 47.43s/it]
                                                       
{'loss': '0.1517', 'grad_norm': '7.292', 'learning_rate': '1.829e-05', 'epoch': '0.104'}

 10%|█         | 520/5000 [6:58:21<59:01:45, 47.43s/it][2026-05-21 23:04:13,028] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1015])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 124])
[2026-05-21 23:04:13,064] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 147])

 10%|█         | 521/5000 [6:59:18<62:23:42, 50.15s/it]
                                                       
{'loss': '0.639', 'grad_norm': '12.86', 'learning_rate': '1.828e-05', 'epoch': '0.1042'}

 10%|█         | 521/5000 [6:59:18<62:23:42, 50.15s/it][2026-05-21 23:05:09,384] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 157])
[2026-05-21 23:05:09,581] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 274])

 10%|█         | 522/5000 [7:00:27<69:25:59, 55.82s/it]
                                                       
{'loss': '0.7235', 'grad_norm': '9.985', 'learning_rate': '1.828e-05', 'epoch': '0.1044'}

 10%|█         | 522/5000 [7:00:27<69:25:59, 55.82s/it][2026-05-21 23:06:22,879] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 23:06:24,883] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 10%|█         | 523/5000 [7:01:35<73:58:22, 59.48s/it]
                                                       
{'loss': '0.7862', 'grad_norm': '12.09', 'learning_rate': '1.827e-05', 'epoch': '0.1046'}

 10%|█         | 523/5000 [7:01:35<73:58:22, 59.48s/it][2026-05-21 23:07:29,744] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-21 23:07:31,589] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 10%|█         | 524/5000 [7:02:44<77:41:36, 62.49s/it]
                                                       
{'loss': '0.5221', 'grad_norm': '15.69', 'learning_rate': '1.827e-05', 'epoch': '0.1048'}

 10%|█         | 524/5000 [7:02:44<77:41:36, 62.49s/it][2026-05-21 23:08:45,868] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1020])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 197])
[2026-05-21 23:08:46,773] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 180])

 10%|█         | 525/5000 [7:03:55<80:35:23, 64.83s/it]
                                                       
{'loss': '0.6339', 'grad_norm': '12.43', 'learning_rate': '1.827e-05', 'epoch': '0.105'}

 10%|█         | 525/5000 [7:03:55<80:35:23, 64.83s/it][2026-05-21 23:09:58,617] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1015])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 208])
[2026-05-21 23:09:59,363] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 11%|█         | 526/5000 [7:05:05<82:34:58, 66.45s/it]
                                                       
{'loss': '0.4827', 'grad_norm': '13.24', 'learning_rate': '1.826e-05', 'epoch': '0.1052'}

 11%|█         | 526/5000 [7:05:05<82:34:58, 66.45s/it][2026-05-21 23:11:26,153] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 216])
[2026-05-21 23:11:27,311] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 228])

 11%|█         | 527/5000 [7:05:57<77:12:05, 62.13s/it]
                                                       
{'loss': '0.4034', 'grad_norm': '10.19', 'learning_rate': '1.826e-05', 'epoch': '0.1054'}

 11%|█         | 527/5000 [7:05:57<77:12:05, 62.13s/it][2026-05-21 23:12:21,113] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 125])
[2026-05-21 23:12:26,476] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 222])

 11%|█         | 528/5000 [7:06:54<75:24:08, 60.70s/it]
                                                       
{'loss': '0.3778', 'grad_norm': '9.786', 'learning_rate': '1.825e-05', 'epoch': '0.1056'}

 11%|█         | 528/5000 [7:06:54<75:24:08, 60.70s/it][2026-05-21 23:12:51,404] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 86])
[2026-05-21 23:12:51,799] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])

 11%|█         | 529/5000 [7:07:46<71:56:27, 57.93s/it]
                                                       
{'loss': '0.3252', 'grad_norm': '10.34', 'learning_rate': '1.825e-05', 'epoch': '0.1058'}

 11%|█         | 529/5000 [7:07:46<71:56:27, 57.93s/it][2026-05-21 23:13:44,859] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 986])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 174])
[2026-05-21 23:13:45,287] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 69])

 11%|█         | 530/5000 [7:08:37<69:36:17, 56.06s/it]
                                                       
{'loss': '0.1371', 'grad_norm': '5.001', 'learning_rate': '1.824e-05', 'epoch': '0.106'}

 11%|█         | 530/5000 [7:08:37<69:36:17, 56.06s/it][2026-05-21 23:14:25,924] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 294])
[2026-05-21 23:14:26,747] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 267])

 11%|█         | 531/5000 [7:09:28<67:44:18, 54.57s/it]
                                                       
{'loss': '0.3658', 'grad_norm': '10.35', 'learning_rate': '1.824e-05', 'epoch': '0.1062'}

 11%|█         | 531/5000 [7:09:28<67:44:18, 54.57s/it][2026-05-21 23:15:17,844] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 342])
[2026-05-21 23:15:18,254] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 265])

 11%|█         | 532/5000 [7:10:20<66:32:41, 53.62s/it]
                                                       
{'loss': '0.4914', 'grad_norm': '12.07', 'learning_rate': '1.824e-05', 'epoch': '0.1064'}

 11%|█         | 532/5000 [7:10:20<66:32:41, 53.62s/it][2026-05-21 23:16:09,349] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])
[2026-05-21 23:16:11,345] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 60])

 11%|█         | 533/5000 [7:11:12<65:55:52, 53.13s/it]
                                                       
{'loss': '0.3997', 'grad_norm': '12.32', 'learning_rate': '1.823e-05', 'epoch': '0.1066'}

 11%|█         | 533/5000 [7:11:12<65:55:52, 53.13s/it][2026-05-21 23:17:00,900] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 61])
[2026-05-21 23:17:01,866] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 87])

 11%|█         | 534/5000 [7:12:02<64:50:33, 52.27s/it]
                                                       
{'loss': '0.3619', 'grad_norm': '9.137', 'learning_rate': '1.823e-05', 'epoch': '0.1068'}

 11%|█         | 534/5000 [7:12:02<64:50:33, 52.27s/it][2026-05-21 23:17:49,602] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 23:17:50,053] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 338])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 11%|█         | 535/5000 [7:12:49<62:45:25, 50.60s/it]
                                                       
{'loss': '0.2133', 'grad_norm': '7.119', 'learning_rate': '1.822e-05', 'epoch': '0.107'}

 11%|█         | 535/5000 [7:12:49<62:45:25, 50.60s/it][2026-05-21 23:18:36,679] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 23:18:36,759] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 11%|█         | 536/5000 [7:13:35<61:10:42, 49.34s/it]
                                                       
{'loss': '0.1208', 'grad_norm': '6.142', 'learning_rate': '1.822e-05', 'epoch': '0.1072'}

 11%|█         | 536/5000 [7:13:35<61:10:42, 49.34s/it][2026-05-21 23:19:23,105] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 23:19:23,690] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 11%|█         | 537/5000 [7:14:22<60:19:08, 48.66s/it]
                                                       
{'loss': '0.4446', 'grad_norm': '12.22', 'learning_rate': '1.822e-05', 'epoch': '0.1074'}

 11%|█         | 537/5000 [7:14:22<60:19:08, 48.66s/it][2026-05-21 23:20:10,678] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 23:20:10,724] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 11%|█         | 538/5000 [7:15:10<60:07:12, 48.51s/it]
                                                       
{'loss': '0.392', 'grad_norm': '14.01', 'learning_rate': '1.821e-05', 'epoch': '0.1076'}

 11%|█         | 538/5000 [7:15:10<60:07:12, 48.51s/it][2026-05-21 23:21:00,582] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 60])
[2026-05-21 23:21:01,963] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 55])

 11%|█         | 539/5000 [7:15:59<60:01:41, 48.44s/it]
                                                       
{'loss': '0.3564', 'grad_norm': '13.03', 'learning_rate': '1.821e-05', 'epoch': '0.1078'}

 11%|█         | 539/5000 [7:15:59<60:01:41, 48.44s/it][2026-05-21 23:21:48,941] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 65])
[2026-05-21 23:21:49,141] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 59])

 11%|█         | 540/5000 [7:16:47<59:50:00, 48.30s/it]
                                                       
{'loss': '0.3448', 'grad_norm': '18.71', 'learning_rate': '1.82e-05', 'epoch': '0.108'}

 11%|█         | 540/5000 [7:16:47<59:50:00, 48.30s/it][2026-05-21 23:22:36,287] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 23:22:37,073] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 424])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])

 11%|█         | 541/5000 [7:17:18<53:31:25, 43.21s/it]
                                                       
{'loss': '0.9119', 'grad_norm': '20.33', 'learning_rate': '1.82e-05', 'epoch': '0.1082'}

 11%|█         | 541/5000 [7:17:18<53:31:25, 43.21s/it][2026-05-21 23:23:07,588] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 180])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 23:23:08,561] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 428])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])

 11%|█         | 542/5000 [7:17:55<51:18:35, 41.43s/it]
                                                       
{'loss': '0.726', 'grad_norm': '19.11', 'learning_rate': '1.82e-05', 'epoch': '0.1084'}

 11%|█         | 542/5000 [7:17:55<51:18:35, 41.43s/it][2026-05-21 23:23:50,715] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 23:23:52,745] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 11%|█         | 543/5000 [7:18:29<48:33:17, 39.22s/it]
                                                       
{'loss': '1.104', 'grad_norm': '19.69', 'learning_rate': '1.819e-05', 'epoch': '0.1086'}

 11%|█         | 543/5000 [7:18:29<48:33:17, 39.22s/it][2026-05-21 23:24:23,959] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 975])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 23:24:24,920] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 974])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 11%|█         | 544/5000 [7:19:04<46:41:59, 37.73s/it]
                                                       
{'loss': '0.6638', 'grad_norm': '14.62', 'learning_rate': '1.819e-05', 'epoch': '0.1088'}

 11%|█         | 544/5000 [7:19:04<46:41:59, 37.73s/it][2026-05-21 23:24:53,467] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 973])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])
[2026-05-21 23:24:53,947] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 11%|█         | 545/5000 [7:19:51<50:13:01, 40.58s/it]
                                                       
{'loss': '0.8521', 'grad_norm': '31.22', 'learning_rate': '1.818e-05', 'epoch': '0.109'}

 11%|█         | 545/5000 [7:19:51<50:13:01, 40.58s/it][2026-05-21 23:25:41,153] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-21 23:25:41,299] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 979])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 11%|█         | 546/5000 [7:20:39<53:09:27, 42.97s/it]
                                                       
{'loss': '0.7785', 'grad_norm': '26.69', 'learning_rate': '1.818e-05', 'epoch': '0.1092'}

 11%|█         | 546/5000 [7:20:39<53:09:27, 42.97s/it][2026-05-21 23:26:27,161] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 668])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 25])
[2026-05-21 23:26:27,270] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 664])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 17])

 11%|█         | 547/5000 [7:21:30<55:47:55, 45.11s/it]
                                                       
{'loss': '0.7178', 'grad_norm': '10.68', 'learning_rate': '1.818e-05', 'epoch': '0.1094'}

 11%|█         | 547/5000 [7:21:30<55:47:55, 45.11s/it][2026-05-21 23:27:17,304] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 689])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 32])
[2026-05-21 23:27:17,351] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 636])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 16])

 11%|█         | 548/5000 [7:22:20<57:45:51, 46.71s/it]
                                                       
{'loss': '0.7558', 'grad_norm': '9.821', 'learning_rate': '1.817e-05', 'epoch': '0.1096'}

 11%|█         | 548/5000 [7:22:20<57:45:51, 46.71s/it][2026-05-21 23:28:07,132] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 258])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 18])
[2026-05-21 23:28:07,147] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 28])

 11%|█         | 549/5000 [7:23:29<66:00:28, 53.39s/it]
                                                       
{'loss': '1.237', 'grad_norm': '14.5', 'learning_rate': '1.817e-05', 'epoch': '0.1098'}

 11%|█         | 549/5000 [7:23:29<66:00:28, 53.39s/it][2026-05-21 23:29:16,098] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 16])
[2026-05-21 23:29:16,139] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 278])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 19])

 11%|█         | 550/5000 [7:24:38<71:38:29, 57.96s/it]
                                                       
{'loss': '1.197', 'grad_norm': '8.8', 'learning_rate': '1.816e-05', 'epoch': '0.11'}

 11%|█         | 550/5000 [7:24:38<71:38:29, 57.96s/it][2026-05-21 23:30:24,609] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 274])
[2026-05-21 23:30:25,021] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 284])

 11%|█         | 551/5000 [7:25:28<68:46:15, 55.65s/it]
                                                       
{'loss': '0.3671', 'grad_norm': '9.098', 'learning_rate': '1.816e-05', 'epoch': '0.1102'}

 11%|█         | 551/5000 [7:25:28<68:46:15, 55.65s/it][2026-05-21 23:31:15,286] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])
[2026-05-21 23:31:15,399] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])

 11%|█         | 552/5000 [7:26:18<66:46:13, 54.04s/it]
                                                       
{'loss': '0.09492', 'grad_norm': '4.961', 'learning_rate': '1.816e-05', 'epoch': '0.1104'}

 11%|█         | 552/5000 [7:26:18<66:46:13, 54.04s/it][2026-05-21 23:32:05,658] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-21 23:32:05,686] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 417])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 149])

 11%|█         | 553/5000 [7:26:51<58:52:15, 47.66s/it]
                                                       
{'loss': '2.182', 'grad_norm': '13.16', 'learning_rate': '1.815e-05', 'epoch': '0.1106'}

 11%|█         | 553/5000 [7:26:51<58:52:15, 47.66s/it][2026-05-21 23:32:38,364] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 179])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 92])
[2026-05-21 23:32:39,621] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 464])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 86])

 11%|█         | 554/5000 [7:27:24<53:37:15, 43.42s/it]
                                                       
{'loss': '2.195', 'grad_norm': '15.63', 'learning_rate': '1.815e-05', 'epoch': '0.1108'}

 11%|█         | 554/5000 [7:27:24<53:37:15, 43.42s/it][2026-05-21 23:33:20,927] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 638])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 23:33:21,355] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 629])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 980])

 11%|█         | 555/5000 [7:27:50<46:54:13, 37.99s/it]
                                                       
{'loss': '1.004', 'grad_norm': '21.2', 'learning_rate': '1.814e-05', 'epoch': '0.111'}

 11%|█         | 555/5000 [7:27:50<46:54:13, 37.99s/it][2026-05-21 23:33:46,937] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 642])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 23:33:46,969] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 623])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 11%|█         | 556/5000 [7:28:15<42:09:15, 34.15s/it]
                                                       
{'loss': '0.8676', 'grad_norm': '16.22', 'learning_rate': '1.814e-05', 'epoch': '0.1112'}

 11%|█         | 556/5000 [7:28:15<42:09:15, 34.15s/it][2026-05-21 23:34:02,935] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 28])
[2026-05-21 23:34:03,118] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 23])

 11%|█         | 557/5000 [7:28:43<39:50:59, 32.29s/it]
                                                       
{'loss': '0.3288', 'grad_norm': '11.02', 'learning_rate': '1.813e-05', 'epoch': '0.1114'}

 11%|█         | 557/5000 [7:28:43<39:50:59, 32.29s/it][2026-05-21 23:34:30,866] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 944])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 21])
[2026-05-21 23:34:31,042] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 30])

 11%|█         | 558/5000 [7:29:12<38:48:59, 31.46s/it]
                                                       
{'loss': '0.2357', 'grad_norm': '12.57', 'learning_rate': '1.813e-05', 'epoch': '0.1116'}

 11%|█         | 558/5000 [7:29:12<38:48:59, 31.46s/it][2026-05-21 23:34:59,670] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 202])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 23:34:59,733] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 218])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 11%|█         | 559/5000 [7:29:39<37:05:09, 30.06s/it]
                                                       
{'loss': '0.3349', 'grad_norm': '13.62', 'learning_rate': '1.813e-05', 'epoch': '0.1118'}

 11%|█         | 559/5000 [7:29:39<37:05:09, 30.06s/it][2026-05-21 23:35:26,532] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 202])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 23:35:26,560] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 202])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 11%|█         | 560/5000 [7:30:07<36:19:09, 29.45s/it]
                                                       
{'loss': '0.2846', 'grad_norm': '12.05', 'learning_rate': '1.812e-05', 'epoch': '0.112'}

 11%|█         | 560/5000 [7:30:07<36:19:09, 29.45s/it][2026-05-21 23:35:57,402] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 23:35:58,036] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 11%|█         | 561/5000 [7:31:02<45:43:12, 37.08s/it]
                                                       
{'loss': '1.891', 'grad_norm': '12.04', 'learning_rate': '1.812e-05', 'epoch': '0.1122'}

 11%|█         | 561/5000 [7:31:02<45:43:12, 37.08s/it][2026-05-21 23:36:52,488] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 23:36:52,844] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 11%|█         | 562/5000 [7:31:58<52:48:59, 42.84s/it]
                                                       
{'loss': '1.785', 'grad_norm': '12.14', 'learning_rate': '1.811e-05', 'epoch': '0.1124'}

 11%|█         | 562/5000 [7:31:58<52:48:59, 42.84s/it][2026-05-21 23:37:45,079] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 211])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 23:37:45,424] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 259])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])

 11%|█▏        | 563/5000 [7:32:42<52:58:53, 42.99s/it]
                                                       
{'loss': '1.274', 'grad_norm': '14.69', 'learning_rate': '1.811e-05', 'epoch': '0.1126'}

 11%|█▏        | 563/5000 [7:32:42<52:58:53, 42.99s/it][2026-05-21 23:38:28,563] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 266])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-21 23:38:28,585] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 312])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 11%|█▏        | 564/5000 [7:33:25<53:06:56, 43.11s/it]
                                                       
{'loss': '1.163', 'grad_norm': '14.72', 'learning_rate': '1.811e-05', 'epoch': '0.1128'}

 11%|█▏        | 564/5000 [7:33:25<53:06:56, 43.11s/it][2026-05-21 23:39:12,213] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 240])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 23:39:12,578] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 317])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 11%|█▏        | 565/5000 [7:34:10<53:44:47, 43.63s/it]
                                                       
{'loss': '2.853', 'grad_norm': '14.32', 'learning_rate': '1.81e-05', 'epoch': '0.113'}

 11%|█▏        | 565/5000 [7:34:10<53:44:47, 43.63s/it][2026-05-21 23:39:57,276] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 280])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 23:39:57,321] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 307])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 11%|█▏        | 566/5000 [7:34:55<54:22:52, 44.15s/it]
                                                       
{'loss': '2.831', 'grad_norm': '11.83', 'learning_rate': '1.81e-05', 'epoch': '0.1132'}

 11%|█▏        | 566/5000 [7:34:55<54:22:52, 44.15s/it][2026-05-21 23:40:43,165] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 272])
[2026-05-21 23:40:43,276] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])

 11%|█▏        | 567/5000 [7:35:44<55:52:24, 45.37s/it]
                                                       
{'loss': '0.9268', 'grad_norm': '12.2', 'learning_rate': '1.809e-05', 'epoch': '0.1134'}

 11%|█▏        | 567/5000 [7:35:44<55:52:24, 45.37s/it][2026-05-21 23:41:31,393] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 473])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 307])
[2026-05-21 23:41:32,060] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 505])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])

 11%|█▏        | 568/5000 [7:36:34<57:47:48, 46.95s/it]
                                                       
{'loss': '0.3435', 'grad_norm': '6.607', 'learning_rate': '1.809e-05', 'epoch': '0.1136'}

 11%|█▏        | 568/5000 [7:36:34<57:47:48, 46.95s/it][2026-05-21 23:42:21,570] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 281])
[2026-05-21 23:42:22,596] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 471])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])

 11%|█▏        | 569/5000 [7:37:03<51:12:54, 41.61s/it]
                                                       
{'loss': '0.4331', 'grad_norm': '10.81', 'learning_rate': '1.809e-05', 'epoch': '0.1138'}

 11%|█▏        | 569/5000 [7:37:03<51:12:54, 41.61s/it][2026-05-21 23:42:51,761] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 294])
[2026-05-21 23:42:52,321] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 502])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 337])

 11%|█▏        | 570/5000 [7:37:33<46:54:32, 38.12s/it]
                                                       
{'loss': '0.3536', 'grad_norm': '10.22', 'learning_rate': '1.808e-05', 'epoch': '0.114'}

 11%|█▏        | 570/5000 [7:37:33<46:54:32, 38.12s/it][2026-05-21 23:43:23,658] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 254])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 23:43:24,086] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 230])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 11%|█▏        | 571/5000 [7:38:03<43:57:33, 35.73s/it]
                                                       
{'loss': '2.37', 'grad_norm': '7.86', 'learning_rate': '1.808e-05', 'epoch': '0.1142'}

 11%|█▏        | 571/5000 [7:38:03<43:57:33, 35.73s/it][2026-05-21 23:43:53,882] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 279])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 23:43:54,825] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 219])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 11%|█▏        | 572/5000 [7:38:34<41:52:02, 34.04s/it]
                                                       
{'loss': '2.277', 'grad_norm': '6.792', 'learning_rate': '1.807e-05', 'epoch': '0.1144'}

 11%|█▏        | 572/5000 [7:38:34<41:52:02, 34.04s/it][2026-05-21 23:44:20,873] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 23:44:21,817] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 11%|█▏        | 573/5000 [7:39:02<39:45:10, 32.33s/it]
                                                       
{'loss': '0.4244', 'grad_norm': '10.35', 'learning_rate': '1.807e-05', 'epoch': '0.1146'}

 11%|█▏        | 573/5000 [7:39:02<39:45:10, 32.33s/it][2026-05-21 23:44:50,091] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-21 23:44:50,119] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 11%|█▏        | 574/5000 [7:39:32<38:55:57, 31.67s/it]
                                                       
{'loss': '0.03905', 'grad_norm': '3.643', 'learning_rate': '1.807e-05', 'epoch': '0.1148'}

 11%|█▏        | 574/5000 [7:39:32<38:55:57, 31.67s/it][2026-05-21 23:45:19,002] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 984])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-21 23:45:19,662] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 557])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 148])

 12%|█▏        | 575/5000 [7:40:01<37:49:38, 30.77s/it]
                                                       
{'loss': '0.378', 'grad_norm': '9.74', 'learning_rate': '1.806e-05', 'epoch': '0.115'}

 12%|█▏        | 575/5000 [7:40:01<37:49:38, 30.77s/it][2026-05-21 23:45:47,708] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 927])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 153])
[2026-05-21 23:45:47,750] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 67])

 12%|█▏        | 576/5000 [7:40:31<37:43:11, 30.69s/it]
                                                       
{'loss': '0.06636', 'grad_norm': '6.017', 'learning_rate': '1.806e-05', 'epoch': '0.1152'}

 12%|█▏        | 576/5000 [7:40:31<37:43:11, 30.69s/it][2026-05-21 23:46:20,960] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 23:46:22,868] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 555])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

 12%|█▏        | 577/5000 [7:41:20<44:29:01, 36.21s/it]
                                                       
{'loss': '1.4', 'grad_norm': '11.54', 'learning_rate': '1.805e-05', 'epoch': '0.1154'}

 12%|█▏        | 577/5000 [7:41:20<44:29:01, 36.21s/it][2026-05-21 23:47:09,520] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 23:47:10,622] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 951])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 12%|█▏        | 578/5000 [7:42:09<49:03:41, 39.94s/it]
                                                       
{'loss': '0.6504', 'grad_norm': '8.401', 'learning_rate': '1.805e-05', 'epoch': '0.1156'}

 12%|█▏        | 578/5000 [7:42:09<49:03:41, 39.94s/it][2026-05-21 23:47:58,311] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 23:47:58,434] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 412])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 12%|█▏        | 579/5000 [7:42:47<48:12:06, 39.25s/it]
                                                       
{'loss': '3.95', 'grad_norm': '18.12', 'learning_rate': '1.804e-05', 'epoch': '0.1158'}

 12%|█▏        | 579/5000 [7:42:47<48:12:06, 39.25s/it][2026-05-21 23:48:35,546] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 493])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 764])
[2026-05-21 23:48:37,080] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 12%|█▏        | 580/5000 [7:43:25<47:46:08, 38.91s/it]
                                                       
{'loss': '3.763', 'grad_norm': '15.45', 'learning_rate': '1.804e-05', 'epoch': '0.116'}

 12%|█▏        | 580/5000 [7:43:25<47:46:08, 38.91s/it][2026-05-21 23:49:14,314] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 267])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-21 23:49:15,641] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 282])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 12%|█▏        | 581/5000 [7:44:10<50:10:56, 40.88s/it]
                                                       
{'loss': '0.8', 'grad_norm': '20.35', 'learning_rate': '1.804e-05', 'epoch': '0.1162'}

 12%|█▏        | 581/5000 [7:44:10<50:10:56, 40.88s/it][2026-05-21 23:50:00,982] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 285])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])
[2026-05-21 23:50:01,104] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 306])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 12%|█▏        | 582/5000 [7:44:56<52:08:43, 42.49s/it]
                                                       
{'loss': '0.7529', 'grad_norm': '21.42', 'learning_rate': '1.803e-05', 'epoch': '0.1164'}

 12%|█▏        | 582/5000 [7:44:56<52:08:43, 42.49s/it][2026-05-21 23:50:43,920] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 23:50:44,200] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 250])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 12%|█▏        | 583/5000 [7:46:04<61:24:39, 50.05s/it]
                                                       
{'loss': '1.472', 'grad_norm': '13.83', 'learning_rate': '1.803e-05', 'epoch': '0.1166'}

 12%|█▏        | 583/5000 [7:46:04<61:24:39, 50.05s/it][2026-05-21 23:51:51,611] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-21 23:51:52,212] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])

 12%|█▏        | 584/5000 [7:47:12<67:55:37, 55.38s/it]
                                                       
{'loss': '0.9365', 'grad_norm': '15.15', 'learning_rate': '1.802e-05', 'epoch': '0.1168'}

 12%|█▏        | 584/5000 [7:47:12<67:55:37, 55.38s/it][2026-05-21 23:52:58,616] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 209])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 307])
[2026-05-21 23:52:59,571] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 292])

 12%|█▏        | 585/5000 [7:48:02<65:59:27, 53.81s/it]
                                                       
{'loss': '0.5201', 'grad_norm': '10.85', 'learning_rate': '1.802e-05', 'epoch': '0.117'}

 12%|█▏        | 585/5000 [7:48:02<65:59:27, 53.81s/it][2026-05-21 23:53:48,999] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])
[2026-05-21 23:53:49,565] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])

 12%|█▏        | 586/5000 [7:48:53<64:46:02, 52.82s/it]
                                                       
{'loss': '0.3313', 'grad_norm': '9.888', 'learning_rate': '1.802e-05', 'epoch': '0.1172'}

 12%|█▏        | 586/5000 [7:48:53<64:46:02, 52.82s/it][2026-05-21 23:54:40,916] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 398])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])
[2026-05-21 23:54:41,332] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 12%|█▏        | 587/5000 [7:49:40<62:41:27, 51.14s/it]
                                                       
{'loss': '0.4022', 'grad_norm': '11.01', 'learning_rate': '1.801e-05', 'epoch': '0.1174'}

 12%|█▏        | 587/5000 [7:49:40<62:41:27, 51.14s/it][2026-05-21 23:55:27,929] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 23:55:28,725] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

 12%|█▏        | 588/5000 [7:50:27<61:22:40, 50.08s/it]
                                                       
{'loss': '0.3726', 'grad_norm': '5.216', 'learning_rate': '1.801e-05', 'epoch': '0.1176'}

 12%|█▏        | 588/5000 [7:50:27<61:22:40, 50.08s/it][2026-05-21 23:56:15,772] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 23:56:17,179] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 12%|█▏        | 589/5000 [7:50:58<54:13:20, 44.25s/it]
                                                       
{'loss': '0.5011', 'grad_norm': '11.8', 'learning_rate': '1.8e-05', 'epoch': '0.1178'}

 12%|█▏        | 589/5000 [7:50:58<54:13:20, 44.25s/it][2026-05-21 23:56:46,761] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 177])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 23:56:49,141] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 12%|█▏        | 590/5000 [7:51:30<49:35:21, 40.48s/it]
                                                       
{'loss': '0.2053', 'grad_norm': '8.734', 'learning_rate': '1.8e-05', 'epoch': '0.118'}

 12%|█▏        | 590/5000 [7:51:30<49:35:21, 40.48s/it][2026-05-21 23:57:21,336] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 145])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-21 23:57:21,390] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 145])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

 12%|█▏        | 591/5000 [7:52:00<45:59:24, 37.55s/it]
                                                       
{'loss': '0.718', 'grad_norm': '18.12', 'learning_rate': '1.8e-05', 'epoch': '0.1182'}

 12%|█▏        | 591/5000 [7:52:00<45:59:24, 37.55s/it][2026-05-21 23:57:49,754] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 136])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-21 23:57:51,827] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 218])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 12%|█▏        | 592/5000 [7:52:30<43:05:19, 35.19s/it]
                                                       
{'loss': '0.5298', 'grad_norm': '17.28', 'learning_rate': '1.799e-05', 'epoch': '0.1184'}

 12%|█▏        | 592/5000 [7:52:30<43:05:19, 35.19s/it][2026-05-21 23:58:18,672] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 170])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 23:58:20,607] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 160])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 12%|█▏        | 593/5000 [7:53:21<48:41:52, 39.78s/it]
                                                       
{'loss': '2.675', 'grad_norm': '38.81', 'learning_rate': '1.799e-05', 'epoch': '0.1186'}

 12%|█▏        | 593/5000 [7:53:21<48:41:52, 39.78s/it][2026-05-21 23:59:09,496] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 178])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-21 23:59:11,035] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 142])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 12%|█▏        | 594/5000 [7:54:11<52:30:02, 42.90s/it]
                                                       
{'loss': '2.701', 'grad_norm': '38.7', 'learning_rate': '1.798e-05', 'epoch': '0.1188'}

 12%|█▏        | 594/5000 [7:54:11<52:30:02, 42.90s/it][2026-05-21 23:59:59,106] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 694])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-21 23:59:59,358] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 12%|█▏        | 595/5000 [7:54:59<54:25:27, 44.48s/it]
                                                       
{'loss': '0.6304', 'grad_norm': '9.609', 'learning_rate': '1.798e-05', 'epoch': '0.119'}

 12%|█▏        | 595/5000 [7:54:59<54:25:27, 44.48s/it][2026-05-22 00:00:47,165] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 00:00:47,809] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 668])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 12%|█▏        | 596/5000 [7:55:49<56:34:12, 46.24s/it]
                                                       
{'loss': '0.4033', 'grad_norm': '8.442', 'learning_rate': '1.798e-05', 'epoch': '0.1192'}

 12%|█▏        | 596/5000 [7:55:49<56:34:12, 46.24s/it][2026-05-22 00:01:38,746] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 272])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 00:01:39,289] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 280])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])

 12%|█▏        | 597/5000 [7:56:37<57:12:08, 46.77s/it]
                                                       
{'loss': '1.548', 'grad_norm': '10.54', 'learning_rate': '1.797e-05', 'epoch': '0.1194'}

 12%|█▏        | 597/5000 [7:56:37<57:12:08, 46.77s/it][2026-05-22 00:02:26,349] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 256])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])
[2026-05-22 00:02:27,988] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 288])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 12%|█▏        | 598/5000 [7:57:26<57:45:54, 47.24s/it]
                                                       
{'loss': '1.442', 'grad_norm': '11.58', 'learning_rate': '1.797e-05', 'epoch': '0.1196'}

 12%|█▏        | 598/5000 [7:57:26<57:45:54, 47.24s/it][2026-05-22 00:03:19,102] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 00:03:20,610] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 975])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

 12%|█▏        | 599/5000 [7:58:13<57:52:47, 47.35s/it]
                                                       
{'loss': '1.521', 'grad_norm': '9.683', 'learning_rate': '1.796e-05', 'epoch': '0.1198'}

 12%|█▏        | 599/5000 [7:58:13<57:52:47, 47.35s/it][2026-05-22 00:04:07,111] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-22 00:04:08,232] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 977])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 12%|█▏        | 600/5000 [7:59:01<58:03:04, 47.50s/it]
                                                       
{'loss': '1.241', 'grad_norm': '8.751', 'learning_rate': '1.796e-05', 'epoch': '0.12'}

 12%|█▏        | 600/5000 [7:59:01<58:03:04, 47.50s/it][2026-05-22 00:04:49,786] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 975])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 00:04:49,789] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 12%|█▏        | 601/5000 [7:59:39<54:28:50, 44.59s/it]
                                                       
{'loss': '3.823', 'grad_norm': '9.264', 'learning_rate': '1.796e-05', 'epoch': '0.1202'}

 12%|█▏        | 601/5000 [7:59:39<54:28:50, 44.59s/it][2026-05-22 00:05:27,455] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 00:05:27,780] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 981])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 12%|█▏        | 602/5000 [8:00:18<52:26:14, 42.92s/it]
                                                       
{'loss': '3.686', 'grad_norm': '5.871', 'learning_rate': '1.795e-05', 'epoch': '0.1204'}

 12%|█▏        | 602/5000 [8:00:18<52:26:14, 42.92s/it][2026-05-22 00:06:07,835] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 211])
[2026-05-22 00:06:07,918] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 12%|█▏        | 603/5000 [8:01:07<54:35:42, 44.70s/it]
                                                       
{'loss': '0.305', 'grad_norm': '10.98', 'learning_rate': '1.795e-05', 'epoch': '0.1206'}

 12%|█▏        | 603/5000 [8:01:07<54:35:42, 44.70s/it][2026-05-22 00:06:56,570] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 128])
[2026-05-22 00:06:57,598] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 158])

 12%|█▏        | 604/5000 [8:01:55<55:55:12, 45.79s/it]
                                                       
{'loss': '0.4975', 'grad_norm': '10.53', 'learning_rate': '1.794e-05', 'epoch': '0.1208'}

 12%|█▏        | 604/5000 [8:01:55<55:55:12, 45.79s/it][2026-05-22 00:07:51,714] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])
[2026-05-22 00:07:51,907] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])

 12%|█▏        | 605/5000 [8:02:41<55:53:40, 45.78s/it]
                                                       
{'loss': '0.3345', 'grad_norm': '14.22', 'learning_rate': '1.794e-05', 'epoch': '0.121'}

 12%|█▏        | 605/5000 [8:02:41<55:53:40, 45.78s/it][2026-05-22 00:08:37,521] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 330])
[2026-05-22 00:08:41,114] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 248])

 12%|█▏        | 606/5000 [8:03:27<55:51:47, 45.77s/it]
                                                       
{'loss': '0.3309', 'grad_norm': '12.62', 'learning_rate': '1.793e-05', 'epoch': '0.1212'}

 12%|█▏        | 606/5000 [8:03:27<55:51:47, 45.77s/it][2026-05-22 00:09:17,621] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 100])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 00:09:18,780] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 96])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 12%|█▏        | 607/5000 [8:04:15<56:51:57, 46.60s/it]
                                                       
{'loss': '0.3979', 'grad_norm': '13.75', 'learning_rate': '1.793e-05', 'epoch': '0.1214'}

 12%|█▏        | 607/5000 [8:04:15<56:51:57, 46.60s/it][2026-05-22 00:10:08,106] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 96])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])
[2026-05-22 00:10:09,244] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 84])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

 12%|█▏        | 608/5000 [8:05:02<56:52:28, 46.62s/it]
                                                       
{'loss': '0.3683', 'grad_norm': '13.21', 'learning_rate': '1.793e-05', 'epoch': '0.1216'}

 12%|█▏        | 608/5000 [8:05:02<56:52:28, 46.62s/it][2026-05-22 00:10:50,597] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 328])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 00:10:50,729] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 238])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 12%|█▏        | 609/5000 [8:05:48<56:47:51, 46.57s/it]
                                                       
{'loss': '0.1918', 'grad_norm': '6.716', 'learning_rate': '1.792e-05', 'epoch': '0.1218'}

 12%|█▏        | 609/5000 [8:05:48<56:47:51, 46.57s/it][2026-05-22 00:11:37,171] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 215])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 00:11:37,497] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 216])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 12%|█▏        | 610/5000 [8:06:35<56:46:44, 46.56s/it]
                                                       
{'loss': '0.1284', 'grad_norm': '6.491', 'learning_rate': '1.792e-05', 'epoch': '0.122'}

 12%|█▏        | 610/5000 [8:06:35<56:46:44, 46.56s/it][2026-05-22 00:12:24,612] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 207])
[2026-05-22 00:12:24,629] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 142])

 12%|█▏        | 611/5000 [8:07:23<57:16:11, 46.97s/it]
                                                       
{'loss': '0.4355', 'grad_norm': '13', 'learning_rate': '1.791e-05', 'epoch': '0.1222'}

 12%|█▏        | 611/5000 [8:07:23<57:16:11, 46.97s/it][2026-05-22 00:13:12,779] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 00:13:13,516] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 115])

 12%|█▏        | 612/5000 [8:08:13<58:23:53, 47.91s/it]
                                                       
{'loss': '0.0673', 'grad_norm': '4.836', 'learning_rate': '1.791e-05', 'epoch': '0.1224'}

 12%|█▏        | 612/5000 [8:08:13<58:23:53, 47.91s/it][2026-05-22 00:14:11,424] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-22 00:14:11,856] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 12%|█▏        | 613/5000 [8:09:03<59:18:17, 48.67s/it]
                                                       
{'loss': '0.9614', 'grad_norm': '12.5', 'learning_rate': '1.791e-05', 'epoch': '0.1226'}

 12%|█▏        | 613/5000 [8:09:03<59:18:17, 48.67s/it][2026-05-22 00:15:01,094] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 00:15:01,108] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 12%|█▏        | 614/5000 [8:09:53<59:51:38, 49.13s/it]
                                                       
{'loss': '1.223', 'grad_norm': '16.23', 'learning_rate': '1.79e-05', 'epoch': '0.1228'}

 12%|█▏        | 614/5000 [8:09:53<59:51:38, 49.13s/it][2026-05-22 00:15:45,085] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 135])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 00:15:45,341] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 101])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 972])

 12%|█▏        | 615/5000 [8:10:44<60:15:44, 49.47s/it]
                                                       
{'loss': '0.4556', 'grad_norm': '12.14', 'learning_rate': '1.79e-05', 'epoch': '0.123'}

 12%|█▏        | 615/5000 [8:10:44<60:15:44, 49.47s/it][2026-05-22 00:16:36,303] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 82])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])
[2026-05-22 00:16:36,956] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 116])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 12%|█▏        | 616/5000 [8:11:32<59:50:19, 49.14s/it]
                                                       
{'loss': '0.7772', 'grad_norm': '11.85', 'learning_rate': '1.789e-05', 'epoch': '0.1232'}

 12%|█▏        | 616/5000 [8:11:32<59:50:19, 49.14s/it][2026-05-22 00:17:20,337] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])
[2026-05-22 00:17:21,598] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 417])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 12%|█▏        | 617/5000 [8:12:20<59:21:15, 48.75s/it]
                                                       
{'loss': '0.1758', 'grad_norm': '7.201', 'learning_rate': '1.789e-05', 'epoch': '0.1234'}

 12%|█▏        | 617/5000 [8:12:20<59:21:15, 48.75s/it][2026-05-22 00:18:09,195] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 166])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 00:18:10,374] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 505])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 12%|█▏        | 618/5000 [8:13:09<59:16:32, 48.70s/it]
                                                       
{'loss': '0.1845', 'grad_norm': '8.672', 'learning_rate': '1.789e-05', 'epoch': '0.1236'}

 12%|█▏        | 618/5000 [8:13:09<59:16:32, 48.70s/it][2026-05-22 00:18:57,144] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 255])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 00:18:57,704] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 254])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 12%|█▏        | 619/5000 [8:14:19<67:11:47, 55.22s/it]
                                                       
{'loss': '0.7564', 'grad_norm': '9.117', 'learning_rate': '1.788e-05', 'epoch': '0.1238'}

 12%|█▏        | 619/5000 [8:14:19<67:11:47, 55.22s/it][2026-05-22 00:20:07,499] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 00:20:07,508] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 237])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 12%|█▏        | 620/5000 [8:15:29<72:36:40, 59.68s/it]
                                                       
{'loss': '0.8381', 'grad_norm': '10.78', 'learning_rate': '1.788e-05', 'epoch': '0.124'}

 12%|█▏        | 620/5000 [8:15:29<72:36:40, 59.68s/it][2026-05-22 00:21:21,718] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1015])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 00:21:22,345] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 12%|█▏        | 621/5000 [8:16:20<69:22:09, 57.03s/it]
                                                       
{'loss': '0.5493', 'grad_norm': '12.2', 'learning_rate': '1.787e-05', 'epoch': '0.1242'}

 12%|█▏        | 621/5000 [8:16:20<69:22:09, 57.03s/it][2026-05-22 00:22:12,948] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 00:22:13,031] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 12%|█▏        | 622/5000 [8:17:10<66:50:37, 54.97s/it]
                                                       
{'loss': '1.208', 'grad_norm': '16', 'learning_rate': '1.787e-05', 'epoch': '0.1244'}

 12%|█▏        | 622/5000 [8:17:10<66:50:37, 54.97s/it][2026-05-22 00:23:18,440] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 87])
[2026-05-22 00:23:19,017] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 170])

 12%|█▏        | 623/5000 [8:17:57<64:01:44, 52.66s/it]
                                                       
{'loss': '0.2562', 'grad_norm': '7.62', 'learning_rate': '1.787e-05', 'epoch': '0.1246'}

 12%|█▏        | 623/5000 [8:17:57<64:01:44, 52.66s/it][2026-05-22 00:24:04,508] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 123])
[2026-05-22 00:24:06,425] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 80])

 12%|█▏        | 624/5000 [8:18:49<63:29:43, 52.24s/it]
                                                       
{'loss': '0.4371', 'grad_norm': '7.898', 'learning_rate': '1.786e-05', 'epoch': '0.1248'}

 12%|█▏        | 624/5000 [8:18:49<63:29:43, 52.24s/it][2026-05-22 00:24:35,868] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 554])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 18])
[2026-05-22 00:24:35,899] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 16])

 12%|█▎        | 625/5000 [8:19:35<61:29:38, 50.60s/it]
                                                       
{'loss': '0.3691', 'grad_norm': '10.92', 'learning_rate': '1.786e-05', 'epoch': '0.125'}

 12%|█▎        | 625/5000 [8:19:35<61:29:38, 50.60s/it][2026-05-22 00:25:22,565] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 926])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 14])
[2026-05-22 00:25:22,583] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 970])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 8])

 13%|█▎        | 626/5000 [8:20:21<59:46:48, 49.20s/it]
                                                       
{'loss': '0.7716', 'grad_norm': '11.56', 'learning_rate': '1.785e-05', 'epoch': '0.1252'}

 13%|█▎        | 626/5000 [8:20:21<59:46:48, 49.20s/it][2026-05-22 00:26:11,090] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])
[2026-05-22 00:26:11,129] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 97])

 13%|█▎        | 627/5000 [8:21:32<67:28:45, 55.55s/it]
                                                       
{'loss': '0.5991', 'grad_norm': '9.599', 'learning_rate': '1.785e-05', 'epoch': '0.1254'}

 13%|█▎        | 627/5000 [8:21:32<67:28:45, 55.55s/it][2026-05-22 00:27:21,314] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 368])
[2026-05-22 00:27:21,469] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 155])

 13%|█▎        | 628/5000 [8:22:41<72:39:14, 59.83s/it]
                                                       
{'loss': '0.9171', 'grad_norm': '11', 'learning_rate': '1.784e-05', 'epoch': '0.1256'}

 13%|█▎        | 628/5000 [8:22:41<72:39:14, 59.83s/it][2026-05-22 00:28:42,094] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 129])
[2026-05-22 00:28:43,433] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 281])

 13%|█▎        | 629/5000 [8:23:31<68:50:00, 56.69s/it]
                                                       
{'loss': '0.5126', 'grad_norm': '11.74', 'learning_rate': '1.784e-05', 'epoch': '0.1258'}

 13%|█▎        | 629/5000 [8:23:31<68:50:00, 56.69s/it][2026-05-22 00:29:30,392] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 134])
[2026-05-22 00:29:36,366] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])

 13%|█▎        | 630/5000 [8:24:19<65:49:21, 54.22s/it]
                                                       
{'loss': '0.5009', 'grad_norm': '9.241', 'learning_rate': '1.784e-05', 'epoch': '0.126'}

 13%|█▎        | 630/5000 [8:24:19<65:49:21, 54.22s/it][2026-05-22 00:30:30,618] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 100])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])
[2026-05-22 00:30:34,150] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 93])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])

 13%|█▎        | 631/5000 [8:25:09<64:00:47, 52.75s/it]
                                                       
{'loss': '1.099', 'grad_norm': '15.02', 'learning_rate': '1.783e-05', 'epoch': '0.1262'}

 13%|█▎        | 631/5000 [8:25:09<64:00:47, 52.75s/it][2026-05-22 00:31:21,712] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 85])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])
[2026-05-22 00:31:27,042] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 108])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 13%|█▎        | 632/5000 [8:25:56<62:02:13, 51.13s/it]
                                                       
{'loss': '0.8669', 'grad_norm': '14.1', 'learning_rate': '1.783e-05', 'epoch': '0.1264'}

 13%|█▎        | 632/5000 [8:25:56<62:02:13, 51.13s/it][2026-05-22 00:31:55,562] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])
[2026-05-22 00:31:58,442] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 13%|█▎        | 633/5000 [8:26:52<63:39:59, 52.48s/it]
                                                       
{'loss': '1.96', 'grad_norm': '32.65', 'learning_rate': '1.782e-05', 'epoch': '0.1266'}

 13%|█▎        | 633/5000 [8:26:52<63:39:59, 52.48s/it][2026-05-22 00:32:51,267] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 00:32:51,868] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 13%|█▎        | 634/5000 [8:27:50<65:41:32, 54.17s/it]
                                                       
{'loss': '2.695', 'grad_norm': '38.36', 'learning_rate': '1.782e-05', 'epoch': '0.1268'}

 13%|█▎        | 634/5000 [8:27:50<65:41:32, 54.17s/it][2026-05-22 00:33:55,360] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 295])
[2026-05-22 00:33:58,376] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 13%|█▎        | 635/5000 [8:28:49<67:36:48, 55.76s/it]
                                                       
{'loss': '0.4112', 'grad_norm': '7.989', 'learning_rate': '1.782e-05', 'epoch': '0.127'}

 13%|█▎        | 635/5000 [8:28:49<67:36:48, 55.76s/it][2026-05-22 00:34:55,891] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])
[2026-05-22 00:34:57,842] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 142])

 13%|█▎        | 636/5000 [8:29:45<67:30:04, 55.68s/it]
                                                       
{'loss': '0.4363', 'grad_norm': '14.48', 'learning_rate': '1.781e-05', 'epoch': '0.1272'}

 13%|█▎        | 636/5000 [8:29:45<67:30:04, 55.68s/it][2026-05-22 00:35:32,195] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 94])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 970])
[2026-05-22 00:35:34,830] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 89])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 13%|█▎        | 637/5000 [8:30:35<65:29:21, 54.04s/it]
                                                       
{'loss': '1.314', 'grad_norm': '20.3', 'learning_rate': '1.781e-05', 'epoch': '0.1274'}

 13%|█▎        | 637/5000 [8:30:35<65:29:21, 54.04s/it][2026-05-22 00:36:23,518] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 90])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 00:36:25,279] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 83])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 13%|█▎        | 638/5000 [8:31:28<65:05:12, 53.72s/it]
                                                       
{'loss': '1.041', 'grad_norm': '14.6', 'learning_rate': '1.78e-05', 'epoch': '0.1276'}

 13%|█▎        | 638/5000 [8:31:28<65:05:12, 53.72s/it][2026-05-22 00:37:28,465] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 354])
[2026-05-22 00:37:32,256] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 140])

 13%|█▎        | 639/5000 [8:32:42<72:28:42, 59.83s/it]
                                                       
{'loss': '0.6324', 'grad_norm': '11.52', 'learning_rate': '1.78e-05', 'epoch': '0.1278'}

 13%|█▎        | 639/5000 [8:32:42<72:28:42, 59.83s/it][2026-05-22 00:38:42,164] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 98])
[2026-05-22 00:38:45,037] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 120])

 13%|█▎        | 640/5000 [8:33:55<77:18:37, 63.83s/it]
                                                       
{'loss': '0.3592', 'grad_norm': '8.412', 'learning_rate': '1.78e-05', 'epoch': '0.128'}

 13%|█▎        | 640/5000 [8:33:55<77:18:37, 63.83s/it][2026-05-22 00:39:46,930] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 218])
[2026-05-22 00:39:47,471] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 149])

 13%|█▎        | 641/5000 [8:34:47<72:59:46, 60.29s/it]
                                                       
{'loss': '0.2633', 'grad_norm': '7.596', 'learning_rate': '1.779e-05', 'epoch': '0.1282'}

 13%|█▎        | 641/5000 [8:34:47<72:59:46, 60.29s/it][2026-05-22 00:40:38,995] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 100])
[2026-05-22 00:40:39,315] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 184])

 13%|█▎        | 642/5000 [8:35:38<69:36:24, 57.50s/it]
                                                       
{'loss': '0.2844', 'grad_norm': '7.652', 'learning_rate': '1.779e-05', 'epoch': '0.1284'}

 13%|█▎        | 642/5000 [8:35:38<69:36:24, 57.50s/it][2026-05-22 00:41:46,816] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1020])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 144])
[2026-05-22 00:41:51,679] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 13%|█▎        | 643/5000 [8:36:25<65:48:38, 54.38s/it]
                                                       
{'loss': '0.9319', 'grad_norm': '14.94', 'learning_rate': '1.778e-05', 'epoch': '0.1286'}

 13%|█▎        | 643/5000 [8:36:25<65:48:38, 54.38s/it][2026-05-22 00:42:32,588] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 123])
[2026-05-22 00:42:33,136] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

 13%|█▎        | 644/5000 [8:37:13<63:26:33, 52.43s/it]
                                                       
{'loss': '0.8582', 'grad_norm': '12', 'learning_rate': '1.778e-05', 'epoch': '0.1288'}

 13%|█▎        | 644/5000 [8:37:13<63:26:33, 52.43s/it][2026-05-22 00:43:01,721] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 94])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 00:43:03,783] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 86])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 13%|█▎        | 645/5000 [8:38:06<63:28:31, 52.47s/it]
                                                       
{'loss': '1.121', 'grad_norm': '11.71', 'learning_rate': '1.778e-05', 'epoch': '0.129'}

 13%|█▎        | 645/5000 [8:38:06<63:28:31, 52.47s/it][2026-05-22 00:43:55,150] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 92])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])
[2026-05-22 00:43:55,370] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 104])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 972])

 13%|█▎        | 646/5000 [8:38:54<62:07:25, 51.37s/it]
                                                       
{'loss': '1.032', 'grad_norm': '9.905', 'learning_rate': '1.777e-05', 'epoch': '0.1292'}

 13%|█▎        | 646/5000 [8:38:54<62:07:25, 51.37s/it][2026-05-22 00:44:45,931] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 91])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 00:44:49,115] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 90])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])

 13%|█▎        | 647/5000 [8:39:47<62:34:30, 51.75s/it]
                                                       
{'loss': '1.212', 'grad_norm': '12.29', 'learning_rate': '1.777e-05', 'epoch': '0.1294'}

 13%|█▎        | 647/5000 [8:39:47<62:34:30, 51.75s/it][2026-05-22 00:45:38,324] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 83])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])
[2026-05-22 00:45:40,680] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 91])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 13%|█▎        | 648/5000 [8:40:38<62:14:43, 51.49s/it]
                                                       
{'loss': '1.091', 'grad_norm': '10.55', 'learning_rate': '1.776e-05', 'epoch': '0.1296'}

 13%|█▎        | 648/5000 [8:40:38<62:14:43, 51.49s/it][2026-05-22 00:46:28,762] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 99])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 00:46:29,396] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 97])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

 13%|█▎        | 649/5000 [8:41:29<61:58:14, 51.27s/it]
                                                       
{'loss': '0.2923', 'grad_norm': '8.968', 'learning_rate': '1.776e-05', 'epoch': '0.1298'}

 13%|█▎        | 649/5000 [8:41:29<61:58:14, 51.27s/it][2026-05-22 00:47:19,927] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 98])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 00:47:19,965] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 115])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 13%|█▎        | 650/5000 [8:42:39<68:49:41, 56.96s/it]
                                                       
{'loss': '0.4742', 'grad_norm': '9.769', 'learning_rate': '1.776e-05', 'epoch': '0.13'}

 13%|█▎        | 650/5000 [8:42:39<68:49:41, 56.96s/it][2026-05-22 00:48:33,275] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 00:48:33,355] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 13%|█▎        | 651/5000 [8:43:30<66:33:22, 55.09s/it]
                                                       
{'loss': '0.7792', 'grad_norm': '12.72', 'learning_rate': '1.775e-05', 'epoch': '0.1302'}

 13%|█▎        | 651/5000 [8:43:30<66:33:22, 55.09s/it][2026-05-22 00:49:24,029] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 00:49:25,162] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 13%|█▎        | 652/5000 [8:44:20<64:43:46, 53.59s/it]
                                                       
{'loss': '0.8887', 'grad_norm': '10.01', 'learning_rate': '1.775e-05', 'epoch': '0.1304'}

 13%|█▎        | 652/5000 [8:44:20<64:43:46, 53.59s/it][2026-05-22 00:50:18,968] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])
[2026-05-22 00:50:20,488] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])

 13%|█▎        | 653/5000 [8:45:10<63:23:38, 52.50s/it]
                                                       
{'loss': '0.8619', 'grad_norm': '13.54', 'learning_rate': '1.774e-05', 'epoch': '0.1306'}

 13%|█▎        | 653/5000 [8:45:10<63:23:38, 52.50s/it][2026-05-22 00:51:10,065] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 299])
[2026-05-22 00:51:10,090] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 338])

 13%|█▎        | 654/5000 [8:45:59<62:21:49, 51.66s/it]
                                                       
{'loss': '1.411', 'grad_norm': '14.15', 'learning_rate': '1.774e-05', 'epoch': '0.1308'}

 13%|█▎        | 654/5000 [8:45:59<62:21:49, 51.66s/it][2026-05-22 00:51:54,637] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 00:51:56,764] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

 13%|█▎        | 655/5000 [8:46:50<61:50:20, 51.24s/it]
                                                       
{'loss': '1.04', 'grad_norm': '11.17', 'learning_rate': '1.773e-05', 'epoch': '0.131'}

 13%|█▎        | 655/5000 [8:46:50<61:50:20, 51.24s/it][2026-05-22 00:52:42,442] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 00:52:43,330] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 13%|█▎        | 656/5000 [8:47:41<61:46:54, 51.20s/it]
                                                       
{'loss': '1.099', 'grad_norm': '12.42', 'learning_rate': '1.773e-05', 'epoch': '0.1312'}

 13%|█▎        | 656/5000 [8:47:41<61:46:54, 51.20s/it][2026-05-22 00:53:42,805] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 00:53:46,992] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

 13%|█▎        | 657/5000 [8:48:54<69:39:25, 57.74s/it]
                                                       
{'loss': '0.5681', 'grad_norm': '13.52', 'learning_rate': '1.773e-05', 'epoch': '0.1314'}

 13%|█▎        | 657/5000 [8:48:54<69:39:25, 57.74s/it][2026-05-22 00:54:52,623] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 975])
[2026-05-22 00:54:56,333] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

 13%|█▎        | 658/5000 [8:50:04<74:16:10, 61.58s/it]
                                                       
{'loss': '0.4838', 'grad_norm': '15.76', 'learning_rate': '1.772e-05', 'epoch': '0.1316'}

 13%|█▎        | 658/5000 [8:50:04<74:16:10, 61.58s/it][2026-05-22 00:56:25,009] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 00:56:25,492] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 13%|█▎        | 659/5000 [8:50:59<71:41:28, 59.45s/it]
                                                       
{'loss': '0.6955', 'grad_norm': '12.21', 'learning_rate': '1.772e-05', 'epoch': '0.1318'}

 13%|█▎        | 659/5000 [8:50:59<71:41:28, 59.45s/it][2026-05-22 00:57:28,619] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 00:57:31,581] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 13%|█▎        | 660/5000 [8:51:48<67:48:24, 56.25s/it]
                                                       
{'loss': '0.09368', 'grad_norm': '8.498', 'learning_rate': '1.771e-05', 'epoch': '0.132'}

 13%|█▎        | 660/5000 [8:51:48<67:48:24, 56.25s/it][2026-05-22 00:57:38,411] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 00:57:38,614] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 13%|█▎        | 661/5000 [8:53:03<74:45:14, 62.02s/it]
                                                       
{'loss': '0.9441', 'grad_norm': '13.74', 'learning_rate': '1.771e-05', 'epoch': '0.1322'}

 13%|█▎        | 661/5000 [8:53:03<74:45:14, 62.02s/it][2026-05-22 00:58:53,309] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 00:58:53,498] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 13%|█▎        | 662/5000 [8:54:14<77:53:06, 64.64s/it]
                                                       
{'loss': '0.5242', 'grad_norm': '13.5', 'learning_rate': '1.771e-05', 'epoch': '0.1324'}

 13%|█▎        | 662/5000 [8:54:14<77:53:06, 64.64s/it][2026-05-22 01:00:01,877] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 278])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 01:00:02,467] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 267])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 13%|█▎        | 663/5000 [8:55:25<80:03:19, 66.45s/it]
                                                       
{'loss': '0.6262', 'grad_norm': '11.27', 'learning_rate': '1.77e-05', 'epoch': '0.1326'}

 13%|█▎        | 663/5000 [8:55:25<80:03:19, 66.45s/it][2026-05-22 01:01:12,856] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 318])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 01:01:13,467] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 498])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 13%|█▎        | 664/5000 [8:56:34<81:13:34, 67.44s/it]
                                                       
{'loss': '0.7476', 'grad_norm': '11.8', 'learning_rate': '1.77e-05', 'epoch': '0.1328'}

 13%|█▎        | 664/5000 [8:56:34<81:13:34, 67.44s/it][2026-05-22 01:02:25,515] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 229])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])
[2026-05-22 01:02:26,357] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

 13%|█▎        | 665/5000 [8:57:48<83:30:13, 69.35s/it]
                                                       
{'loss': '0.4226', 'grad_norm': '11.16', 'learning_rate': '1.769e-05', 'epoch': '0.133'}

 13%|█▎        | 665/5000 [8:57:48<83:30:13, 69.35s/it][2026-05-22 01:03:37,462] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 01:03:38,051] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 197])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 13%|█▎        | 666/5000 [8:58:59<84:12:30, 69.95s/it]
                                                       
{'loss': '0.2356', 'grad_norm': '8.681', 'learning_rate': '1.769e-05', 'epoch': '0.1332'}

 13%|█▎        | 666/5000 [8:58:59<84:12:30, 69.95s/it][2026-05-22 01:04:50,566] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-22 01:04:50,939] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 13%|█▎        | 667/5000 [8:59:50<77:20:24, 64.26s/it]
                                                       
{'loss': '0.4475', 'grad_norm': '11.95', 'learning_rate': '1.769e-05', 'epoch': '0.1334'}

 13%|█▎        | 667/5000 [8:59:50<77:20:24, 64.26s/it][2026-05-22 01:05:41,006] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 01:05:42,746] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 13%|█▎        | 668/5000 [9:00:40<72:12:14, 60.00s/it]
                                                       
{'loss': '0.4789', 'grad_norm': '7.358', 'learning_rate': '1.768e-05', 'epoch': '0.1336'}

 13%|█▎        | 668/5000 [9:00:40<72:12:14, 60.00s/it][2026-05-22 01:06:32,106] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])
[2026-05-22 01:06:32,891] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 225])

 13%|█▎        | 669/5000 [9:01:11<61:28:54, 51.10s/it]
                                                       
{'loss': '0.2329', 'grad_norm': '7.641', 'learning_rate': '1.768e-05', 'epoch': '0.1338'}

 13%|█▎        | 669/5000 [9:01:11<61:28:54, 51.10s/it][2026-05-22 01:07:02,690] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 273])
[2026-05-22 01:07:02,957] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 233])

 13%|█▎        | 670/5000 [9:01:44<55:08:45, 45.85s/it]
                                                       
{'loss': '0.06401', 'grad_norm': '5.624', 'learning_rate': '1.767e-05', 'epoch': '0.134'}

 13%|█▎        | 670/5000 [9:01:44<55:08:45, 45.85s/it][2026-05-22 01:07:40,393] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 339])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 958])
[2026-05-22 01:07:42,277] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 261])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 13%|█▎        | 671/5000 [9:02:33<56:04:46, 46.64s/it]
                                                       
{'loss': '1.24', 'grad_norm': '14.17', 'learning_rate': '1.767e-05', 'epoch': '0.1342'}

 13%|█▎        | 671/5000 [9:02:33<56:04:46, 46.64s/it][2026-05-22 01:08:29,024] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 285])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 01:08:30,937] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 298])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 13%|█▎        | 672/5000 [9:03:25<58:02:01, 48.27s/it]
                                                       
{'loss': '0.981', 'grad_norm': '11.21', 'learning_rate': '1.767e-05', 'epoch': '0.1344'}

 13%|█▎        | 672/5000 [9:03:25<58:02:01, 48.27s/it][2026-05-22 01:09:14,942] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 559])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 01:09:15,509] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 13%|█▎        | 673/5000 [9:04:15<58:35:53, 48.75s/it]
                                                       
{'loss': '0.7594', 'grad_norm': '9.62', 'learning_rate': '1.766e-05', 'epoch': '0.1346'}

 13%|█▎        | 673/5000 [9:04:15<58:35:53, 48.75s/it][2026-05-22 01:10:05,390] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 01:10:05,437] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 977])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 13%|█▎        | 674/5000 [9:05:06<59:36:00, 49.60s/it]
                                                       
{'loss': '0.7324', 'grad_norm': '9.037', 'learning_rate': '1.766e-05', 'epoch': '0.1348'}

 13%|█▎        | 674/5000 [9:05:06<59:36:00, 49.60s/it][2026-05-22 01:10:53,887] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 425])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 01:10:54,249] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 14%|█▎        | 675/5000 [9:05:54<58:53:18, 49.02s/it]
                                                       
{'loss': '0.363', 'grad_norm': '8.841', 'learning_rate': '1.765e-05', 'epoch': '0.135'}

 14%|█▎        | 675/5000 [9:05:54<58:53:18, 49.02s/it][2026-05-22 01:11:41,645] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 01:11:42,227] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 14%|█▎        | 676/5000 [9:06:43<58:42:48, 48.88s/it]
                                                       
{'loss': '0.1857', 'grad_norm': '5.025', 'learning_rate': '1.765e-05', 'epoch': '0.1352'}

 14%|█▎        | 676/5000 [9:06:43<58:42:48, 48.88s/it][2026-05-22 01:12:29,838] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 287])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 01:12:30,066] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 272])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 14%|█▎        | 677/5000 [9:07:30<58:01:26, 48.32s/it]
                                                       
{'loss': '0.4186', 'grad_norm': '10.67', 'learning_rate': '1.764e-05', 'epoch': '0.1354'}

 14%|█▎        | 677/5000 [9:07:30<58:01:26, 48.32s/it][2026-05-22 01:13:16,587] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 264])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 01:13:17,118] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 238])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 14%|█▎        | 678/5000 [9:08:14<56:45:10, 47.27s/it]
                                                       
{'loss': '0.427', 'grad_norm': '5.48', 'learning_rate': '1.764e-05', 'epoch': '0.1356'}

 14%|█▎        | 678/5000 [9:08:14<56:45:10, 47.27s/it][2026-05-22 01:14:03,757] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 102])
[2026-05-22 01:14:03,795] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])

 14%|█▎        | 679/5000 [9:09:22<64:09:24, 53.45s/it]
                                                       
{'loss': '0.9341', 'grad_norm': '13.38', 'learning_rate': '1.764e-05', 'epoch': '0.1358'}

 14%|█▎        | 679/5000 [9:09:22<64:09:24, 53.45s/it][2026-05-22 01:15:11,611] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])
[2026-05-22 01:15:12,317] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 75])

 14%|█▎        | 680/5000 [9:10:31<69:38:36, 58.04s/it]
                                                       
{'loss': '0.8007', 'grad_norm': '11.61', 'learning_rate': '1.763e-05', 'epoch': '0.136'}

 14%|█▎        | 680/5000 [9:10:31<69:38:36, 58.04s/it][2026-05-22 01:16:28,635] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 198])
[2026-05-22 01:16:29,390] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 315])

 14%|█▎        | 681/5000 [9:11:02<59:54:21, 49.93s/it]
                                                       
{'loss': '0.6441', 'grad_norm': '21.15', 'learning_rate': '1.763e-05', 'epoch': '0.1362'}

 14%|█▎        | 681/5000 [9:11:02<59:54:21, 49.93s/it][2026-05-22 01:16:58,200] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 183])
[2026-05-22 01:16:59,364] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])

 14%|█▎        | 682/5000 [9:11:33<53:12:34, 44.36s/it]
                                                       
{'loss': '0.604', 'grad_norm': '18.47', 'learning_rate': '1.762e-05', 'epoch': '0.1364'}

 14%|█▎        | 682/5000 [9:11:33<53:12:34, 44.36s/it][2026-05-22 01:17:31,156] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 288])
[2026-05-22 01:17:33,291] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])

 14%|█▎        | 683/5000 [9:12:06<48:57:37, 40.83s/it]
                                                       
{'loss': '2.541', 'grad_norm': '9.197', 'learning_rate': '1.762e-05', 'epoch': '0.1366'}

 14%|█▎        | 683/5000 [9:12:06<48:57:37, 40.83s/it][2026-05-22 01:18:04,341] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 265])
[2026-05-22 01:18:07,637] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])

 14%|█▎        | 684/5000 [9:12:37<45:22:34, 37.85s/it]
                                                       
{'loss': '2.437', 'grad_norm': '8.945', 'learning_rate': '1.762e-05', 'epoch': '0.1368'}

 14%|█▎        | 684/5000 [9:12:37<45:22:34, 37.85s/it][2026-05-22 01:18:24,033] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])
[2026-05-22 01:18:24,861] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 247])

 14%|█▎        | 685/5000 [9:13:24<48:46:46, 40.70s/it]
                                                       
{'loss': '2.222', 'grad_norm': '33.25', 'learning_rate': '1.761e-05', 'epoch': '0.137'}

 14%|█▎        | 685/5000 [9:13:24<48:46:46, 40.70s/it][2026-05-22 01:19:11,625] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 953])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 238])
[2026-05-22 01:19:12,105] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])

 14%|█▎        | 686/5000 [9:14:11<50:50:21, 42.42s/it]
                                                       
{'loss': '2.226', 'grad_norm': '27.15', 'learning_rate': '1.761e-05', 'epoch': '0.1372'}

 14%|█▎        | 686/5000 [9:14:11<50:50:21, 42.42s/it][2026-05-22 01:19:58,022] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 284])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 01:19:58,842] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 296])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 14%|█▎        | 687/5000 [9:14:58<52:31:19, 43.84s/it]
                                                       
{'loss': '0.4249', 'grad_norm': '10.97', 'learning_rate': '1.76e-05', 'epoch': '0.1374'}

 14%|█▎        | 687/5000 [9:14:58<52:31:19, 43.84s/it][2026-05-22 01:20:45,838] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 287])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 01:20:46,607] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 352])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 14%|█▍        | 688/5000 [9:15:45<53:33:31, 44.72s/it]
                                                       
{'loss': '0.3195', 'grad_norm': '10.67', 'learning_rate': '1.76e-05', 'epoch': '0.1376'}

 14%|█▍        | 688/5000 [9:15:45<53:33:31, 44.72s/it][2026-05-22 01:21:35,681] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 01:21:36,745] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 14%|█▍        | 689/5000 [9:16:36<55:49:55, 46.62s/it]
                                                       
{'loss': '0.4475', 'grad_norm': '10.7', 'learning_rate': '1.76e-05', 'epoch': '0.1378'}

 14%|█▍        | 689/5000 [9:16:36<55:49:55, 46.62s/it][2026-05-22 01:22:27,782] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 01:22:28,345] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 14%|█▍        | 690/5000 [9:17:28<57:41:23, 48.19s/it]
                                                       
{'loss': '0.2304', 'grad_norm': '8.484', 'learning_rate': '1.759e-05', 'epoch': '0.138'}

 14%|█▍        | 690/5000 [9:17:28<57:41:23, 48.19s/it][2026-05-22 01:23:23,817] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 230])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 01:23:26,006] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 352])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 14%|█▍        | 691/5000 [9:18:13<56:36:35, 47.30s/it]
                                                       
{'loss': '0.6706', 'grad_norm': '15.87', 'learning_rate': '1.759e-05', 'epoch': '0.1382'}

 14%|█▍        | 691/5000 [9:18:13<56:36:35, 47.30s/it][2026-05-22 01:24:10,267] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 254])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-22 01:24:10,822] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 252])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 14%|█▍        | 692/5000 [9:18:58<55:53:26, 46.71s/it]
                                                       
{'loss': '0.4381', 'grad_norm': '12.62', 'learning_rate': '1.758e-05', 'epoch': '0.1384'}

 14%|█▍        | 692/5000 [9:18:58<55:53:26, 46.71s/it][2026-05-22 01:24:48,085] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 929])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 971])
[2026-05-22 01:24:48,832] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 560])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

 14%|█▍        | 693/5000 [9:19:28<49:47:49, 41.62s/it]
                                                       
{'loss': '0.3794', 'grad_norm': '9.628', 'learning_rate': '1.758e-05', 'epoch': '0.1386'}

 14%|█▍        | 693/5000 [9:19:28<49:47:49, 41.62s/it][2026-05-22 01:25:17,700] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 969])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 01:25:18,247] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 14%|█▍        | 694/5000 [9:20:00<46:27:43, 38.84s/it]
                                                       
{'loss': '0.04529', 'grad_norm': '4.624', 'learning_rate': '1.758e-05', 'epoch': '0.1388'}

 14%|█▍        | 694/5000 [9:20:00<46:27:43, 38.84s/it][2026-05-22 01:25:59,108] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 95])
[2026-05-22 01:26:03,293] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 154])

 14%|█▍        | 695/5000 [9:20:52<50:55:55, 42.59s/it]
                                                       
{'loss': '0.3755', 'grad_norm': '9.119', 'learning_rate': '1.757e-05', 'epoch': '0.139'}

 14%|█▍        | 695/5000 [9:20:52<50:55:55, 42.59s/it][2026-05-22 01:26:50,316] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 79])
[2026-05-22 01:26:52,948] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])

 14%|█▍        | 696/5000 [9:21:43<54:03:36, 45.22s/it]
                                                       
{'loss': '0.341', 'grad_norm': '9.939', 'learning_rate': '1.757e-05', 'epoch': '0.1392'}

 14%|█▍        | 696/5000 [9:21:43<54:03:36, 45.22s/it][2026-05-22 01:27:30,280] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 417])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 56])
[2026-05-22 01:27:30,813] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])

 14%|█▍        | 697/5000 [9:22:30<54:51:49, 45.90s/it]
                                                       
{'loss': '0.7067', 'grad_norm': '12.63', 'learning_rate': '1.756e-05', 'epoch': '0.1394'}

 14%|█▍        | 697/5000 [9:22:30<54:51:49, 45.90s/it][2026-05-22 01:28:18,248] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 175])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])
[2026-05-22 01:28:19,392] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 479])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])

 14%|█▍        | 698/5000 [9:23:17<55:12:32, 46.20s/it]
                                                       
{'loss': '0.7674', 'grad_norm': '12', 'learning_rate': '1.756e-05', 'epoch': '0.1396'}

 14%|█▍        | 698/5000 [9:23:17<55:12:32, 46.20s/it][2026-05-22 01:29:07,213] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 286])
[2026-05-22 01:29:07,263] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 272])

 14%|█▍        | 699/5000 [9:24:22<61:58:35, 51.88s/it]
                                                       
{'loss': '1.082', 'grad_norm': '13.59', 'learning_rate': '1.756e-05', 'epoch': '0.1398'}

 14%|█▍        | 699/5000 [9:24:22<61:58:35, 51.88s/it][2026-05-22 01:30:12,155] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 299])
[2026-05-22 01:30:12,934] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])

 14%|█▍        | 700/5000 [9:25:31<67:57:06, 56.89s/it]
                                                       
{'loss': '0.5651', 'grad_norm': '14.13', 'learning_rate': '1.755e-05', 'epoch': '0.14'}

 14%|█▍        | 700/5000 [9:25:31<67:57:06, 56.89s/it][2026-05-22 01:31:29,617] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 01:31:31,224] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 14%|█▍        | 701/5000 [9:26:19<64:47:27, 54.26s/it]
                                                       
{'loss': '0.258', 'grad_norm': '10.08', 'learning_rate': '1.755e-05', 'epoch': '0.1402'}

 14%|█▍        | 701/5000 [9:26:19<64:47:27, 54.26s/it][2026-05-22 01:32:20,797] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 01:32:22,844] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 14%|█▍        | 702/5000 [9:27:09<63:06:24, 52.86s/it]
                                                       
{'loss': '0.1645', 'grad_norm': '8.026', 'learning_rate': '1.754e-05', 'epoch': '0.1404'}

 14%|█▍        | 702/5000 [9:27:09<63:06:24, 52.86s/it][2026-05-22 01:33:02,422] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 01:33:03,589] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 14%|█▍        | 703/5000 [9:27:36<53:58:51, 45.22s/it]
                                                       
{'loss': '0.1751', 'grad_norm': '9.83', 'learning_rate': '1.754e-05', 'epoch': '0.1406'}

 14%|█▍        | 703/5000 [9:27:36<53:58:51, 45.22s/it][2026-05-22 01:33:29,556] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-22 01:33:30,182] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])

 14%|█▍        | 704/5000 [9:28:07<48:42:04, 40.81s/it]
                                                       
{'loss': '0.1312', 'grad_norm': '8.216', 'learning_rate': '1.753e-05', 'epoch': '0.1408'}

 14%|█▍        | 704/5000 [9:28:07<48:42:04, 40.81s/it][2026-05-22 01:34:05,155] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 738])
[2026-05-22 01:34:06,288] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 14%|█▍        | 705/5000 [9:28:55<51:13:05, 42.93s/it]
                                                       
{'loss': '0.3804', 'grad_norm': '8.542', 'learning_rate': '1.753e-05', 'epoch': '0.141'}

 14%|█▍        | 705/5000 [9:28:55<51:13:05, 42.93s/it][2026-05-22 01:34:54,199] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])
[2026-05-22 01:34:55,106] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 14%|█▍        | 706/5000 [9:29:45<54:04:26, 45.33s/it]
                                                       
{'loss': '0.1054', 'grad_norm': '5.608', 'learning_rate': '1.753e-05', 'epoch': '0.1412'}

 14%|█▍        | 706/5000 [9:29:45<54:04:26, 45.33s/it][2026-05-22 01:35:41,290] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 61])
[2026-05-22 01:35:46,145] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 73])

 14%|█▍        | 707/5000 [9:30:54<62:22:23, 52.30s/it]
                                                       
{'loss': '1.12', 'grad_norm': '10.96', 'learning_rate': '1.752e-05', 'epoch': '0.1414'}

 14%|█▍        | 707/5000 [9:30:54<62:22:23, 52.30s/it][2026-05-22 01:36:46,472] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 35])
[2026-05-22 01:36:48,434] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 148])

 14%|█▍        | 708/5000 [9:32:04<68:37:22, 57.56s/it]
                                                       
{'loss': '0.4606', 'grad_norm': '12.32', 'learning_rate': '1.752e-05', 'epoch': '0.1416'}

 14%|█▍        | 708/5000 [9:32:04<68:37:22, 57.56s/it][2026-05-22 01:37:53,801] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 205])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 01:37:54,131] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 117])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 14%|█▍        | 709/5000 [9:33:14<73:06:50, 61.34s/it]
                                                       
{'loss': '1.182', 'grad_norm': '8.906', 'learning_rate': '1.751e-05', 'epoch': '0.1418'}

 14%|█▍        | 709/5000 [9:33:14<73:06:50, 61.34s/it][2026-05-22 01:39:03,456] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 125])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 01:39:04,947] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 197])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

 14%|█▍        | 710/5000 [9:34:24<76:21:04, 64.07s/it]
                                                       
{'loss': '0.8878', 'grad_norm': '11.27', 'learning_rate': '1.751e-05', 'epoch': '0.142'}

 14%|█▍        | 710/5000 [9:34:24<76:21:04, 64.07s/it][2026-05-22 01:40:12,059] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 188])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-22 01:40:12,816] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 189])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 183])

 14%|█▍        | 711/5000 [9:35:31<77:12:38, 64.81s/it]
                                                       
{'loss': '1.075', 'grad_norm': '14.09', 'learning_rate': '1.751e-05', 'epoch': '0.1422'}

 14%|█▍        | 711/5000 [9:35:31<77:12:38, 64.81s/it][2026-05-22 01:41:19,078] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 173])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-22 01:41:19,288] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 235])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 14%|█▍        | 712/5000 [9:36:41<78:56:02, 66.27s/it]
                                                       
{'loss': '0.9497', 'grad_norm': '14.56', 'learning_rate': '1.75e-05', 'epoch': '0.1424'}

 14%|█▍        | 712/5000 [9:36:41<78:56:02, 66.27s/it][2026-05-22 01:42:27,846] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 284])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 01:42:28,932] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 294])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 14%|█▍        | 713/5000 [9:37:24<70:45:55, 59.43s/it]
                                                       
{'loss': '0.6176', 'grad_norm': '16.51', 'learning_rate': '1.75e-05', 'epoch': '0.1426'}

 14%|█▍        | 713/5000 [9:37:24<70:45:55, 59.43s/it][2026-05-22 01:43:11,906] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 299])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 01:43:12,249] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 315])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 14%|█▍        | 714/5000 [9:38:08<65:07:27, 54.70s/it]
                                                       
{'loss': '0.5802', 'grad_norm': '15.55', 'learning_rate': '1.749e-05', 'epoch': '0.1428'}

 14%|█▍        | 714/5000 [9:38:08<65:07:27, 54.70s/it][2026-05-22 01:43:57,280] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 428])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 01:43:58,194] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 14%|█▍        | 715/5000 [9:38:56<62:50:40, 52.80s/it]
                                                       
{'loss': '0.5384', 'grad_norm': '16.16', 'learning_rate': '1.749e-05', 'epoch': '0.143'}

 14%|█▍        | 715/5000 [9:38:56<62:50:40, 52.80s/it][2026-05-22 01:44:44,124] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 202])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 01:44:46,456] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 507])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

 14%|█▍        | 716/5000 [9:39:44<61:08:49, 51.38s/it]
                                                       
{'loss': '1.126', 'grad_norm': '13.48', 'learning_rate': '1.749e-05', 'epoch': '0.1432'}

 14%|█▍        | 716/5000 [9:39:44<61:08:49, 51.38s/it][2026-05-22 01:45:34,471] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 285])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 01:45:35,025] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 246])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 14%|█▍        | 717/5000 [9:40:10<51:49:46, 43.56s/it]
                                                       
{'loss': '0.4544', 'grad_norm': '11.72', 'learning_rate': '1.748e-05', 'epoch': '0.1434'}

 14%|█▍        | 717/5000 [9:40:10<51:49:46, 43.56s/it][2026-05-22 01:45:58,862] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 352])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 01:46:00,098] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 266])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 14%|█▍        | 718/5000 [9:40:35<45:24:44, 38.18s/it]
                                                       
{'loss': '0.4539', 'grad_norm': '11.29', 'learning_rate': '1.748e-05', 'epoch': '0.1436'}

 14%|█▍        | 718/5000 [9:40:35<45:24:44, 38.18s/it][2026-05-22 01:46:22,911] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 62])
[2026-05-22 01:46:22,994] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 64])

 14%|█▍        | 719/5000 [9:41:03<41:36:21, 34.99s/it]
                                                       
{'loss': '0.2009', 'grad_norm': '6.764', 'learning_rate': '1.747e-05', 'epoch': '0.1438'}

 14%|█▍        | 719/5000 [9:41:03<41:36:21, 34.99s/it][2026-05-22 01:46:49,826] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 244])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])
[2026-05-22 01:46:51,473] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 70])

 14%|█▍        | 720/5000 [9:41:33<39:45:51, 33.45s/it]
                                                       
{'loss': '0.08834', 'grad_norm': '5.741', 'learning_rate': '1.747e-05', 'epoch': '0.144'}

 14%|█▍        | 720/5000 [9:41:33<39:45:51, 33.45s/it][2026-05-22 01:47:23,540] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 122])
[2026-05-22 01:47:23,567] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])

 14%|█▍        | 721/5000 [9:42:22<45:35:02, 38.35s/it]
                                                       
{'loss': '0.248', 'grad_norm': '9.331', 'learning_rate': '1.747e-05', 'epoch': '0.1442'}

 14%|█▍        | 721/5000 [9:42:22<45:35:02, 38.35s/it][2026-05-22 01:48:13,450] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 168])
[2026-05-22 01:48:14,134] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 147])

 14%|█▍        | 722/5000 [9:43:16<50:59:08, 42.91s/it]
                                                       
{'loss': '0.5163', 'grad_norm': '7.406', 'learning_rate': '1.746e-05', 'epoch': '0.1444'}

 14%|█▍        | 722/5000 [9:43:16<50:59:08, 42.91s/it][2026-05-22 01:49:37,477] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 155])
[2026-05-22 01:49:42,008] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])

 14%|█▍        | 723/5000 [9:44:04<52:45:59, 44.41s/it]
                                                       
{'loss': '0.5838', 'grad_norm': '14.72', 'learning_rate': '1.746e-05', 'epoch': '0.1446'}

 14%|█▍        | 723/5000 [9:44:04<52:45:59, 44.41s/it][2026-05-22 01:50:25,726] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 168])
[2026-05-22 01:50:30,197] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 91])

 14%|█▍        | 724/5000 [9:44:54<54:47:59, 46.14s/it]
                                                       
{'loss': '0.3793', 'grad_norm': '10', 'learning_rate': '1.745e-05', 'epoch': '0.1448'}

 14%|█▍        | 724/5000 [9:44:54<54:47:59, 46.14s/it][2026-05-22 01:50:41,630] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 267])
[2026-05-22 01:50:42,256] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 467])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 294])

 14%|█▍        | 725/5000 [9:45:22<48:23:01, 40.74s/it]
                                                       
{'loss': '0.3258', 'grad_norm': '10.42', 'learning_rate': '1.745e-05', 'epoch': '0.145'}

 14%|█▍        | 725/5000 [9:45:22<48:23:01, 40.74s/it][2026-05-22 01:51:10,173] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 302])
[2026-05-22 01:51:10,821] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 426])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])

 15%|█▍        | 726/5000 [9:45:52<44:27:52, 37.45s/it]
                                                       
{'loss': '0.3101', 'grad_norm': '9.635', 'learning_rate': '1.744e-05', 'epoch': '0.1452'}

 15%|█▍        | 726/5000 [9:45:52<44:27:52, 37.45s/it][2026-05-22 01:51:53,724] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 383])
[2026-05-22 01:51:54,313] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 121])

 15%|█▍        | 727/5000 [9:46:46<50:32:30, 42.58s/it]
                                                       
{'loss': '0.644', 'grad_norm': '13.94', 'learning_rate': '1.744e-05', 'epoch': '0.1454'}

 15%|█▍        | 727/5000 [9:46:46<50:32:30, 42.58s/it][2026-05-22 01:52:48,019] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 224])
[2026-05-22 01:52:50,967] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 206])

 15%|█▍        | 728/5000 [9:47:39<54:14:44, 45.71s/it]
                                                       
{'loss': '0.5394', 'grad_norm': '9.162', 'learning_rate': '1.744e-05', 'epoch': '0.1456'}

 15%|█▍        | 728/5000 [9:47:39<54:14:44, 45.71s/it][2026-05-22 01:53:37,351] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 01:53:37,912] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])

 15%|█▍        | 729/5000 [9:48:29<55:34:12, 46.84s/it]
                                                       
{'loss': '0.4912', 'grad_norm': '13.08', 'learning_rate': '1.743e-05', 'epoch': '0.1458'}

 15%|█▍        | 729/5000 [9:48:29<55:34:12, 46.84s/it][2026-05-22 01:54:23,317] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 01:54:26,538] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 15%|█▍        | 730/5000 [9:49:19<56:31:27, 47.66s/it]
                                                       
{'loss': '0.3858', 'grad_norm': '11.14', 'learning_rate': '1.743e-05', 'epoch': '0.146'}

 15%|█▍        | 730/5000 [9:49:19<56:31:27, 47.66s/it][2026-05-22 01:55:05,867] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])
[2026-05-22 01:55:05,888] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 15%|█▍        | 731/5000 [9:49:48<49:55:24, 42.10s/it]
                                                       
{'loss': '0.4448', 'grad_norm': '13.23', 'learning_rate': '1.742e-05', 'epoch': '0.1462'}

 15%|█▍        | 731/5000 [9:49:48<49:55:24, 42.10s/it][2026-05-22 01:55:34,362] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 168])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-22 01:55:35,472] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 426])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 15%|█▍        | 732/5000 [9:50:16<45:04:38, 38.02s/it]
                                                       
{'loss': '0.2457', 'grad_norm': '11', 'learning_rate': '1.742e-05', 'epoch': '0.1464'}

 15%|█▍        | 732/5000 [9:50:16<45:04:38, 38.02s/it][2026-05-22 01:56:03,558] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])
[2026-05-22 01:56:03,606] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 668])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 56])

 15%|█▍        | 733/5000 [9:51:08<49:49:17, 42.03s/it]
                                                       
{'loss': '0.5006', 'grad_norm': '15.3', 'learning_rate': '1.742e-05', 'epoch': '0.1466'}

 15%|█▍        | 733/5000 [9:51:08<49:49:17, 42.03s/it][2026-05-22 01:56:55,315] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 63])
[2026-05-22 01:56:55,550] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 58])

 15%|█▍        | 734/5000 [9:51:57<52:32:16, 44.34s/it]
                                                       
{'loss': '0.3849', 'grad_norm': '10.69', 'learning_rate': '1.741e-05', 'epoch': '0.1468'}

 15%|█▍        | 734/5000 [9:51:57<52:32:16, 44.34s/it][2026-05-22 01:57:46,439] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1023])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 73])
[2026-05-22 01:57:46,450] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])

 15%|█▍        | 735/5000 [9:52:47<54:31:38, 46.03s/it]
                                                       
{'loss': '0.2797', 'grad_norm': '9.324', 'learning_rate': '1.741e-05', 'epoch': '0.147'}

 15%|█▍        | 735/5000 [9:52:47<54:31:38, 46.03s/it][2026-05-22 01:58:36,035] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 473])
[2026-05-22 01:58:36,593] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 96])

 15%|█▍        | 736/5000 [9:53:37<55:52:28, 47.17s/it]
                                                       
{'loss': '0.4265', 'grad_norm': '11.01', 'learning_rate': '1.74e-05', 'epoch': '0.1472'}

 15%|█▍        | 736/5000 [9:53:37<55:52:28, 47.17s/it][2026-05-22 01:59:37,512] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])
[2026-05-22 01:59:40,348] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])

 15%|█▍        | 737/5000 [9:54:05<49:07:00, 41.48s/it]
                                                       
{'loss': '0.4778', 'grad_norm': '16.15', 'learning_rate': '1.74e-05', 'epoch': '0.1474'}

 15%|█▍        | 737/5000 [9:54:05<49:07:00, 41.48s/it][2026-05-22 02:00:04,557] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])
[2026-05-22 02:00:06,297] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 15%|█▍        | 738/5000 [9:54:37<45:44:33, 38.64s/it]
                                                       
{'loss': '0.5122', 'grad_norm': '16.58', 'learning_rate': '1.74e-05', 'epoch': '0.1476'}

 15%|█▍        | 738/5000 [9:54:37<45:44:33, 38.64s/it][2026-05-22 02:00:30,868] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 116])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 02:00:32,161] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 105])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 15%|█▍        | 739/5000 [9:55:12<44:22:11, 37.49s/it]
                                                       
{'loss': '3.657', 'grad_norm': '6.308', 'learning_rate': '1.739e-05', 'epoch': '0.1478'}

 15%|█▍        | 739/5000 [9:55:12<44:22:11, 37.49s/it][2026-05-22 02:01:05,733] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 106])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 02:01:06,033] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 100])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])

 15%|█▍        | 740/5000 [9:55:46<43:06:22, 36.43s/it]
                                                       
{'loss': '3.65', 'grad_norm': '6.102', 'learning_rate': '1.739e-05', 'epoch': '0.148'}

 15%|█▍        | 740/5000 [9:55:46<43:06:22, 36.43s/it][2026-05-22 02:01:35,035] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 02:01:37,964] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])

 15%|█▍        | 741/5000 [9:56:33<46:58:17, 39.70s/it]
                                                       
{'loss': '0.4169', 'grad_norm': '10.99', 'learning_rate': '1.738e-05', 'epoch': '0.1482'}

 15%|█▍        | 741/5000 [9:56:33<46:58:17, 39.70s/it][2026-05-22 02:02:22,743] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-22 02:02:24,515] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 15%|█▍        | 742/5000 [9:57:24<50:57:27, 43.08s/it]
                                                       
{'loss': '0.4147', 'grad_norm': '10.66', 'learning_rate': '1.738e-05', 'epoch': '0.1484'}

 15%|█▍        | 742/5000 [9:57:24<50:57:27, 43.08s/it][2026-05-22 02:03:11,570] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 185])
[2026-05-22 02:03:11,944] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 282])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 15%|█▍        | 743/5000 [9:58:34<60:30:19, 51.17s/it]
                                                       
{'loss': '1.389', 'grad_norm': '16.19', 'learning_rate': '1.738e-05', 'epoch': '0.1486'}

 15%|█▍        | 743/5000 [9:58:34<60:30:19, 51.17s/it][2026-05-22 02:04:21,257] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 317])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 02:04:21,997] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 254])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 15%|█▍        | 744/5000 [9:59:43<66:49:44, 56.53s/it]
                                                       
{'loss': '0.9457', 'grad_norm': '11.51', 'learning_rate': '1.737e-05', 'epoch': '0.1488'}

 15%|█▍        | 744/5000 [9:59:43<66:49:44, 56.53s/it][2026-05-22 02:05:33,654] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 252])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])
[2026-05-22 02:05:34,025] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 252])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 15%|█▍        | 745/5000 [10:00:32<64:04:46, 54.22s/it]
                                                        
{'loss': '1.472', 'grad_norm': '11.56', 'learning_rate': '1.737e-05', 'epoch': '0.149'}

 15%|█▍        | 745/5000 [10:00:32<64:04:46, 54.22s/it][2026-05-22 02:06:21,506] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 279])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 02:06:22,090] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 303])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])

 15%|█▍        | 746/5000 [10:01:21<62:04:40, 52.53s/it]
                                                        
{'loss': '1.081', 'grad_norm': '12.37', 'learning_rate': '1.736e-05', 'epoch': '0.1492'}

 15%|█▍        | 746/5000 [10:01:21<62:04:40, 52.53s/it][2026-05-22 02:07:22,467] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 93])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 02:07:26,347] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 102])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])

 15%|█▍        | 747/5000 [10:02:13<61:51:35, 52.36s/it]
                                                        
{'loss': '1.182', 'grad_norm': '23.45', 'learning_rate': '1.736e-05', 'epoch': '0.1494'}

 15%|█▍        | 747/5000 [10:02:13<61:51:35, 52.36s/it][2026-05-22 02:08:09,888] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 120])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 02:08:12,797] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 83])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

 15%|█▍        | 748/5000 [10:03:04<61:35:22, 52.15s/it]
                                                        
{'loss': '0.9267', 'grad_norm': '21.95', 'learning_rate': '1.736e-05', 'epoch': '0.1496'}

 15%|█▍        | 748/5000 [10:03:04<61:35:22, 52.15s/it][2026-05-22 02:08:54,865] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 02:08:55,748] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 486])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 15%|█▍        | 749/5000 [10:03:33<53:16:16, 45.11s/it]
                                                        
{'loss': '0.4367', 'grad_norm': '10.65', 'learning_rate': '1.735e-05', 'epoch': '0.1498'}

 15%|█▍        | 749/5000 [10:03:33<53:16:16, 45.11s/it][2026-05-22 02:09:23,424] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 473])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 02:09:23,984] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 446])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 15%|█▌        | 750/5000 [10:04:03<47:57:46, 40.63s/it]
                                                        
{'loss': '0.3914', 'grad_norm': '10.09', 'learning_rate': '1.735e-05', 'epoch': '0.15'}

 15%|█▌        | 750/5000 [10:04:03<47:57:46, 40.63s/it][2026-05-22 02:09:48,455] INFO [src.utils.basic_utils:19] Saving model to /workspace/VLM2Vec/outputs/Qwen3.5-08b-continued-mixed-alltasks/checkpoint-750
[2026-05-22 02:09:48,619] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-22 02:09:48,629] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-22 02:09:54,162] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1020])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 134])
[2026-05-22 02:09:55,329] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 107])

 15%|█▌        | 751/5000 [10:04:53<51:04:08, 43.27s/it]
                                                        
{'loss': '0.6838', 'grad_norm': '13.36', 'learning_rate': '1.734e-05', 'epoch': '0.1502'}

 15%|█▌        | 751/5000 [10:04:53<51:04:08, 43.27s/it][2026-05-22 02:10:43,087] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 94])
[2026-05-22 02:10:44,026] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 131])

 15%|█▌        | 752/5000 [10:05:41<52:52:00, 44.80s/it]
                                                        
{'loss': '1.094', 'grad_norm': '8.37', 'learning_rate': '1.734e-05', 'epoch': '0.1504'}

 15%|█▌        | 752/5000 [10:05:41<52:52:00, 44.80s/it][2026-05-22 02:11:40,402] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 02:11:41,075] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 15%|█▌        | 753/5000 [10:06:32<54:58:34, 46.60s/it]
                                                        
{'loss': '0.5633', 'grad_norm': '14.44', 'learning_rate': '1.733e-05', 'epoch': '0.1506'}

 15%|█▌        | 753/5000 [10:06:32<54:58:34, 46.60s/it][2026-05-22 02:12:29,055] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 02:12:31,575] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 15%|█▌        | 754/5000 [10:07:22<56:10:49, 47.63s/it]
                                                        
{'loss': '1.305', 'grad_norm': '10.48', 'learning_rate': '1.733e-05', 'epoch': '0.1508'}

 15%|█▌        | 754/5000 [10:07:22<56:10:49, 47.63s/it][2026-05-22 02:13:23,857] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])
[2026-05-22 02:13:24,755] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 15%|█▌        | 755/5000 [10:08:15<58:04:48, 49.26s/it]
                                                        
{'loss': '0.6421', 'grad_norm': '13.89', 'learning_rate': '1.733e-05', 'epoch': '0.151'}

 15%|█▌        | 755/5000 [10:08:15<58:04:48, 49.26s/it][2026-05-22 02:14:16,781] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 02:14:16,794] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 15%|█▌        | 756/5000 [10:09:08<59:24:19, 50.39s/it]
                                                        
{'loss': '0.4527', 'grad_norm': '7.414', 'learning_rate': '1.732e-05', 'epoch': '0.1512'}

 15%|█▌        | 756/5000 [10:09:08<59:24:19, 50.39s/it][2026-05-22 02:14:57,983] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 291])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 02:15:00,112] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 271])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])

 15%|█▌        | 757/5000 [10:09:59<59:27:44, 50.45s/it]
                                                        
{'loss': '2.967', 'grad_norm': '40.82', 'learning_rate': '1.732e-05', 'epoch': '0.1514'}

 15%|█▌        | 757/5000 [10:09:59<59:27:44, 50.45s/it][2026-05-22 02:15:47,770] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 347])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 02:15:48,052] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 215])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 15%|█▌        | 758/5000 [10:10:46<58:27:10, 49.61s/it]
                                                        
{'loss': '2.804', 'grad_norm': '19.51', 'learning_rate': '1.731e-05', 'epoch': '0.1516'}

 15%|█▌        | 758/5000 [10:10:46<58:27:10, 49.61s/it][2026-05-22 02:16:35,470] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 262])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 278])
[2026-05-22 02:16:36,214] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 15%|█▌        | 759/5000 [10:11:55<65:16:25, 55.41s/it]
                                                        
{'loss': '1.214', 'grad_norm': '11.16', 'learning_rate': '1.731e-05', 'epoch': '0.1518'}

 15%|█▌        | 759/5000 [10:11:55<65:16:25, 55.41s/it][2026-05-22 02:17:43,950] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 02:17:44,574] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 15%|█▌        | 760/5000 [10:13:03<69:36:38, 59.10s/it]
                                                        
{'loss': '1.084', 'grad_norm': '10.13', 'learning_rate': '1.731e-05', 'epoch': '0.152'}

 15%|█▌        | 760/5000 [10:13:03<69:36:38, 59.10s/it][2026-05-22 02:18:51,879] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 248])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 02:18:52,531] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 240])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])

 15%|█▌        | 761/5000 [10:13:55<66:59:56, 56.90s/it]
                                                        
{'loss': '0.5194', 'grad_norm': '21.12', 'learning_rate': '1.73e-05', 'epoch': '0.1522'}

 15%|█▌        | 761/5000 [10:13:55<66:59:56, 56.90s/it][2026-05-22 02:19:43,670] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 278])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 02:19:44,809] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

 15%|█▌        | 762/5000 [10:14:48<65:46:59, 55.88s/it]
                                                        
{'loss': '0.3336', 'grad_norm': '7.548', 'learning_rate': '1.73e-05', 'epoch': '0.1524'}

 15%|█▌        | 762/5000 [10:14:48<65:46:59, 55.88s/it][2026-05-22 02:20:38,158] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 02:20:38,281] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 964])

 15%|█▌        | 763/5000 [10:15:36<63:02:46, 53.57s/it]
                                                        
{'loss': '0.9451', 'grad_norm': '10.1', 'learning_rate': '1.729e-05', 'epoch': '0.1526'}

 15%|█▌        | 763/5000 [10:15:36<63:02:46, 53.57s/it][2026-05-22 02:21:25,530] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 502])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 975])
[2026-05-22 02:21:26,604] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 183])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 15%|█▌        | 764/5000 [10:16:25<61:22:55, 52.17s/it]
                                                        
{'loss': '1.062', 'grad_norm': '8.723', 'learning_rate': '1.729e-05', 'epoch': '0.1528'}

 15%|█▌        | 764/5000 [10:16:25<61:22:55, 52.17s/it][2026-05-22 02:22:14,693] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 317])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 02:22:16,271] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 302])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

 15%|█▌        | 765/5000 [10:16:59<54:51:55, 46.64s/it]
                                                        
{'loss': '0.4902', 'grad_norm': '11.09', 'learning_rate': '1.729e-05', 'epoch': '0.153'}

 15%|█▌        | 765/5000 [10:16:59<54:51:55, 46.64s/it][2026-05-22 02:22:48,975] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 278])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-22 02:22:49,790] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 288])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 15%|█▌        | 766/5000 [10:17:32<49:58:29, 42.49s/it]
                                                        
{'loss': '0.07966', 'grad_norm': '7.985', 'learning_rate': '1.728e-05', 'epoch': '0.1532'}

 15%|█▌        | 766/5000 [10:17:32<49:58:29, 42.49s/it][2026-05-22 02:23:32,590] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-22 02:23:38,851] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 15%|█▌        | 767/5000 [10:18:20<52:08:30, 44.34s/it]
                                                        
{'loss': '1.271', 'grad_norm': '9.461', 'learning_rate': '1.728e-05', 'epoch': '0.1534'}

 15%|█▌        | 767/5000 [10:18:21<52:08:30, 44.34s/it][2026-05-22 02:24:21,582] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 02:24:22,267] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 15%|█▌        | 768/5000 [10:19:11<54:27:53, 46.33s/it]
                                                        
{'loss': '0.8111', 'grad_norm': '7.521', 'learning_rate': '1.727e-05', 'epoch': '0.1536'}

 15%|█▌        | 768/5000 [10:19:11<54:27:53, 46.33s/it][2026-05-22 02:25:02,837] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 102])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 02:25:03,093] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 115])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 15%|█▌        | 769/5000 [10:19:59<54:48:21, 46.63s/it]
                                                        
{'loss': '0.5826', 'grad_norm': '7.449', 'learning_rate': '1.727e-05', 'epoch': '0.1538'}

 15%|█▌        | 769/5000 [10:19:59<54:48:21, 46.63s/it][2026-05-22 02:25:48,700] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 104])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 02:25:50,578] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 114])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 15%|█▌        | 770/5000 [10:20:51<56:45:48, 48.31s/it]
                                                        
{'loss': '0.2265', 'grad_norm': '7.702', 'learning_rate': '1.727e-05', 'epoch': '0.154'}

 15%|█▌        | 770/5000 [10:20:51<56:45:48, 48.31s/it][2026-05-22 02:26:45,767] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 124])
[2026-05-22 02:26:46,158] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 145])

 15%|█▌        | 771/5000 [10:21:44<58:33:20, 49.85s/it]
                                                        
{'loss': '0.8635', 'grad_norm': '9.337', 'learning_rate': '1.726e-05', 'epoch': '0.1542'}

 15%|█▌        | 771/5000 [10:21:44<58:33:20, 49.85s/it][2026-05-22 02:27:39,234] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 108])
[2026-05-22 02:27:40,217] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 89])

 15%|█▌        | 772/5000 [10:22:41<60:47:05, 51.76s/it]
                                                        
{'loss': '0.2234', 'grad_norm': '8.136', 'learning_rate': '1.726e-05', 'epoch': '0.1544'}

 15%|█▌        | 772/5000 [10:22:41<60:47:05, 51.76s/it][2026-05-22 02:28:53,435] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 185])
[2026-05-22 02:28:55,413] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 108])

 15%|█▌        | 773/5000 [10:23:52<67:35:58, 57.57s/it]
                                                        
{'loss': '0.7765', 'grad_norm': '10.46', 'learning_rate': '1.725e-05', 'epoch': '0.1546'}

 15%|█▌        | 773/5000 [10:23:52<67:35:58, 57.57s/it][2026-05-22 02:30:05,815] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 96])
[2026-05-22 02:30:11,709] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 117])

 15%|█▌        | 774/5000 [10:25:04<72:50:40, 62.05s/it]
                                                        
{'loss': '0.4418', 'grad_norm': '11.1', 'learning_rate': '1.725e-05', 'epoch': '0.1548'}

 15%|█▌        | 774/5000 [10:25:04<72:50:40, 62.05s/it][2026-05-22 02:31:03,924] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 77])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 02:31:05,913] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 81])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])

 16%|█▌        | 775/5000 [10:25:58<69:50:25, 59.51s/it]
                                                        
{'loss': '1.457', 'grad_norm': '11.34', 'learning_rate': '1.724e-05', 'epoch': '0.155'}

 16%|█▌        | 775/5000 [10:25:58<69:50:25, 59.51s/it][2026-05-22 02:31:59,153] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 121])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 02:32:04,712] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 162])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 16%|█▌        | 776/5000 [10:26:48<66:25:59, 56.62s/it]
                                                        
{'loss': '1.606', 'grad_norm': '14.06', 'learning_rate': '1.724e-05', 'epoch': '0.1552'}

 16%|█▌        | 776/5000 [10:26:48<66:25:59, 56.62s/it][2026-05-22 02:32:39,139] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 327])
[2026-05-22 02:32:39,320] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 200])

 16%|█▌        | 777/5000 [10:27:41<65:16:18, 55.64s/it]
                                                        
{'loss': '0.3134', 'grad_norm': '17.56', 'learning_rate': '1.724e-05', 'epoch': '0.1554'}

 16%|█▌        | 777/5000 [10:27:41<65:16:18, 55.64s/it][2026-05-22 02:33:35,474] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 152])
[2026-05-22 02:33:36,093] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])

 16%|█▌        | 778/5000 [10:28:31<63:11:17, 53.88s/it]
                                                        
{'loss': '0.3497', 'grad_norm': '10.76', 'learning_rate': '1.723e-05', 'epoch': '0.1556'}

 16%|█▌        | 778/5000 [10:28:31<63:11:17, 53.88s/it][2026-05-22 02:34:26,475] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 635])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 111])
[2026-05-22 02:34:27,715] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 652])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])

 16%|█▌        | 779/5000 [10:29:17<60:21:54, 51.48s/it]
                                                        
{'loss': '0.09682', 'grad_norm': '4.928', 'learning_rate': '1.723e-05', 'epoch': '0.1558'}

 16%|█▌        | 779/5000 [10:29:17<60:21:54, 51.48s/it][2026-05-22 02:35:11,074] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 636])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-22 02:35:11,075] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 642])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])

 16%|█▌        | 780/5000 [10:30:03<58:38:44, 50.03s/it]
                                                        
{'loss': '0.07303', 'grad_norm': '4.87', 'learning_rate': '1.722e-05', 'epoch': '0.156'}

 16%|█▌        | 780/5000 [10:30:03<58:38:44, 50.03s/it][2026-05-22 02:35:51,221] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 695])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 61])
[2026-05-22 02:35:51,254] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 680])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 67])

 16%|█▌        | 781/5000 [10:30:51<57:37:25, 49.17s/it]
                                                        
{'loss': '0.758', 'grad_norm': '13.17', 'learning_rate': '1.722e-05', 'epoch': '0.1562'}

 16%|█▌        | 781/5000 [10:30:51<57:37:25, 49.17s/it][2026-05-22 02:36:38,356] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 661])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 81])
[2026-05-22 02:36:40,177] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 679])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])

 16%|█▌        | 782/5000 [10:31:36<56:09:36, 47.93s/it]
                                                        
{'loss': '1.278', 'grad_norm': '11.8', 'learning_rate': '1.722e-05', 'epoch': '0.1564'}

 16%|█▌        | 782/5000 [10:31:36<56:09:36, 47.93s/it][2026-05-22 02:37:38,477] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 394])
[2026-05-22 02:37:40,533] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1015])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 361])

 16%|█▌        | 783/5000 [10:32:28<57:44:25, 49.29s/it]
                                                        
{'loss': '0.4582', 'grad_norm': '11.71', 'learning_rate': '1.721e-05', 'epoch': '0.1566'}

 16%|█▌        | 783/5000 [10:32:28<57:44:25, 49.29s/it][2026-05-22 02:38:27,864] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 122])
[2026-05-22 02:38:33,441] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1014])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 16%|█▌        | 784/5000 [10:33:20<58:45:41, 50.18s/it]
                                                        
{'loss': '0.3251', 'grad_norm': '18.04', 'learning_rate': '1.721e-05', 'epoch': '0.1568'}

 16%|█▌        | 784/5000 [10:33:20<58:45:41, 50.18s/it][2026-05-22 02:39:20,813] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 02:39:21,731] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

 16%|█▌        | 785/5000 [10:33:56<53:45:00, 45.91s/it]
                                                        
{'loss': '1.21', 'grad_norm': '17.4', 'learning_rate': '1.72e-05', 'epoch': '0.157'}

 16%|█▌        | 785/5000 [10:33:56<53:45:00, 45.91s/it][2026-05-22 02:39:56,892] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 02:39:57,465] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 16%|█▌        | 786/5000 [10:34:32<50:02:40, 42.75s/it]
                                                        
{'loss': '1.094', 'grad_norm': '15.4', 'learning_rate': '1.72e-05', 'epoch': '0.1572'}

 16%|█▌        | 786/5000 [10:34:32<50:02:40, 42.75s/it][2026-05-22 02:40:21,268] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 30])
[2026-05-22 02:40:23,089] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 483])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 21])

 16%|█▌        | 787/5000 [10:35:05<46:49:05, 40.01s/it]
                                                        
{'loss': '0.8278', 'grad_norm': '8.311', 'learning_rate': '1.72e-05', 'epoch': '0.1574'}

 16%|█▌        | 787/5000 [10:35:05<46:49:05, 40.01s/it][2026-05-22 02:40:53,547] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 72])
[2026-05-22 02:40:54,090] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 29])

 16%|█▌        | 788/5000 [10:35:39<44:39:49, 38.17s/it]
                                                        
{'loss': '0.8447', 'grad_norm': '8.335', 'learning_rate': '1.719e-05', 'epoch': '0.1576'}

 16%|█▌        | 788/5000 [10:35:39<44:39:49, 38.17s/it][2026-05-22 02:41:28,137] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 02:41:29,762] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 16%|█▌        | 789/5000 [10:36:33<50:13:04, 42.93s/it]
                                                        
{'loss': '0.387', 'grad_norm': '8.584', 'learning_rate': '1.719e-05', 'epoch': '0.1578'}

 16%|█▌        | 789/5000 [10:36:33<50:13:04, 42.93s/it][2026-05-22 02:42:23,474] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-22 02:42:24,600] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 16%|█▌        | 790/5000 [10:37:25<53:16:15, 45.55s/it]
                                                        
{'loss': '0.5164', 'grad_norm': '9.955', 'learning_rate': '1.718e-05', 'epoch': '0.158'}

 16%|█▌        | 790/5000 [10:37:25<53:16:15, 45.55s/it][2026-05-22 02:43:19,172] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 02:43:19,380] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 16%|█▌        | 791/5000 [10:38:35<61:49:32, 52.88s/it]
                                                        
{'loss': '1.117', 'grad_norm': '9.959', 'learning_rate': '1.718e-05', 'epoch': '0.1582'}

 16%|█▌        | 791/5000 [10:38:35<61:49:32, 52.88s/it][2026-05-22 02:44:27,222] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 177])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 02:44:27,945] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 16%|█▌        | 792/5000 [10:39:46<68:02:43, 58.21s/it]
                                                        
{'loss': '0.9977', 'grad_norm': '8.765', 'learning_rate': '1.718e-05', 'epoch': '0.1584'}

 16%|█▌        | 792/5000 [10:39:46<68:02:43, 58.21s/it][2026-05-22 02:45:36,118] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 02:45:36,826] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

 16%|█▌        | 793/5000 [10:40:34<64:26:47, 55.15s/it]
                                                        
{'loss': '1.371', 'grad_norm': '23.06', 'learning_rate': '1.717e-05', 'epoch': '0.1586'}

 16%|█▌        | 793/5000 [10:40:34<64:26:47, 55.15s/it][2026-05-22 02:46:22,616] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 197])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 02:46:24,016] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 474])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 16%|█▌        | 794/5000 [10:41:20<61:20:21, 52.50s/it]
                                                        
{'loss': '1.36', 'grad_norm': '23.94', 'learning_rate': '1.717e-05', 'epoch': '0.1588'}

 16%|█▌        | 794/5000 [10:41:20<61:20:21, 52.50s/it][2026-05-22 02:47:20,076] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 233])
[2026-05-22 02:47:22,028] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 350])

 16%|█▌        | 795/5000 [10:42:12<61:16:47, 52.46s/it]
                                                        
{'loss': '1.706', 'grad_norm': '18.21', 'learning_rate': '1.716e-05', 'epoch': '0.159'}

 16%|█▌        | 795/5000 [10:42:12<61:16:47, 52.46s/it][2026-05-22 02:48:13,114] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 323])
[2026-05-22 02:48:14,693] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 232])

 16%|█▌        | 796/5000 [10:43:05<61:24:09, 52.58s/it]
                                                        
{'loss': '1.368', 'grad_norm': '16.02', 'learning_rate': '1.716e-05', 'epoch': '0.1592'}

 16%|█▌        | 796/5000 [10:43:05<61:24:09, 52.58s/it][2026-05-22 02:48:52,977] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 278])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 286])
[2026-05-22 02:48:53,254] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 16%|█▌        | 797/5000 [10:43:52<59:26:47, 50.92s/it]
                                                        
{'loss': '0.9236', 'grad_norm': '8.323', 'learning_rate': '1.716e-05', 'epoch': '0.1594'}

 16%|█▌        | 797/5000 [10:43:52<59:26:47, 50.92s/it][2026-05-22 02:49:40,494] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 525])
[2026-05-22 02:49:41,043] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 16%|█▌        | 798/5000 [10:44:45<59:58:15, 51.38s/it]
                                                        
{'loss': '0.5938', 'grad_norm': '10.29', 'learning_rate': '1.715e-05', 'epoch': '0.1596'}

 16%|█▌        | 798/5000 [10:44:45<59:58:15, 51.38s/it][2026-05-22 02:50:31,954] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 290])
[2026-05-22 02:50:33,151] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])

 16%|█▌        | 799/5000 [10:45:32<58:35:12, 50.21s/it]
                                                        
{'loss': '0.864', 'grad_norm': '11.36', 'learning_rate': '1.715e-05', 'epoch': '0.1598'}

 16%|█▌        | 799/5000 [10:45:32<58:35:12, 50.21s/it][2026-05-22 02:51:20,328] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 520])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 274])
[2026-05-22 02:51:20,586] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 507])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 260])

 16%|█▌        | 800/5000 [10:46:24<59:07:21, 50.68s/it]
                                                        
{'loss': '0.8463', 'grad_norm': '7.655', 'learning_rate': '1.714e-05', 'epoch': '0.16'}

 16%|█▌        | 800/5000 [10:46:24<59:07:21, 50.68s/it][2026-05-22 02:52:11,684] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 222])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 237])
[2026-05-22 02:52:11,837] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])

 16%|█▌        | 801/5000 [10:47:13<58:33:29, 50.20s/it]
                                                        
{'loss': '0.3656', 'grad_norm': '8.577', 'learning_rate': '1.714e-05', 'epoch': '0.1602'}

 16%|█▌        | 801/5000 [10:47:13<58:33:29, 50.20s/it][2026-05-22 02:53:00,858] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])
[2026-05-22 02:53:02,007] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 336])

 16%|█▌        | 802/5000 [10:48:02<58:12:15, 49.91s/it]
                                                        
{'loss': '0.2255', 'grad_norm': '6.199', 'learning_rate': '1.713e-05', 'epoch': '0.1604'}

 16%|█▌        | 802/5000 [10:48:02<58:12:15, 49.91s/it][2026-05-22 02:53:51,288] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 02:53:52,643] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 16%|█▌        | 803/5000 [10:48:33<51:34:02, 44.23s/it]
                                                        
{'loss': '0.3604', 'grad_norm': '8.522', 'learning_rate': '1.713e-05', 'epoch': '0.1606'}

 16%|█▌        | 803/5000 [10:48:33<51:34:02, 44.23s/it][2026-05-22 02:54:22,672] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 02:54:23,014] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 16%|█▌        | 804/5000 [10:49:07<47:57:41, 41.15s/it]
                                                        
{'loss': '0.04184', 'grad_norm': '4.499', 'learning_rate': '1.713e-05', 'epoch': '0.1608'}

 16%|█▌        | 804/5000 [10:49:07<47:57:41, 41.15s/it][2026-05-22 02:54:56,387] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 281])
[2026-05-22 02:54:57,299] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])

 16%|█▌        | 805/5000 [10:49:38<44:20:52, 38.06s/it]
                                                        
{'loss': '0.2214', 'grad_norm': '7.233', 'learning_rate': '1.712e-05', 'epoch': '0.161'}

 16%|█▌        | 805/5000 [10:49:38<44:20:52, 38.06s/it][2026-05-22 02:55:27,259] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 293])
[2026-05-22 02:55:28,201] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 283])

 16%|█▌        | 806/5000 [10:50:11<42:29:26, 36.47s/it]
                                                        
{'loss': '0.04676', 'grad_norm': '3.767', 'learning_rate': '1.712e-05', 'epoch': '0.1612'}

 16%|█▌        | 806/5000 [10:50:11<42:29:26, 36.47s/it][2026-05-22 02:56:12,467] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 02:56:15,120] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 16%|█▌        | 807/5000 [10:50:44<41:16:56, 35.44s/it]
                                                        
{'loss': '0.3963', 'grad_norm': '8.666', 'learning_rate': '1.711e-05', 'epoch': '0.1614'}

 16%|█▌        | 807/5000 [10:50:44<41:16:56, 35.44s/it][2026-05-22 02:56:46,292] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 02:56:46,549] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])

 16%|█▌        | 808/5000 [10:51:18<40:48:03, 35.04s/it]
                                                        
{'loss': '0.07381', 'grad_norm': '5.754', 'learning_rate': '1.711e-05', 'epoch': '0.1616'}

 16%|█▌        | 808/5000 [10:51:18<40:48:03, 35.04s/it][2026-05-22 02:57:09,227] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 94])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 02:57:10,914] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 128])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 16%|█▌        | 809/5000 [10:52:10<46:46:54, 40.18s/it]
                                                        
{'loss': '0.8166', 'grad_norm': '15', 'learning_rate': '1.711e-05', 'epoch': '0.1618'}

 16%|█▌        | 809/5000 [10:52:10<46:46:54, 40.18s/it][2026-05-22 02:58:02,694] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 94])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])
[2026-05-22 02:58:06,224] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 98])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 16%|█▌        | 810/5000 [10:53:04<51:31:54, 44.28s/it]
                                                        
{'loss': '0.2907', 'grad_norm': '14.97', 'learning_rate': '1.71e-05', 'epoch': '0.162'}

 16%|█▌        | 810/5000 [10:53:04<51:31:54, 44.28s/it][2026-05-22 02:58:53,082] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 239])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 51])
[2026-05-22 02:58:53,813] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 72])

 16%|█▌        | 811/5000 [10:53:51<52:27:10, 45.08s/it]
                                                        
{'loss': '0.681', 'grad_norm': '16.33', 'learning_rate': '1.71e-05', 'epoch': '0.1622'}

 16%|█▌        | 811/5000 [10:53:51<52:27:10, 45.08s/it][2026-05-22 02:59:41,180] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 508])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 112])
[2026-05-22 02:59:41,315] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 177])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 89])

 16%|█▌        | 812/5000 [10:54:39<53:37:15, 46.09s/it]
                                                        
{'loss': '0.3184', 'grad_norm': '11.59', 'learning_rate': '1.709e-05', 'epoch': '0.1624'}

 16%|█▌        | 812/5000 [10:54:39<53:37:15, 46.09s/it][2026-05-22 03:00:26,646] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 399])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 79])
[2026-05-22 03:00:27,301] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 370])

 16%|█▋        | 813/5000 [10:55:50<62:04:23, 53.37s/it]
                                                        
{'loss': '0.3311', 'grad_norm': '11.19', 'learning_rate': '1.709e-05', 'epoch': '0.1626'}

 16%|█▋        | 813/5000 [10:55:50<62:04:23, 53.37s/it][2026-05-22 03:01:36,887] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 338])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 03:01:38,090] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 428])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 72])

 16%|█▋        | 814/5000 [10:56:59<67:41:47, 58.22s/it]
                                                        
{'loss': '0.7536', 'grad_norm': '11.68', 'learning_rate': '1.709e-05', 'epoch': '0.1628'}

 16%|█▋        | 814/5000 [10:56:59<67:41:47, 58.22s/it][2026-05-22 03:02:46,465] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 281])
[2026-05-22 03:02:47,995] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 527])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 285])

 16%|█▋        | 815/5000 [10:57:47<63:59:06, 55.04s/it]
                                                        
{'loss': '0.8892', 'grad_norm': '13.69', 'learning_rate': '1.708e-05', 'epoch': '0.163'}

 16%|█▋        | 815/5000 [10:57:47<63:59:06, 55.04s/it][2026-05-22 03:03:35,163] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 236])
[2026-05-22 03:03:35,855] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])

 16%|█▋        | 816/5000 [10:58:35<61:38:18, 53.04s/it]
                                                        
{'loss': '1.385', 'grad_norm': '14.77', 'learning_rate': '1.708e-05', 'epoch': '0.1632'}

 16%|█▋        | 816/5000 [10:58:35<61:38:18, 53.04s/it][2026-05-22 03:04:26,817] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 298])
[2026-05-22 03:04:27,178] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 242])

 16%|█▋        | 817/5000 [10:59:01<52:07:34, 44.86s/it]
                                                        
{'loss': '0.251', 'grad_norm': '11.22', 'learning_rate': '1.707e-05', 'epoch': '0.1634'}

 16%|█▋        | 817/5000 [10:59:01<52:07:34, 44.86s/it][2026-05-22 03:04:52,861] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 280])
[2026-05-22 03:04:53,375] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1012])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 557])

 16%|█▋        | 818/5000 [10:59:31<46:56:50, 40.41s/it]
                                                        
{'loss': '0.1508', 'grad_norm': '10.17', 'learning_rate': '1.707e-05', 'epoch': '0.1636'}

 16%|█▋        | 818/5000 [10:59:31<46:56:50, 40.41s/it][2026-05-22 03:05:29,927] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1017])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 7])
[2026-05-22 03:05:30,037] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 12])

 16%|█▋        | 819/5000 [11:00:01<43:15:40, 37.25s/it]
                                                        
{'loss': '2.487', 'grad_norm': '63.16', 'learning_rate': '1.707e-05', 'epoch': '0.1638'}

 16%|█▋        | 819/5000 [11:00:01<43:15:40, 37.25s/it][2026-05-22 03:05:57,934] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 973])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 9])
[2026-05-22 03:05:58,964] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 984])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 9])

 16%|█▋        | 820/5000 [11:00:28<39:51:23, 34.33s/it]
                                                        
{'loss': '1.838', 'grad_norm': '29.83', 'learning_rate': '1.706e-05', 'epoch': '0.164'}

 16%|█▋        | 820/5000 [11:00:28<39:51:23, 34.33s/it][2026-05-22 03:06:16,238] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 618])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])
[2026-05-22 03:06:16,323] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 538])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 274])

 16%|█▋        | 821/5000 [11:00:59<38:40:20, 33.31s/it]
                                                        
{'loss': '0.6692', 'grad_norm': '13.82', 'learning_rate': '1.706e-05', 'epoch': '0.1642'}

 16%|█▋        | 821/5000 [11:00:59<38:40:20, 33.31s/it][2026-05-22 03:06:47,254] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 668])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 286])
[2026-05-22 03:06:47,765] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 642])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 266])

 16%|█▋        | 822/5000 [11:01:29<37:25:00, 32.24s/it]
                                                        
{'loss': '0.05396', 'grad_norm': '4.308', 'learning_rate': '1.705e-05', 'epoch': '0.1644'}

 16%|█▋        | 822/5000 [11:01:29<37:25:00, 32.24s/it][2026-05-22 03:07:19,915] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 03:07:21,528] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

 16%|█▋        | 823/5000 [11:02:20<43:53:00, 37.82s/it]
                                                        
{'loss': '0.4968', 'grad_norm': '12.16', 'learning_rate': '1.705e-05', 'epoch': '0.1646'}

 16%|█▋        | 823/5000 [11:02:20<43:53:00, 37.82s/it][2026-05-22 03:08:11,725] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 439])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 03:08:13,499] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 178])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 16%|█▋        | 824/5000 [11:03:15<50:04:04, 43.16s/it]
                                                        
{'loss': '0.269', 'grad_norm': '7.281', 'learning_rate': '1.704e-05', 'epoch': '0.1648'}

 16%|█▋        | 824/5000 [11:03:15<50:04:04, 43.16s/it][2026-05-22 03:09:06,968] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 306])
[2026-05-22 03:09:07,628] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])

 16%|█▋        | 825/5000 [11:04:02<51:23:55, 44.32s/it]
                                                        
{'loss': '2.184', 'grad_norm': '15.23', 'learning_rate': '1.704e-05', 'epoch': '0.165'}

 16%|█▋        | 825/5000 [11:04:02<51:23:55, 44.32s/it][2026-05-22 03:09:53,828] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 153])
[2026-05-22 03:09:54,840] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 197])

 17%|█▋        | 826/5000 [11:04:51<52:53:25, 45.62s/it]
                                                        
{'loss': '1.889', 'grad_norm': '11.27', 'learning_rate': '1.704e-05', 'epoch': '0.1652'}

 17%|█▋        | 826/5000 [11:04:51<52:53:25, 45.62s/it][2026-05-22 03:10:49,155] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 03:10:50,069] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 185])

 17%|█▋        | 827/5000 [11:05:24<48:20:29, 41.70s/it]
                                                        
{'loss': '1.946', 'grad_norm': '9.804', 'learning_rate': '1.703e-05', 'epoch': '0.1654'}

 17%|█▋        | 827/5000 [11:05:24<48:20:29, 41.70s/it][2026-05-22 03:11:20,845] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 78])
[2026-05-22 03:11:22,679] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 83])

 17%|█▋        | 828/5000 [11:05:59<46:11:32, 39.86s/it]
                                                        
{'loss': '1.622', 'grad_norm': '6.175', 'learning_rate': '1.703e-05', 'epoch': '0.1656'}

 17%|█▋        | 828/5000 [11:05:59<46:11:32, 39.86s/it][2026-05-22 03:12:06,918] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 241])
[2026-05-22 03:12:09,889] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 251])

 17%|█▋        | 829/5000 [11:06:48<49:15:27, 42.51s/it]
                                                        
{'loss': '1.236', 'grad_norm': '9.704', 'learning_rate': '1.702e-05', 'epoch': '0.1658'}

 17%|█▋        | 829/5000 [11:06:48<49:15:27, 42.51s/it][2026-05-22 03:12:57,322] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 299])
[2026-05-22 03:12:59,903] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 299])

 17%|█▋        | 830/5000 [11:07:39<52:06:31, 44.99s/it]
                                                        
{'loss': '0.7977', 'grad_norm': '8.402', 'learning_rate': '1.702e-05', 'epoch': '0.166'}

 17%|█▋        | 830/5000 [11:07:39<52:06:31, 44.99s/it][2026-05-22 03:13:26,723] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-22 03:13:27,456] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])

 17%|█▋        | 831/5000 [11:08:27<53:17:18, 46.02s/it]
                                                        
{'loss': '0.3021', 'grad_norm': '10.1', 'learning_rate': '1.702e-05', 'epoch': '0.1662'}

 17%|█▋        | 831/5000 [11:08:27<53:17:18, 46.02s/it][2026-05-22 03:14:15,197] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 386])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 03:14:16,108] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 972])

 17%|█▋        | 832/5000 [11:09:15<54:00:37, 46.65s/it]
                                                        
{'loss': '0.2413', 'grad_norm': '7.125', 'learning_rate': '1.701e-05', 'epoch': '0.1664'}

 17%|█▋        | 832/5000 [11:09:15<54:00:37, 46.65s/it][2026-05-22 03:15:09,502] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 03:15:11,203] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 17%|█▋        | 833/5000 [11:10:02<53:55:37, 46.59s/it]
                                                        
{'loss': '0.1917', 'grad_norm': '8.426', 'learning_rate': '1.701e-05', 'epoch': '0.1666'}

 17%|█▋        | 833/5000 [11:10:02<53:55:37, 46.59s/it][2026-05-22 03:15:58,200] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])
[2026-05-22 03:15:58,941] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 17%|█▋        | 834/5000 [11:10:48<53:49:24, 46.51s/it]
                                                        
{'loss': '0.1249', 'grad_norm': '5.593', 'learning_rate': '1.7e-05', 'epoch': '0.1668'}

 17%|█▋        | 834/5000 [11:10:48<53:49:24, 46.51s/it][2026-05-22 03:16:35,206] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 03:16:36,511] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 252])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 17%|█▋        | 835/5000 [11:11:34<53:40:23, 46.39s/it]
                                                        
{'loss': '0.3184', 'grad_norm': '7.666', 'learning_rate': '1.7e-05', 'epoch': '0.167'}

 17%|█▋        | 835/5000 [11:11:34<53:40:23, 46.39s/it][2026-05-22 03:17:22,258] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 299])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 03:17:22,856] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 501])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 17%|█▋        | 836/5000 [11:12:22<54:00:41, 46.70s/it]
                                                        
{'loss': '0.1477', 'grad_norm': '7.162', 'learning_rate': '1.7e-05', 'epoch': '0.1672'}

 17%|█▋        | 836/5000 [11:12:22<54:00:41, 46.70s/it][2026-05-22 03:18:10,030] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 638])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 03:18:10,428] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 657])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 17%|█▋        | 837/5000 [11:13:09<54:08:26, 46.82s/it]
                                                        
{'loss': '1.222', 'grad_norm': '23.6', 'learning_rate': '1.699e-05', 'epoch': '0.1674'}

 17%|█▋        | 837/5000 [11:13:09<54:08:26, 46.82s/it][2026-05-22 03:18:57,052] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 639])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 03:18:57,531] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 658])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 17%|█▋        | 838/5000 [11:13:58<54:58:06, 47.55s/it]
                                                        
{'loss': '1.171', 'grad_norm': '19.6', 'learning_rate': '1.699e-05', 'epoch': '0.1676'}

 17%|█▋        | 838/5000 [11:13:58<54:58:06, 47.55s/it][2026-05-22 03:19:57,070] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 106])
[2026-05-22 03:19:57,630] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 173])

 17%|█▋        | 839/5000 [11:14:49<56:07:03, 48.55s/it]
                                                        
{'loss': '0.7313', 'grad_norm': '10.16', 'learning_rate': '1.698e-05', 'epoch': '0.1678'}

 17%|█▋        | 839/5000 [11:14:49<56:07:03, 48.55s/it][2026-05-22 03:20:46,323] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 215])
[2026-05-22 03:20:47,446] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 163])

 17%|█▋        | 840/5000 [11:15:42<57:51:14, 50.07s/it]
                                                        
{'loss': '0.2107', 'grad_norm': '6.103', 'learning_rate': '1.698e-05', 'epoch': '0.168'}

 17%|█▋        | 840/5000 [11:15:42<57:51:14, 50.07s/it][2026-05-22 03:21:33,441] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 11])
[2026-05-22 03:21:35,716] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 17])

 17%|█▋        | 841/5000 [11:16:13<51:03:26, 44.19s/it]
                                                        
{'loss': '0.1834', 'grad_norm': '7.401', 'learning_rate': '1.698e-05', 'epoch': '0.1682'}

 17%|█▋        | 841/5000 [11:16:13<51:03:26, 44.19s/it][2026-05-22 03:22:03,566] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 13])
[2026-05-22 03:22:05,272] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 15])

 17%|█▋        | 842/5000 [11:16:50<48:41:09, 42.15s/it]
                                                        
{'loss': '0.06811', 'grad_norm': '5.876', 'learning_rate': '1.697e-05', 'epoch': '0.1684'}

 17%|█▋        | 842/5000 [11:16:50<48:41:09, 42.15s/it][2026-05-22 03:22:40,817] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 211])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 03:22:41,881] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 493])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 17%|█▋        | 843/5000 [11:17:29<47:27:51, 41.10s/it]
                                                        
{'loss': '2.198', 'grad_norm': '15.57', 'learning_rate': '1.697e-05', 'epoch': '0.1686'}

 17%|█▋        | 843/5000 [11:17:29<47:27:51, 41.10s/it][2026-05-22 03:23:21,553] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 470])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 03:23:21,699] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])

 17%|█▋        | 844/5000 [11:18:09<46:55:36, 40.65s/it]
                                                        
{'loss': '2.071', 'grad_norm': '12.34', 'learning_rate': '1.696e-05', 'epoch': '0.1688'}

 17%|█▋        | 844/5000 [11:18:09<46:55:36, 40.65s/it][2026-05-22 03:23:56,445] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 280])
[2026-05-22 03:23:56,892] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 509])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 266])

 17%|█▋        | 845/5000 [11:18:56<49:16:07, 42.69s/it]
                                                        
{'loss': '1.201', 'grad_norm': '9.727', 'learning_rate': '1.696e-05', 'epoch': '0.169'}

 17%|█▋        | 845/5000 [11:18:56<49:16:07, 42.69s/it][2026-05-22 03:24:44,071] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 529])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])
[2026-05-22 03:24:44,858] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])

 17%|█▋        | 846/5000 [11:19:45<51:31:33, 44.65s/it]
                                                        
{'loss': '1.156', 'grad_norm': '11.32', 'learning_rate': '1.696e-05', 'epoch': '0.1692'}

 17%|█▋        | 846/5000 [11:19:45<51:31:33, 44.65s/it][2026-05-22 03:25:32,633] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 410])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 245])
[2026-05-22 03:25:33,812] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 17%|█▋        | 847/5000 [11:20:33<52:27:52, 45.48s/it]
                                                        
{'loss': '1.904', 'grad_norm': '27.52', 'learning_rate': '1.695e-05', 'epoch': '0.1694'}

 17%|█▋        | 847/5000 [11:20:33<52:27:52, 45.48s/it][2026-05-22 03:26:20,892] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 03:26:21,299] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 17%|█▋        | 848/5000 [11:21:19<52:52:00, 45.84s/it]
                                                        
{'loss': '1.594', 'grad_norm': '26.12', 'learning_rate': '1.695e-05', 'epoch': '0.1696'}

 17%|█▋        | 848/5000 [11:21:19<52:52:00, 45.84s/it][2026-05-22 03:27:07,258] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 266])
[2026-05-22 03:27:07,577] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 383])

 17%|█▋        | 849/5000 [11:22:08<54:00:37, 46.84s/it]
                                                        
{'loss': '1.231', 'grad_norm': '10.17', 'learning_rate': '1.694e-05', 'epoch': '0.1698'}

 17%|█▋        | 849/5000 [11:22:08<54:00:37, 46.84s/it][2026-05-22 03:27:56,650] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 370])
[2026-05-22 03:27:57,137] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 472])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 296])

 17%|█▋        | 850/5000 [11:23:00<55:47:01, 48.39s/it]
                                                        
{'loss': '0.3745', 'grad_norm': '10.78', 'learning_rate': '1.694e-05', 'epoch': '0.17'}

 17%|█▋        | 850/5000 [11:23:00<55:47:01, 48.39s/it][2026-05-22 03:28:46,702] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 192])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 22])
[2026-05-22 03:28:47,206] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 202])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 22])

 17%|█▋        | 851/5000 [11:23:29<48:59:38, 42.51s/it]
                                                        
{'loss': '0.3946', 'grad_norm': '8.737', 'learning_rate': '1.693e-05', 'epoch': '0.1702'}

 17%|█▋        | 851/5000 [11:23:29<48:59:38, 42.51s/it][2026-05-22 03:29:15,529] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 186])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 21])
[2026-05-22 03:29:15,984] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 218])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 23])

 17%|█▋        | 852/5000 [11:23:59<44:32:07, 38.65s/it]
                                                        
{'loss': '0.05741', 'grad_norm': '4.686', 'learning_rate': '1.693e-05', 'epoch': '0.1704'}

 17%|█▋        | 852/5000 [11:23:59<44:32:07, 38.65s/it][2026-05-22 03:29:47,577] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 244])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 03:29:48,194] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 148])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 17%|█▋        | 853/5000 [11:24:32<42:37:26, 37.00s/it]
                                                        
{'loss': '0.3465', 'grad_norm': '8.015', 'learning_rate': '1.693e-05', 'epoch': '0.1706'}

 17%|█▋        | 853/5000 [11:24:32<42:37:26, 37.00s/it][2026-05-22 03:30:20,660] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 124])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 03:30:21,964] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 140])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 17%|█▋        | 854/5000 [11:25:06<41:35:59, 36.12s/it]
                                                        
{'loss': '0.04521', 'grad_norm': '4.676', 'learning_rate': '1.692e-05', 'epoch': '0.1708'}

 17%|█▋        | 854/5000 [11:25:06<41:35:59, 36.12s/it][2026-05-22 03:30:54,797] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 548])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 03:30:55,994] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 17%|█▋        | 855/5000 [11:25:37<39:43:13, 34.50s/it]
                                                        
{'loss': '0.5324', 'grad_norm': '12.07', 'learning_rate': '1.692e-05', 'epoch': '0.171'}

 17%|█▋        | 855/5000 [11:25:37<39:43:13, 34.50s/it][2026-05-22 03:31:26,034] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])
[2026-05-22 03:31:26,678] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 944])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 17%|█▋        | 856/5000 [11:26:08<38:30:50, 33.46s/it]
                                                        
{'loss': '0.05053', 'grad_norm': '4.285', 'learning_rate': '1.691e-05', 'epoch': '0.1712'}

 17%|█▋        | 856/5000 [11:26:08<38:30:50, 33.46s/it][2026-05-22 03:31:56,766] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 03:31:57,311] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 17%|█▋        | 857/5000 [11:26:32<35:17:58, 30.67s/it]
                                                        
{'loss': '2.344', 'grad_norm': '8.833', 'learning_rate': '1.691e-05', 'epoch': '0.1714'}

 17%|█▋        | 857/5000 [11:26:32<35:17:58, 30.67s/it][2026-05-22 03:32:20,888] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 03:32:21,585] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 17%|█▋        | 858/5000 [11:26:56<32:58:22, 28.66s/it]
                                                        
{'loss': '2.345', 'grad_norm': '7.152', 'learning_rate': '1.691e-05', 'epoch': '0.1716'}

 17%|█▋        | 858/5000 [11:26:56<32:58:22, 28.66s/it][2026-05-22 03:32:47,188] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 661])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 03:32:48,762] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 628])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 17%|█▋        | 859/5000 [11:27:44<39:29:31, 34.33s/it]
                                                        
{'loss': '1.293', 'grad_norm': '9.483', 'learning_rate': '1.69e-05', 'epoch': '0.1718'}

 17%|█▋        | 859/5000 [11:27:44<39:29:31, 34.33s/it][2026-05-22 03:33:34,776] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 631])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 03:33:36,640] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 653])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 17%|█▋        | 860/5000 [11:28:32<44:10:38, 38.42s/it]
                                                        
{'loss': '0.9568', 'grad_norm': '10.54', 'learning_rate': '1.69e-05', 'epoch': '0.172'}

 17%|█▋        | 860/5000 [11:28:32<44:10:38, 38.42s/it][2026-05-22 03:34:23,405] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 208])
[2026-05-22 03:34:23,765] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 90])

 17%|█▋        | 861/5000 [11:29:13<45:12:10, 39.32s/it]
                                                        
{'loss': '0.5193', 'grad_norm': '12.55', 'learning_rate': '1.689e-05', 'epoch': '0.1722'}

 17%|█▋        | 861/5000 [11:29:13<45:12:10, 39.32s/it][2026-05-22 03:35:04,810] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 154])
[2026-05-22 03:35:05,294] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 126])

 17%|█▋        | 862/5000 [11:30:04<49:20:51, 42.93s/it]
                                                        
{'loss': '0.1847', 'grad_norm': '8.665', 'learning_rate': '1.689e-05', 'epoch': '0.1724'}

 17%|█▋        | 862/5000 [11:30:04<49:20:51, 42.93s/it][2026-05-22 03:36:02,072] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 282])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 03:36:03,070] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 296])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

 17%|█▋        | 863/5000 [11:31:12<57:44:38, 50.25s/it]
                                                        
{'loss': '1.231', 'grad_norm': '11.59', 'learning_rate': '1.689e-05', 'epoch': '0.1726'}

 17%|█▋        | 863/5000 [11:31:12<57:44:38, 50.25s/it][2026-05-22 03:37:08,316] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 03:37:10,311] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 309])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 17%|█▋        | 864/5000 [11:32:20<63:48:56, 55.55s/it]
                                                        
{'loss': '1.374', 'grad_norm': '12.75', 'learning_rate': '1.688e-05', 'epoch': '0.1728'}

 17%|█▋        | 864/5000 [11:32:20<63:48:56, 55.55s/it][2026-05-22 03:38:07,921] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 180])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 03:38:09,607] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 204])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 972])

 17%|█▋        | 865/5000 [11:33:16<64:11:56, 55.89s/it]
                                                        
{'loss': '1.88', 'grad_norm': '11.99', 'learning_rate': '1.688e-05', 'epoch': '0.173'}

 17%|█▋        | 865/5000 [11:33:16<64:11:56, 55.89s/it][2026-05-22 03:39:04,809] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 268])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])
[2026-05-22 03:39:05,978] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 124])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 17%|█▋        | 866/5000 [11:34:13<64:23:32, 56.07s/it]
                                                        
{'loss': '1.399', 'grad_norm': '10.84', 'learning_rate': '1.687e-05', 'epoch': '0.1732'}

 17%|█▋        | 866/5000 [11:34:13<64:23:32, 56.07s/it][2026-05-22 03:40:02,124] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 237])
[2026-05-22 03:40:03,456] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 17%|█▋        | 867/5000 [11:35:02<62:00:11, 54.01s/it]
                                                        
{'loss': '1.006', 'grad_norm': '12.85', 'learning_rate': '1.687e-05', 'epoch': '0.1734'}

 17%|█▋        | 867/5000 [11:35:02<62:00:11, 54.01s/it][2026-05-22 03:40:52,156] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 03:40:52,371] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 17%|█▋        | 868/5000 [11:35:51<60:11:29, 52.44s/it]
                                                        
{'loss': '0.9872', 'grad_norm': '12.91', 'learning_rate': '1.687e-05', 'epoch': '0.1736'}

 17%|█▋        | 868/5000 [11:35:51<60:11:29, 52.44s/it][2026-05-22 03:41:48,633] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 03:41:48,860] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 17%|█▋        | 869/5000 [11:36:39<58:36:23, 51.07s/it]
                                                        
{'loss': '0.8251', 'grad_norm': '10.31', 'learning_rate': '1.686e-05', 'epoch': '0.1738'}

 17%|█▋        | 869/5000 [11:36:39<58:36:23, 51.07s/it][2026-05-22 03:42:36,824] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 03:42:38,804] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 17%|█▋        | 870/5000 [11:37:27<57:42:41, 50.31s/it]
                                                        
{'loss': '0.1379', 'grad_norm': '8.301', 'learning_rate': '1.686e-05', 'epoch': '0.174'}

 17%|█▋        | 870/5000 [11:37:27<57:42:41, 50.31s/it][2026-05-22 03:43:27,909] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 03:43:29,803] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 17%|█▋        | 871/5000 [11:38:15<56:52:01, 49.58s/it]
                                                        
{'loss': '0.8009', 'grad_norm': '10.2', 'learning_rate': '1.685e-05', 'epoch': '0.1742'}

 17%|█▋        | 871/5000 [11:38:15<56:52:01, 49.58s/it][2026-05-22 03:44:16,760] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 941])
[2026-05-22 03:44:20,031] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 17%|█▋        | 872/5000 [11:39:03<56:18:35, 49.11s/it]
                                                        
{'loss': '0.3897', 'grad_norm': '11.25', 'learning_rate': '1.685e-05', 'epoch': '0.1744'}

 17%|█▋        | 872/5000 [11:39:03<56:18:35, 49.11s/it][2026-05-22 03:44:56,812] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 03:44:57,004] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 17%|█▋        | 873/5000 [11:39:56<57:42:15, 50.34s/it]
                                                        
{'loss': '0.3324', 'grad_norm': '10.2', 'learning_rate': '1.684e-05', 'epoch': '0.1746'}

 17%|█▋        | 873/5000 [11:39:56<57:42:15, 50.34s/it][2026-05-22 03:45:49,786] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 03:45:52,922] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 17%|█▋        | 874/5000 [11:40:52<59:38:59, 52.05s/it]
                                                        
{'loss': '0.1667', 'grad_norm': '6.447', 'learning_rate': '1.684e-05', 'epoch': '0.1748'}

 17%|█▋        | 874/5000 [11:40:52<59:38:59, 52.05s/it][2026-05-22 03:46:48,164] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 03:46:48,262] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 399])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 18%|█▊        | 875/5000 [11:41:43<59:10:51, 51.65s/it]
                                                        
{'loss': '0.4006', 'grad_norm': '8.319', 'learning_rate': '1.684e-05', 'epoch': '0.175'}

 18%|█▊        | 875/5000 [11:41:43<59:10:51, 51.65s/it][2026-05-22 03:47:40,468] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 398])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 03:47:42,529] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 18%|█▊        | 876/5000 [11:42:33<58:45:11, 51.29s/it]
                                                        
{'loss': '0.1635', 'grad_norm': '4.881', 'learning_rate': '1.683e-05', 'epoch': '0.1752'}

 18%|█▊        | 876/5000 [11:42:33<58:45:11, 51.29s/it][2026-05-22 03:48:20,557] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 03:48:20,773] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 288])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 18%|█▊        | 877/5000 [11:43:43<64:56:42, 56.71s/it]
                                                        
{'loss': '0.5745', 'grad_norm': '9.221', 'learning_rate': '1.683e-05', 'epoch': '0.1754'}

 18%|█▊        | 877/5000 [11:43:43<64:56:42, 56.71s/it][2026-05-22 03:49:29,646] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 280])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 03:49:30,393] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 304])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 18%|█▊        | 878/5000 [11:44:52<69:16:14, 60.50s/it]
                                                        
{'loss': '0.2814', 'grad_norm': '9.977', 'learning_rate': '1.682e-05', 'epoch': '0.1756'}

 18%|█▊        | 878/5000 [11:44:52<69:16:14, 60.50s/it][2026-05-22 03:50:41,296] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 80])
[2026-05-22 03:50:41,789] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 18%|█▊        | 879/5000 [11:46:03<72:39:18, 63.47s/it]
                                                        
{'loss': '0.3992', 'grad_norm': '10.05', 'learning_rate': '1.682e-05', 'epoch': '0.1758'}

 18%|█▊        | 879/5000 [11:46:03<72:39:18, 63.47s/it][2026-05-22 03:51:51,714] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 120])
[2026-05-22 03:51:52,033] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 104])

 18%|█▊        | 880/5000 [11:47:12<74:51:30, 65.41s/it]
                                                        
{'loss': '0.6644', 'grad_norm': '8.954', 'learning_rate': '1.682e-05', 'epoch': '0.176'}

 18%|█▊        | 880/5000 [11:47:12<74:51:30, 65.41s/it][2026-05-22 03:53:08,921] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 251])
[2026-05-22 03:53:10,586] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 287])

 18%|█▊        | 881/5000 [11:47:45<63:39:20, 55.64s/it]
                                                        
{'loss': '2.248', 'grad_norm': '35.3', 'learning_rate': '1.681e-05', 'epoch': '0.1762'}

 18%|█▊        | 881/5000 [11:47:45<63:39:20, 55.64s/it][2026-05-22 03:53:44,963] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 224])
[2026-05-22 03:53:45,500] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])

 18%|█▊        | 882/5000 [11:48:17<55:22:49, 48.41s/it]
                                                        
{'loss': '1.533', 'grad_norm': '23.37', 'learning_rate': '1.681e-05', 'epoch': '0.1764'}

 18%|█▊        | 882/5000 [11:48:17<55:22:49, 48.41s/it][2026-05-22 03:54:04,185] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 284])
[2026-05-22 03:54:05,481] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 284])

 18%|█▊        | 883/5000 [11:48:45<48:22:14, 42.30s/it]
                                                        
{'loss': '0.6442', 'grad_norm': '20.17', 'learning_rate': '1.68e-05', 'epoch': '0.1766'}

 18%|█▊        | 883/5000 [11:48:45<48:22:14, 42.30s/it][2026-05-22 03:54:33,194] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 434])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])
[2026-05-22 03:54:33,885] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])

 18%|█▊        | 884/5000 [11:49:13<43:34:46, 38.12s/it]
                                                        
{'loss': '0.4463', 'grad_norm': '17.91', 'learning_rate': '1.68e-05', 'epoch': '0.1768'}

 18%|█▊        | 884/5000 [11:49:13<43:34:46, 38.12s/it][2026-05-22 03:55:04,472] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 138])
[2026-05-22 03:55:05,140] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 282])

 18%|█▊        | 885/5000 [11:50:03<47:24:55, 41.48s/it]
                                                        
{'loss': '0.8454', 'grad_norm': '20.25', 'learning_rate': '1.68e-05', 'epoch': '0.177'}

 18%|█▊        | 885/5000 [11:50:03<47:24:55, 41.48s/it][2026-05-22 03:55:53,762] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 245])
[2026-05-22 03:55:54,270] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 89])

 18%|█▊        | 886/5000 [11:50:52<49:59:57, 43.75s/it]
                                                        
{'loss': '0.8923', 'grad_norm': '22.16', 'learning_rate': '1.679e-05', 'epoch': '0.1772'}

 18%|█▊        | 886/5000 [11:50:52<49:59:57, 43.75s/it][2026-05-22 03:57:09,916] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 114])
[2026-05-22 03:57:10,284] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 213])

 18%|█▊        | 887/5000 [11:51:41<51:45:33, 45.30s/it]
                                                        
{'loss': '0.5362', 'grad_norm': '10.84', 'learning_rate': '1.679e-05', 'epoch': '0.1774'}

 18%|█▊        | 887/5000 [11:51:41<51:45:33, 45.30s/it][2026-05-22 03:58:02,286] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 169])
[2026-05-22 03:58:04,516] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 131])

 18%|█▊        | 888/5000 [11:52:30<53:06:32, 46.50s/it]
                                                        
{'loss': '0.1556', 'grad_norm': '4.843', 'learning_rate': '1.678e-05', 'epoch': '0.1776'}

 18%|█▊        | 888/5000 [11:52:30<53:06:32, 46.50s/it][2026-05-22 03:58:19,287] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 212])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 03:58:20,683] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 18%|█▊        | 889/5000 [11:53:00<47:32:22, 41.63s/it]
                                                        
{'loss': '0.2904', 'grad_norm': '8.602', 'learning_rate': '1.678e-05', 'epoch': '0.1778'}

 18%|█▊        | 889/5000 [11:53:00<47:32:22, 41.63s/it][2026-05-22 03:58:49,633] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-22 03:58:50,832] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 18%|█▊        | 890/5000 [11:53:31<43:50:59, 38.41s/it]
                                                        
{'loss': '0.04864', 'grad_norm': '3.944', 'learning_rate': '1.678e-05', 'epoch': '0.178'}

 18%|█▊        | 890/5000 [11:53:31<43:50:59, 38.41s/it][2026-05-22 03:59:20,779] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 03:59:21,535] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

 18%|█▊        | 891/5000 [11:54:19<47:11:53, 41.35s/it]
                                                        
{'loss': '0.1873', 'grad_norm': '7.643', 'learning_rate': '1.677e-05', 'epoch': '0.1782'}

 18%|█▊        | 891/5000 [11:54:19<47:11:53, 41.35s/it][2026-05-22 04:00:10,464] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 434])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-22 04:00:11,585] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 197])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 18%|█▊        | 892/5000 [11:55:09<49:56:42, 43.77s/it]
                                                        
{'loss': '0.2966', 'grad_norm': '10.04', 'learning_rate': '1.677e-05', 'epoch': '0.1784'}

 18%|█▊        | 892/5000 [11:55:09<49:56:42, 43.77s/it][2026-05-22 04:01:04,155] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 105])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 04:01:06,731] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 107])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 18%|█▊        | 893/5000 [11:55:57<51:33:33, 45.19s/it]
                                                        
{'loss': '0.5634', 'grad_norm': '11.42', 'learning_rate': '1.676e-05', 'epoch': '0.1786'}

 18%|█▊        | 893/5000 [11:55:57<51:33:33, 45.19s/it][2026-05-22 04:01:52,098] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 87])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 04:01:53,693] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 106])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 18%|█▊        | 894/5000 [11:56:46<52:58:16, 46.44s/it]
                                                        
{'loss': '0.61', 'grad_norm': '11.06', 'learning_rate': '1.676e-05', 'epoch': '0.1788'}

 18%|█▊        | 894/5000 [11:56:46<52:58:16, 46.44s/it][2026-05-22 04:02:46,896] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])
[2026-05-22 04:02:47,882] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 110])

 18%|█▊        | 895/5000 [11:57:37<54:16:36, 47.60s/it]
                                                        
{'loss': '1.178', 'grad_norm': '10.99', 'learning_rate': '1.676e-05', 'epoch': '0.179'}

 18%|█▊        | 895/5000 [11:57:37<54:16:36, 47.60s/it][2026-05-22 04:03:38,403] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-22 04:03:38,828] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 154])

 18%|█▊        | 896/5000 [11:58:28<55:24:37, 48.61s/it]
                                                        
{'loss': '1.126', 'grad_norm': '13.46', 'learning_rate': '1.675e-05', 'epoch': '0.1792'}

 18%|█▊        | 896/5000 [11:58:28<55:24:37, 48.61s/it][2026-05-22 04:04:16,308] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 920])
[2026-05-22 04:04:16,761] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 18%|█▊        | 897/5000 [11:59:15<54:49:47, 48.11s/it]
                                                        
{'loss': '0.5296', 'grad_norm': '7.429', 'learning_rate': '1.675e-05', 'epoch': '0.1794'}

 18%|█▊        | 897/5000 [11:59:15<54:49:47, 48.11s/it][2026-05-22 04:05:04,246] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 545])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 980])
[2026-05-22 04:05:04,580] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 169])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 18%|█▊        | 898/5000 [12:00:05<55:37:15, 48.81s/it]
                                                        
{'loss': '0.478', 'grad_norm': '12.78', 'learning_rate': '1.674e-05', 'epoch': '0.1796'}

 18%|█▊        | 898/5000 [12:00:05<55:37:15, 48.81s/it][2026-05-22 04:05:54,631] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 984])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 180])
[2026-05-22 04:05:56,137] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 184])

 18%|█▊        | 899/5000 [12:00:56<56:11:23, 49.33s/it]
                                                        
{'loss': '1.238', 'grad_norm': '12.13', 'learning_rate': '1.674e-05', 'epoch': '0.1798'}

 18%|█▊        | 899/5000 [12:00:56<56:11:23, 49.33s/it][2026-05-22 04:06:46,369] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 228])
[2026-05-22 04:06:46,524] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 18%|█▊        | 900/5000 [12:01:44<55:47:50, 48.99s/it]
                                                        
{'loss': '1.556', 'grad_norm': '14.93', 'learning_rate': '1.673e-05', 'epoch': '0.18'}

 18%|█▊        | 900/5000 [12:01:44<55:47:50, 48.99s/it][2026-05-22 04:07:43,613] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 112])
[2026-05-22 04:07:43,940] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 89])

 18%|█▊        | 901/5000 [12:02:34<56:05:15, 49.26s/it]
                                                        
{'loss': '0.2735', 'grad_norm': '7.929', 'learning_rate': '1.673e-05', 'epoch': '0.1802'}

 18%|█▊        | 901/5000 [12:02:34<56:05:15, 49.26s/it][2026-05-22 04:08:31,260] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-22 04:08:34,899] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 65])

 18%|█▊        | 902/5000 [12:03:22<55:47:30, 49.01s/it]
                                                        
{'loss': '0.3268', 'grad_norm': '9.741', 'learning_rate': '1.673e-05', 'epoch': '0.1804'}

 18%|█▊        | 902/5000 [12:03:22<55:47:30, 49.01s/it][2026-05-22 04:09:08,840] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 483])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 106])
[2026-05-22 04:09:09,465] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 969])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 18%|█▊        | 903/5000 [12:04:09<54:55:17, 48.26s/it]
                                                        
{'loss': '1.113', 'grad_norm': '12.81', 'learning_rate': '1.672e-05', 'epoch': '0.1806'}

 18%|█▊        | 903/5000 [12:04:09<54:55:17, 48.26s/it][2026-05-22 04:09:55,933] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 944])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-22 04:09:56,046] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 965])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 59])

 18%|█▊        | 904/5000 [12:05:01<56:23:41, 49.57s/it]
                                                        
{'loss': '0.8165', 'grad_norm': '10.8', 'learning_rate': '1.672e-05', 'epoch': '0.1808'}

 18%|█▊        | 904/5000 [12:05:01<56:23:41, 49.57s/it][2026-05-22 04:10:52,272] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 218])
[2026-05-22 04:10:52,962] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])

 18%|█▊        | 905/5000 [12:05:50<56:04:39, 49.30s/it]
                                                        
{'loss': '0.8083', 'grad_norm': '12.98', 'learning_rate': '1.671e-05', 'epoch': '0.181'}

 18%|█▊        | 905/5000 [12:05:50<56:04:39, 49.30s/it][2026-05-22 04:11:41,162] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 213])
[2026-05-22 04:11:41,221] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 213])

 18%|█▊        | 906/5000 [12:06:39<55:58:58, 49.23s/it]
                                                        
{'loss': '0.8954', 'grad_norm': '15.88', 'learning_rate': '1.671e-05', 'epoch': '0.1812'}

 18%|█▊        | 906/5000 [12:06:39<55:58:58, 49.23s/it][2026-05-22 04:12:39,027] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 04:12:41,484] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])

 18%|█▊        | 907/5000 [12:07:27<55:31:16, 48.83s/it]
                                                        
{'loss': '2.615', 'grad_norm': '7.476', 'learning_rate': '1.671e-05', 'epoch': '0.1814'}

 18%|█▊        | 907/5000 [12:07:27<55:31:16, 48.83s/it][2026-05-22 04:13:29,673] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 04:13:30,195] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 18%|█▊        | 908/5000 [12:08:14<55:02:20, 48.42s/it]
                                                        
{'loss': '2.583', 'grad_norm': '11.76', 'learning_rate': '1.67e-05', 'epoch': '0.1816'}

 18%|█▊        | 908/5000 [12:08:14<55:02:20, 48.42s/it][2026-05-22 04:14:12,475] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-22 04:14:17,297] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 290])

 18%|█▊        | 909/5000 [12:09:01<54:25:24, 47.89s/it]
                                                        
{'loss': '0.466', 'grad_norm': '17.71', 'learning_rate': '1.67e-05', 'epoch': '0.1818'}

 18%|█▊        | 909/5000 [12:09:01<54:25:24, 47.89s/it][2026-05-22 04:15:02,241] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 04:15:06,249] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 132])

 18%|█▊        | 910/5000 [12:09:48<53:58:59, 47.52s/it]
                                                        
{'loss': '0.5477', 'grad_norm': '17.36', 'learning_rate': '1.669e-05', 'epoch': '0.182'}

 18%|█▊        | 910/5000 [12:09:48<53:58:59, 47.52s/it][2026-05-22 04:15:44,063] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 428])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 245])
[2026-05-22 04:15:45,271] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 233])

 18%|█▊        | 911/5000 [12:10:36<54:12:59, 47.73s/it]
                                                        
{'loss': '0.2418', 'grad_norm': '16.39', 'learning_rate': '1.669e-05', 'epoch': '0.1822'}

 18%|█▊        | 911/5000 [12:10:36<54:12:59, 47.73s/it][2026-05-22 04:16:31,835] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 282])
[2026-05-22 04:16:36,523] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 265])

 18%|█▊        | 912/5000 [12:11:25<54:34:53, 48.07s/it]
                                                        
{'loss': '0.3251', 'grad_norm': '9.186', 'learning_rate': '1.669e-05', 'epoch': '0.1824'}

 18%|█▊        | 912/5000 [12:11:25<54:34:53, 48.07s/it][2026-05-22 04:17:14,951] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 433])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 04:17:16,339] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 212])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 18%|█▊        | 913/5000 [12:12:36<62:28:14, 55.03s/it]
                                                        
{'loss': '1.202', 'grad_norm': '11.48', 'learning_rate': '1.668e-05', 'epoch': '0.1826'}

 18%|█▊        | 913/5000 [12:12:36<62:28:14, 55.03s/it][2026-05-22 04:18:25,244] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 04:18:28,589] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 18%|█▊        | 914/5000 [12:13:48<68:05:45, 60.00s/it]
                                                        
{'loss': '1.393', 'grad_norm': '12.86', 'learning_rate': '1.668e-05', 'epoch': '0.1828'}

 18%|█▊        | 914/5000 [12:13:48<68:05:45, 60.00s/it][2026-05-22 04:19:36,590] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 04:19:37,045] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])

 18%|█▊        | 915/5000 [12:14:37<64:35:36, 56.92s/it]
                                                        
{'loss': '0.3099', 'grad_norm': '8.982', 'learning_rate': '1.667e-05', 'epoch': '0.183'}

 18%|█▊        | 915/5000 [12:14:37<64:35:36, 56.92s/it][2026-05-22 04:20:27,273] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 04:20:27,386] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 399])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 18%|█▊        | 916/5000 [12:15:29<62:52:10, 55.42s/it]
                                                        
{'loss': '0.4559', 'grad_norm': '11.4', 'learning_rate': '1.667e-05', 'epoch': '0.1832'}

 18%|█▊        | 916/5000 [12:15:29<62:52:10, 55.42s/it][2026-05-22 04:21:16,315] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 205])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 04:21:16,641] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 296])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 18%|█▊        | 917/5000 [12:15:56<53:12:40, 46.92s/it]
                                                        
{'loss': '0.214', 'grad_norm': '7.866', 'learning_rate': '1.667e-05', 'epoch': '0.1834'}

 18%|█▊        | 917/5000 [12:15:56<53:12:40, 46.92s/it][2026-05-22 04:21:43,374] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 262])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 04:21:43,485] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 270])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 18%|█▊        | 918/5000 [12:16:26<47:25:41, 41.83s/it]
                                                        
{'loss': '0.05811', 'grad_norm': '5.858', 'learning_rate': '1.666e-05', 'epoch': '0.1836'}

 18%|█▊        | 918/5000 [12:16:26<47:25:41, 41.83s/it][2026-05-22 04:22:25,823] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 04:22:29,493] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 242])

 18%|█▊        | 919/5000 [12:17:18<50:54:52, 44.91s/it]
                                                        
{'loss': '0.5007', 'grad_norm': '12.84', 'learning_rate': '1.666e-05', 'epoch': '0.1838'}

 18%|█▊        | 919/5000 [12:17:18<50:54:52, 44.91s/it][2026-05-22 04:23:20,233] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 123])
[2026-05-22 04:23:20,949] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])

 18%|█▊        | 920/5000 [12:18:11<53:19:43, 47.05s/it]
                                                        
{'loss': '1.027', 'grad_norm': '10.27', 'learning_rate': '1.665e-05', 'epoch': '0.184'}

 18%|█▊        | 920/5000 [12:18:11<53:19:43, 47.05s/it][2026-05-22 04:24:01,084] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 04:24:02,571] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 18%|█▊        | 921/5000 [12:18:43<48:18:58, 42.64s/it]
                                                        
{'loss': '1.906', 'grad_norm': '28.42', 'learning_rate': '1.665e-05', 'epoch': '0.1842'}

 18%|█▊        | 921/5000 [12:18:43<48:18:58, 42.64s/it][2026-05-22 04:24:31,669] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 169])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 04:24:32,372] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 440])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 18%|█▊        | 922/5000 [12:19:16<45:12:04, 39.90s/it]
                                                        
{'loss': '1.691', 'grad_norm': '30.98', 'learning_rate': '1.664e-05', 'epoch': '0.1844'}

 18%|█▊        | 922/5000 [12:19:16<45:12:04, 39.90s/it][2026-05-22 04:25:06,061] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 04:25:08,079] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 473])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

 18%|█▊        | 923/5000 [12:19:45<41:11:46, 36.38s/it]
                                                        
{'loss': '0.4462', 'grad_norm': '11.13', 'learning_rate': '1.664e-05', 'epoch': '0.1846'}

 18%|█▊        | 923/5000 [12:19:45<41:11:46, 36.38s/it][2026-05-22 04:25:34,340] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])
[2026-05-22 04:25:35,032] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 18%|█▊        | 924/5000 [12:20:13<38:32:38, 34.04s/it]
                                                        
{'loss': '0.381', 'grad_norm': '9.945', 'learning_rate': '1.664e-05', 'epoch': '0.1848'}

 18%|█▊        | 924/5000 [12:20:13<38:32:38, 34.04s/it][2026-05-22 04:26:02,779] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 136])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 04:26:05,039] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 116])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 18%|█▊        | 925/5000 [12:21:03<43:53:10, 38.77s/it]
                                                        
{'loss': '0.3078', 'grad_norm': '9.018', 'learning_rate': '1.663e-05', 'epoch': '0.185'}

 18%|█▊        | 925/5000 [12:21:03<43:53:10, 38.77s/it][2026-05-22 04:26:52,751] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 100])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 04:26:55,361] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 93])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 19%|█▊        | 926/5000 [12:21:52<47:27:10, 41.93s/it]
                                                        
{'loss': '0.245', 'grad_norm': '7.982', 'learning_rate': '1.663e-05', 'epoch': '0.1852'}

 19%|█▊        | 926/5000 [12:21:52<47:27:10, 41.93s/it][2026-05-22 04:28:03,679] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 169])
[2026-05-22 04:28:04,348] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 353])

 19%|█▊        | 927/5000 [12:22:42<50:11:41, 44.37s/it]
                                                        
{'loss': '0.6621', 'grad_norm': '10.65', 'learning_rate': '1.662e-05', 'epoch': '0.1854'}

 19%|█▊        | 927/5000 [12:22:42<50:11:41, 44.37s/it][2026-05-22 04:28:55,589] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 137])
[2026-05-22 04:28:57,202] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 147])

 19%|█▊        | 928/5000 [12:23:34<52:43:50, 46.62s/it]
                                                        
{'loss': '0.945', 'grad_norm': '8.087', 'learning_rate': '1.662e-05', 'epoch': '0.1856'}

 19%|█▊        | 928/5000 [12:23:34<52:43:50, 46.62s/it][2026-05-22 04:29:25,420] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 94])
[2026-05-22 04:29:26,002] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 128])

 19%|█▊        | 929/5000 [12:24:25<54:14:43, 47.97s/it]
                                                        
{'loss': '0.6572', 'grad_norm': '9.354', 'learning_rate': '1.662e-05', 'epoch': '0.1858'}

 19%|█▊        | 929/5000 [12:24:25<54:14:43, 47.97s/it][2026-05-22 04:30:16,878] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 74])
[2026-05-22 04:30:17,117] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 356])

 19%|█▊        | 930/5000 [12:25:17<55:20:26, 48.95s/it]
                                                        
{'loss': '0.3161', 'grad_norm': '5.978', 'learning_rate': '1.661e-05', 'epoch': '0.186'}

 19%|█▊        | 930/5000 [12:25:17<55:20:26, 48.95s/it][2026-05-22 04:31:11,539] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 253])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 04:31:11,790] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 300])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 19%|█▊        | 931/5000 [12:26:02<54:10:09, 47.93s/it]
                                                        
{'loss': '1.322', 'grad_norm': '9.822', 'learning_rate': '1.661e-05', 'epoch': '0.1862'}

 19%|█▊        | 931/5000 [12:26:02<54:10:09, 47.93s/it][2026-05-22 04:31:58,956] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 284])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 04:32:00,484] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 325])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 19%|█▊        | 932/5000 [12:26:51<54:30:09, 48.23s/it]
                                                        
{'loss': '1.55', 'grad_norm': '16.91', 'learning_rate': '1.66e-05', 'epoch': '0.1864'}

 19%|█▊        | 932/5000 [12:26:51<54:30:09, 48.23s/it][2026-05-22 04:32:38,550] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 04:32:39,086] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 329])

 19%|█▊        | 933/5000 [12:27:41<54:56:55, 48.64s/it]
                                                        
{'loss': '0.3008', 'grad_norm': '10.03', 'learning_rate': '1.66e-05', 'epoch': '0.1866'}

 19%|█▊        | 933/5000 [12:27:41<54:56:55, 48.64s/it][2026-05-22 04:33:28,217] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])
[2026-05-22 04:33:28,967] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 19%|█▊        | 934/5000 [12:28:28<54:33:38, 48.31s/it]
                                                        
{'loss': '0.3403', 'grad_norm': '9.833', 'learning_rate': '1.66e-05', 'epoch': '0.1868'}

 19%|█▊        | 934/5000 [12:28:28<54:33:38, 48.31s/it][2026-05-22 04:34:23,091] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 958])
[2026-05-22 04:34:25,914] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 19%|█▊        | 935/5000 [12:29:16<54:23:57, 48.18s/it]
                                                        
{'loss': '0.3566', 'grad_norm': '10.15', 'learning_rate': '1.659e-05', 'epoch': '0.187'}

 19%|█▊        | 935/5000 [12:29:16<54:23:57, 48.18s/it][2026-05-22 04:35:12,197] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 04:35:12,570] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 446])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 19%|█▊        | 936/5000 [12:30:07<55:14:18, 48.93s/it]
                                                        
{'loss': '0.342', 'grad_norm': '9.864', 'learning_rate': '1.659e-05', 'epoch': '0.1872'}

 19%|█▊        | 936/5000 [12:30:07<55:14:18, 48.93s/it][2026-05-22 04:35:54,836] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])
[2026-05-22 04:35:54,874] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 334])

 19%|█▊        | 937/5000 [12:30:37<49:00:29, 43.42s/it]
                                                        
{'loss': '0.3024', 'grad_norm': '10.93', 'learning_rate': '1.658e-05', 'epoch': '0.1874'}

 19%|█▊        | 937/5000 [12:30:37<49:00:29, 43.42s/it][2026-05-22 04:36:25,479] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 303])
[2026-05-22 04:36:26,229] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 467])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])

 19%|█▉        | 938/5000 [12:31:07<44:25:00, 39.36s/it]
                                                        
{'loss': '0.3001', 'grad_norm': '11.54', 'learning_rate': '1.658e-05', 'epoch': '0.1876'}

 19%|█▉        | 938/5000 [12:31:07<44:25:00, 39.36s/it][2026-05-22 04:37:09,205] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 210])
[2026-05-22 04:37:09,489] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 245])

 19%|█▉        | 939/5000 [12:31:43<43:16:11, 38.36s/it]
                                                        
{'loss': '1.519', 'grad_norm': '26.63', 'learning_rate': '1.658e-05', 'epoch': '0.1878'}

 19%|█▉        | 939/5000 [12:31:43<43:16:11, 38.36s/it][2026-05-22 04:37:45,172] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 128])
[2026-05-22 04:37:46,547] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 236])

 19%|█▉        | 940/5000 [12:32:22<43:28:39, 38.55s/it]
                                                        
{'loss': '1.961', 'grad_norm': '34.58', 'learning_rate': '1.657e-05', 'epoch': '0.188'}

 19%|█▉        | 940/5000 [12:32:22<43:28:39, 38.55s/it][2026-05-22 04:38:20,004] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 86])
[2026-05-22 04:38:21,165] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 987])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 240])

 19%|█▉        | 941/5000 [12:33:15<48:17:55, 42.84s/it]
                                                        
{'loss': '0.3564', 'grad_norm': '12.26', 'learning_rate': '1.657e-05', 'epoch': '0.1882'}

 19%|█▉        | 941/5000 [12:33:15<48:17:55, 42.84s/it][2026-05-22 04:39:13,200] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 84])
[2026-05-22 04:39:17,422] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 95])

 19%|█▉        | 942/5000 [12:34:09<51:54:17, 46.05s/it]
                                                        
{'loss': '0.5578', 'grad_norm': '9.758', 'learning_rate': '1.656e-05', 'epoch': '0.1884'}

 19%|█▉        | 942/5000 [12:34:09<51:54:17, 46.05s/it][2026-05-22 04:39:57,975] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 247])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 04:40:00,472] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 299])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 19%|█▉        | 943/5000 [12:34:40<47:00:29, 41.71s/it]
                                                        
{'loss': '0.7007', 'grad_norm': '11.02', 'learning_rate': '1.656e-05', 'epoch': '0.1886'}

 19%|█▉        | 943/5000 [12:34:40<47:00:29, 41.71s/it][2026-05-22 04:40:31,288] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 263])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 04:40:32,316] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 252])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 19%|█▉        | 944/5000 [12:35:12<43:34:17, 38.67s/it]
                                                        
{'loss': '0.07087', 'grad_norm': '5.027', 'learning_rate': '1.656e-05', 'epoch': '0.1888'}

 19%|█▉        | 944/5000 [12:35:12<43:34:17, 38.67s/it][2026-05-22 04:40:59,479] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 291])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 04:40:59,766] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 319])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 19%|█▉        | 945/5000 [12:36:02<47:23:04, 42.07s/it]
                                                        
{'loss': '0.3838', 'grad_norm': '17.11', 'learning_rate': '1.655e-05', 'epoch': '0.189'}

 19%|█▉        | 945/5000 [12:36:02<47:23:04, 42.07s/it][2026-05-22 04:41:49,821] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 346])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 04:41:49,968] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 303])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 19%|█▉        | 946/5000 [12:36:51<49:52:22, 44.29s/it]
                                                        
{'loss': '0.4232', 'grad_norm': '7.77', 'learning_rate': '1.655e-05', 'epoch': '0.1892'}

 19%|█▉        | 946/5000 [12:36:51<49:52:22, 44.29s/it][2026-05-22 04:42:38,317] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 238])
[2026-05-22 04:42:39,400] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 209])

 19%|█▉        | 947/5000 [12:37:37<50:29:27, 44.85s/it]
                                                        
{'loss': '0.1543', 'grad_norm': '6.788', 'learning_rate': '1.654e-05', 'epoch': '0.1894'}

 19%|█▉        | 947/5000 [12:37:37<50:29:27, 44.85s/it][2026-05-22 04:43:25,463] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])
[2026-05-22 04:43:25,641] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 266])

 19%|█▉        | 948/5000 [12:38:23<50:48:18, 45.14s/it]
                                                        
{'loss': '0.1265', 'grad_norm': '9.387', 'learning_rate': '1.654e-05', 'epoch': '0.1896'}

 19%|█▉        | 948/5000 [12:38:23<50:48:18, 45.14s/it][2026-05-22 04:44:11,761] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 106])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 04:44:13,614] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 92])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 19%|█▉        | 949/5000 [12:39:11<51:52:11, 46.10s/it]
                                                        
{'loss': '1.205', 'grad_norm': '9.608', 'learning_rate': '1.653e-05', 'epoch': '0.1898'}

 19%|█▉        | 949/5000 [12:39:11<51:52:11, 46.10s/it][2026-05-22 04:44:59,931] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 115])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 04:45:00,433] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 108])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 19%|█▉        | 950/5000 [12:40:00<52:43:28, 46.87s/it]
                                                        
{'loss': '0.487', 'grad_norm': '6.797', 'learning_rate': '1.653e-05', 'epoch': '0.19'}

 19%|█▉        | 950/5000 [12:40:00<52:43:28, 46.87s/it][2026-05-22 04:45:53,440] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 977])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 04:45:54,800] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 19%|█▉        | 951/5000 [12:40:34<48:11:16, 42.84s/it]
                                                        
{'loss': '0.5907', 'grad_norm': '11.43', 'learning_rate': '1.653e-05', 'epoch': '0.1902'}

 19%|█▉        | 951/5000 [12:40:34<48:11:16, 42.84s/it][2026-05-22 04:46:27,754] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 04:46:27,792] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 19%|█▉        | 952/5000 [12:41:10<45:54:20, 40.83s/it]
                                                        
{'loss': '0.0575', 'grad_norm': '5.133', 'learning_rate': '1.652e-05', 'epoch': '0.1904'}

 19%|█▉        | 952/5000 [12:41:10<45:54:20, 40.83s/it][2026-05-22 04:46:58,576] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 04:47:00,497] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 19%|█▉        | 953/5000 [12:41:37<41:28:48, 36.90s/it]
                                                        
{'loss': '0.505', 'grad_norm': '12.59', 'learning_rate': '1.652e-05', 'epoch': '0.1906'}

 19%|█▉        | 953/5000 [12:41:37<41:28:48, 36.90s/it][2026-05-22 04:47:27,510] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 197])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 04:47:27,973] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 506])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])

 19%|█▉        | 954/5000 [12:42:09<39:31:48, 35.17s/it]
                                                        
{'loss': '0.1293', 'grad_norm': '8.203', 'learning_rate': '1.651e-05', 'epoch': '0.1908'}

 19%|█▉        | 954/5000 [12:42:09<39:31:48, 35.17s/it][2026-05-22 04:47:59,633] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 04:48:01,847] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 19%|█▉        | 955/5000 [12:43:00<45:00:30, 40.06s/it]
                                                        
{'loss': '0.9935', 'grad_norm': '16.07', 'learning_rate': '1.651e-05', 'epoch': '0.191'}

 19%|█▉        | 955/5000 [12:43:00<45:00:30, 40.06s/it][2026-05-22 04:48:52,625] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])
[2026-05-22 04:48:55,057] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 19%|█▉        | 956/5000 [12:43:52<49:08:28, 43.75s/it]
                                                        
{'loss': '0.9283', 'grad_norm': '12.47', 'learning_rate': '1.651e-05', 'epoch': '0.1912'}

 19%|█▉        | 956/5000 [12:43:52<49:08:28, 43.75s/it][2026-05-22 04:49:52,026] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 94])
[2026-05-22 04:49:54,464] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 140])

 19%|█▉        | 957/5000 [12:45:04<58:29:32, 52.08s/it]
                                                        
{'loss': '0.4253', 'grad_norm': '11.21', 'learning_rate': '1.65e-05', 'epoch': '0.1914'}

 19%|█▉        | 957/5000 [12:45:04<58:29:32, 52.08s/it][2026-05-22 04:51:03,158] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 986])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 100])
[2026-05-22 04:51:03,439] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 223])

 19%|█▉        | 958/5000 [12:46:14<64:35:15, 57.52s/it]
                                                        
{'loss': '0.37', 'grad_norm': '11.12', 'learning_rate': '1.65e-05', 'epoch': '0.1916'}

 19%|█▉        | 958/5000 [12:46:14<64:35:15, 57.52s/it][2026-05-22 04:52:03,370] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 74])
[2026-05-22 04:52:04,389] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 58])

 19%|█▉        | 959/5000 [12:47:01<61:03:50, 54.40s/it]
                                                        
{'loss': '0.2996', 'grad_norm': '7.703', 'learning_rate': '1.649e-05', 'epoch': '0.1918'}

 19%|█▉        | 959/5000 [12:47:01<61:03:50, 54.40s/it][2026-05-22 04:52:51,415] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 978])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 59])
[2026-05-22 04:52:51,796] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 978])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 84])

 19%|█▉        | 960/5000 [12:47:53<60:00:04, 53.47s/it]
                                                        
{'loss': '0.1259', 'grad_norm': '7.195', 'learning_rate': '1.649e-05', 'epoch': '0.192'}

 19%|█▉        | 960/5000 [12:47:53<60:00:04, 53.47s/it][2026-05-22 04:53:41,866] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 17])
[2026-05-22 04:53:42,307] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 23])

 19%|█▉        | 961/5000 [12:49:02<65:23:56, 58.29s/it]
                                                        
{'loss': '0.7969', 'grad_norm': '9.512', 'learning_rate': '1.649e-05', 'epoch': '0.1922'}

 19%|█▉        | 961/5000 [12:49:02<65:23:56, 58.29s/it][2026-05-22 04:54:51,518] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 12])
[2026-05-22 04:54:52,158] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 976])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 14])

 19%|█▉        | 962/5000 [12:50:12<69:18:32, 61.79s/it]
                                                        
{'loss': '0.5877', 'grad_norm': '9.612', 'learning_rate': '1.648e-05', 'epoch': '0.1924'}

 19%|█▉        | 962/5000 [12:50:12<69:18:32, 61.79s/it][2026-05-22 04:55:59,905] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 18])
[2026-05-22 04:56:00,424] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 44])

 19%|█▉        | 963/5000 [12:51:00<64:34:48, 57.59s/it]
                                                        
{'loss': '2.168', 'grad_norm': '13.07', 'learning_rate': '1.648e-05', 'epoch': '0.1926'}

 19%|█▉        | 963/5000 [12:51:00<64:34:48, 57.59s/it][2026-05-22 04:56:47,321] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 19])
[2026-05-22 04:56:47,708] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 21])

 19%|█▉        | 964/5000 [12:51:47<61:10:06, 54.56s/it]
                                                        
{'loss': '2.052', 'grad_norm': '13.34', 'learning_rate': '1.647e-05', 'epoch': '0.1928'}

 19%|█▉        | 964/5000 [12:51:47<61:10:06, 54.56s/it][2026-05-22 04:57:39,873] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 108])
[2026-05-22 04:57:40,126] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 179])

 19%|█▉        | 965/5000 [12:52:34<58:38:57, 52.33s/it]
                                                        
{'loss': '0.1189', 'grad_norm': '6.807', 'learning_rate': '1.647e-05', 'epoch': '0.193'}

 19%|█▉        | 965/5000 [12:52:34<58:38:57, 52.33s/it][2026-05-22 04:58:26,700] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1012])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 290])
[2026-05-22 04:58:26,756] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 134])

 19%|█▉        | 966/5000 [12:53:23<57:21:27, 51.19s/it]
                                                        
{'loss': '0.09799', 'grad_norm': '7.75', 'learning_rate': '1.647e-05', 'epoch': '0.1932'}

 19%|█▉        | 966/5000 [12:53:23<57:21:27, 51.19s/it][2026-05-22 04:59:20,023] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])
[2026-05-22 04:59:25,352] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 19%|█▉        | 967/5000 [12:54:09<55:42:37, 49.73s/it]
                                                        
{'loss': '3.698', 'grad_norm': '7.558', 'learning_rate': '1.646e-05', 'epoch': '0.1934'}

 19%|█▉        | 967/5000 [12:54:09<55:42:37, 49.73s/it][2026-05-22 05:00:10,546] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 05:00:11,460] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 19%|█▉        | 968/5000 [12:54:55<54:26:13, 48.60s/it]
                                                        
{'loss': '3.672', 'grad_norm': '9.673', 'learning_rate': '1.646e-05', 'epoch': '0.1936'}

 19%|█▉        | 968/5000 [12:54:55<54:26:13, 48.60s/it][2026-05-22 05:00:50,570] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 123])
[2026-05-22 05:00:52,906] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 201])

 19%|█▉        | 969/5000 [12:55:39<52:51:00, 47.20s/it]
                                                        
{'loss': '1.577', 'grad_norm': '11.87', 'learning_rate': '1.645e-05', 'epoch': '0.1938'}

 19%|█▉        | 969/5000 [12:55:39<52:51:00, 47.20s/it][2026-05-22 05:01:34,092] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 209])
[2026-05-22 05:01:36,518] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 116])

 19%|█▉        | 970/5000 [12:56:23<51:41:49, 46.18s/it]
                                                        
{'loss': '1.491', 'grad_norm': '9.945', 'learning_rate': '1.645e-05', 'epoch': '0.194'}

 19%|█▉        | 970/5000 [12:56:23<51:41:49, 46.18s/it][2026-05-22 05:02:16,330] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 98])
[2026-05-22 05:02:16,497] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 99])

 19%|█▉        | 971/5000 [12:57:17<54:23:05, 48.59s/it]
                                                        
{'loss': '1.406', 'grad_norm': '8.902', 'learning_rate': '1.644e-05', 'epoch': '0.1942'}

 19%|█▉        | 971/5000 [12:57:17<54:23:05, 48.59s/it][2026-05-22 05:03:11,796] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 186])
[2026-05-22 05:03:11,986] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 133])

 19%|█▉        | 972/5000 [12:58:10<55:46:17, 49.85s/it]
                                                        
{'loss': '1.432', 'grad_norm': '8.481', 'learning_rate': '1.644e-05', 'epoch': '0.1944'}

 19%|█▉        | 972/5000 [12:58:10<55:46:17, 49.85s/it][2026-05-22 05:04:06,808] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 118])
[2026-05-22 05:04:13,401] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 987])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 203])

 19%|█▉        | 973/5000 [12:59:20<62:26:05, 55.81s/it]
                                                        
{'loss': '0.5925', 'grad_norm': '8.42', 'learning_rate': '1.644e-05', 'epoch': '0.1946'}

 19%|█▉        | 973/5000 [12:59:20<62:26:05, 55.81s/it][2026-05-22 05:05:18,139] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 137])
[2026-05-22 05:05:18,234] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 121])

 19%|█▉        | 974/5000 [13:00:30<67:11:25, 60.08s/it]
                                                        
{'loss': '1.253', 'grad_norm': '9.28', 'learning_rate': '1.643e-05', 'epoch': '0.1948'}

 19%|█▉        | 974/5000 [13:00:30<67:11:25, 60.08s/it][2026-05-22 05:06:19,704] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 125])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 05:06:22,087] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 117])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 20%|█▉        | 975/5000 [13:01:17<62:55:54, 56.29s/it]
                                                        
{'loss': '2.43', 'grad_norm': '11.59', 'learning_rate': '1.643e-05', 'epoch': '0.195'}

 20%|█▉        | 975/5000 [13:01:17<62:55:54, 56.29s/it][2026-05-22 05:07:07,564] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 173])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 05:07:07,620] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 133])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 20%|█▉        | 976/5000 [13:02:04<59:46:03, 53.47s/it]
                                                        
{'loss': '2.4', 'grad_norm': '12.08', 'learning_rate': '1.642e-05', 'epoch': '0.1952'}

 20%|█▉        | 976/5000 [13:02:04<59:46:03, 53.47s/it][2026-05-22 05:07:54,845] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 256])
[2026-05-22 05:07:55,958] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 124])

 20%|█▉        | 977/5000 [13:02:54<58:42:03, 52.53s/it]
                                                        
{'loss': '0.7741', 'grad_norm': '15.53', 'learning_rate': '1.642e-05', 'epoch': '0.1954'}

 20%|█▉        | 977/5000 [13:02:54<58:42:03, 52.53s/it][2026-05-22 05:08:45,177] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 166])
[2026-05-22 05:08:45,357] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 169])

 20%|█▉        | 978/5000 [13:03:45<58:01:51, 51.94s/it]
                                                        
{'loss': '0.838', 'grad_norm': '21.24', 'learning_rate': '1.642e-05', 'epoch': '0.1956'}

 20%|█▉        | 978/5000 [13:03:45<58:01:51, 51.94s/it][2026-05-22 05:10:07,431] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])
[2026-05-22 05:10:08,649] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 155])

 20%|█▉        | 979/5000 [13:04:36<57:36:16, 51.57s/it]
                                                        
{'loss': '2.8', 'grad_norm': '45.15', 'learning_rate': '1.641e-05', 'epoch': '0.1958'}

 20%|█▉        | 979/5000 [13:04:36<57:36:16, 51.57s/it][2026-05-22 05:10:56,223] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 200])
[2026-05-22 05:11:03,113] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 170])

 20%|█▉        | 980/5000 [13:05:26<57:02:03, 51.08s/it]
                                                        
{'loss': '2.602', 'grad_norm': '12.8', 'learning_rate': '1.641e-05', 'epoch': '0.196'}

 20%|█▉        | 980/5000 [13:05:26<57:02:03, 51.08s/it][2026-05-22 05:11:22,706] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 67])
[2026-05-22 05:11:24,315] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 46])

 20%|█▉        | 981/5000 [13:06:15<56:24:10, 50.52s/it]
                                                        
{'loss': '0.8426', 'grad_norm': '12.49', 'learning_rate': '1.64e-05', 'epoch': '0.1962'}

 20%|█▉        | 981/5000 [13:06:15<56:24:10, 50.52s/it][2026-05-22 05:12:12,012] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 128])
[2026-05-22 05:12:13,664] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 122])

 20%|█▉        | 982/5000 [13:07:02<55:13:50, 49.48s/it]
                                                        
{'loss': '1.266', 'grad_norm': '11.24', 'learning_rate': '1.64e-05', 'epoch': '0.1964'}

 20%|█▉        | 982/5000 [13:07:02<55:13:50, 49.48s/it][2026-05-22 05:12:53,189] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 05:12:53,371] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 20%|█▉        | 983/5000 [13:07:53<55:37:40, 49.85s/it]
                                                        
{'loss': '0.4855', 'grad_norm': '13.94', 'learning_rate': '1.64e-05', 'epoch': '0.1966'}

 20%|█▉        | 983/5000 [13:07:53<55:37:40, 49.85s/it][2026-05-22 05:13:45,333] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1021])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 05:13:49,145] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 20%|█▉        | 984/5000 [13:08:45<56:30:03, 50.65s/it]
                                                        
{'loss': '0.3952', 'grad_norm': '12.57', 'learning_rate': '1.639e-05', 'epoch': '0.1968'}

 20%|█▉        | 984/5000 [13:08:45<56:30:03, 50.65s/it][2026-05-22 05:14:32,386] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 05:14:32,517] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 20%|█▉        | 985/5000 [13:09:36<56:30:10, 50.66s/it]
                                                        
{'loss': '0.4149', 'grad_norm': '12.43', 'learning_rate': '1.639e-05', 'epoch': '0.197'}

 20%|█▉        | 985/5000 [13:09:36<56:30:10, 50.66s/it][2026-05-22 05:15:23,277] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 05:15:23,334] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 20%|█▉        | 986/5000 [13:10:26<56:17:13, 50.48s/it]
                                                        
{'loss': '0.5537', 'grad_norm': '12.36', 'learning_rate': '1.638e-05', 'epoch': '0.1972'}

 20%|█▉        | 986/5000 [13:10:26<56:17:13, 50.48s/it][2026-05-22 05:16:26,358] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 94])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])
[2026-05-22 05:16:33,936] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 94])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 20%|█▉        | 987/5000 [13:11:14<55:36:08, 49.88s/it]
                                                        
{'loss': '0.5329', 'grad_norm': '12.69', 'learning_rate': '1.638e-05', 'epoch': '0.1974'}

 20%|█▉        | 987/5000 [13:11:14<55:36:08, 49.88s/it][2026-05-22 05:17:14,827] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 100])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 05:17:16,148] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 95])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 20%|█▉        | 988/5000 [13:12:02<54:58:19, 49.33s/it]
                                                        
{'loss': '0.4987', 'grad_norm': '12.42', 'learning_rate': '1.638e-05', 'epoch': '0.1976'}

 20%|█▉        | 988/5000 [13:12:02<54:58:19, 49.33s/it][2026-05-22 05:17:55,956] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 05:17:56,059] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])

 20%|█▉        | 989/5000 [13:12:59<57:14:17, 51.37s/it]
                                                        
{'loss': '0.9303', 'grad_norm': '21.37', 'learning_rate': '1.637e-05', 'epoch': '0.1978'}

 20%|█▉        | 989/5000 [13:12:59<57:14:17, 51.37s/it][2026-05-22 05:18:52,368] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 05:18:54,657] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 20%|█▉        | 990/5000 [13:13:54<58:31:10, 52.54s/it]
                                                        
{'loss': '0.6141', 'grad_norm': '17.46', 'learning_rate': '1.637e-05', 'epoch': '0.198'}

 20%|█▉        | 990/5000 [13:13:54<58:31:10, 52.54s/it][2026-05-22 05:20:01,534] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 265])
[2026-05-22 05:20:02,186] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 242])

 20%|█▉        | 991/5000 [13:14:44<57:37:24, 51.74s/it]
                                                        
{'loss': '0.5383', 'grad_norm': '13.78', 'learning_rate': '1.636e-05', 'epoch': '0.1982'}

 20%|█▉        | 991/5000 [13:14:44<57:37:24, 51.74s/it][2026-05-22 05:20:52,420] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])
[2026-05-22 05:20:54,555] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 301])

 20%|█▉        | 992/5000 [13:15:34<57:15:12, 51.43s/it]
                                                        
{'loss': '0.3811', 'grad_norm': '11.86', 'learning_rate': '1.636e-05', 'epoch': '0.1984'}

 20%|█▉        | 992/5000 [13:15:34<57:15:12, 51.43s/it][2026-05-22 05:21:21,861] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 263])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 213])
[2026-05-22 05:21:22,669] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 450])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 525])

 20%|█▉        | 993/5000 [13:16:23<56:16:40, 50.56s/it]
                                                        
{'loss': '1.638', 'grad_norm': '13.55', 'learning_rate': '1.636e-05', 'epoch': '0.1986'}

 20%|█▉        | 993/5000 [13:16:23<56:16:40, 50.56s/it][2026-05-22 05:22:11,129] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 05:22:11,322] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 20%|█▉        | 994/5000 [13:17:11<55:23:37, 49.78s/it]
                                                        
{'loss': '1.083', 'grad_norm': '13.63', 'learning_rate': '1.635e-05', 'epoch': '0.1988'}

 20%|█▉        | 994/5000 [13:17:11<55:23:37, 49.78s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 05:23:00,247] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 299])
[2026-05-22 05:23:00,500] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 224])

 20%|█▉        | 995/5000 [13:18:21<62:12:00, 55.91s/it]
                                                        
{'loss': '0.7082', 'grad_norm': '13.76', 'learning_rate': '1.635e-05', 'epoch': '0.199'}

 20%|█▉        | 995/5000 [13:18:21<62:12:00, 55.91s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 05:24:10,437] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 329])
[2026-05-22 05:24:10,658] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])

 20%|█▉        | 996/5000 [13:19:32<67:04:01, 60.30s/it]
                                                        
{'loss': '0.4434', 'grad_norm': '11.2', 'learning_rate': '1.634e-05', 'epoch': '0.1992'}

 20%|█▉        | 996/5000 [13:19:32<67:04:01, 60.30s/it][2026-05-22 05:25:28,360] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 55])
[2026-05-22 05:25:30,533] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 987])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 157])

 20%|█▉        | 997/5000 [13:20:20<63:07:01, 56.76s/it]
                                                        
{'loss': '0.4624', 'grad_norm': '9.911', 'learning_rate': '1.634e-05', 'epoch': '0.1994'}

 20%|█▉        | 997/5000 [13:20:20<63:07:01, 56.76s/it][2026-05-22 05:26:14,383] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 53])
[2026-05-22 05:26:17,497] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])

 20%|█▉        | 998/5000 [13:21:10<60:50:43, 54.73s/it]
                                                        
{'loss': '0.207', 'grad_norm': '6.29', 'learning_rate': '1.633e-05', 'epoch': '0.1996'}

 20%|█▉        | 998/5000 [13:21:10<60:50:43, 54.73s/it][2026-05-22 05:27:10,998] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 05:27:11,522] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 20%|█▉        | 999/5000 [13:21:47<54:45:33, 49.27s/it]
                                                        
{'loss': '0.4178', 'grad_norm': '8.491', 'learning_rate': '1.633e-05', 'epoch': '0.1998'}

 20%|█▉        | 999/5000 [13:21:47<54:45:33, 49.27s/it][2026-05-22 05:27:47,260] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 05:27:48,901] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 20%|██        | 1000/5000 [13:22:22<50:15:19, 45.23s/it]
                                                         
{'loss': '0.04209', 'grad_norm': '6.667', 'learning_rate': '1.633e-05', 'epoch': '0.2'}

 20%|██        | 1000/5000 [13:22:22<50:15:19, 45.23s/it][2026-05-22 05:28:07,639] INFO [src.utils.basic_utils:19] Saving model to /workspace/VLM2Vec/outputs/Qwen3.5-08b-continued-mixed-alltasks/checkpoint-1000
[2026-05-22 05:28:07,927] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-22 05:28:07,936] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-22 05:28:11,704] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 120])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 05:28:12,582] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 96])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 20%|██        | 1001/5000 [13:23:10<51:00:47, 45.92s/it]
                                                         
{'loss': '0.1679', 'grad_norm': '7.661', 'learning_rate': '1.632e-05', 'epoch': '0.2002'}

 20%|██        | 1001/5000 [13:23:10<51:00:47, 45.92s/it][2026-05-22 05:28:58,208] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 103])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 05:29:00,176] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 75])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 20%|██        | 1002/5000 [13:23:58<51:43:44, 46.58s/it]
                                                         
{'loss': '0.1941', 'grad_norm': '10.02', 'learning_rate': '1.632e-05', 'epoch': '0.2004'}

 20%|██        | 1002/5000 [13:23:58<51:43:44, 46.58s/it][2026-05-22 05:29:59,365] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 155])
[2026-05-22 05:29:59,754] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 154])

 20%|██        | 1003/5000 [13:24:44<51:36:32, 46.48s/it]
                                                         
{'loss': '0.3411', 'grad_norm': '9.231', 'learning_rate': '1.631e-05', 'epoch': '0.2006'}

 20%|██        | 1003/5000 [13:24:44<51:36:32, 46.48s/it][2026-05-22 05:30:43,613] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 224])
[2026-05-22 05:30:47,935] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 204])

 20%|██        | 1004/5000 [13:25:31<51:45:38, 46.63s/it]
                                                         
{'loss': '0.3611', 'grad_norm': '8.615', 'learning_rate': '1.631e-05', 'epoch': '0.2008'}

 20%|██        | 1004/5000 [13:25:31<51:45:38, 46.63s/it][2026-05-22 05:31:17,918] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 221])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 44])
[2026-05-22 05:31:17,930] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 62])

 20%|██        | 1005/5000 [13:26:41<59:20:02, 53.47s/it]
                                                         
{'loss': '0.4259', 'grad_norm': '12.73', 'learning_rate': '1.631e-05', 'epoch': '0.201'}

 20%|██        | 1005/5000 [13:26:41<59:20:02, 53.47s/it][2026-05-22 05:32:27,355] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 178])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 56])
[2026-05-22 05:32:27,898] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 430])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 74])

 20%|██        | 1006/5000 [13:27:50<64:38:38, 58.27s/it]
                                                         
{'loss': '0.5811', 'grad_norm': '8.248', 'learning_rate': '1.63e-05', 'epoch': '0.2012'}

 20%|██        | 1006/5000 [13:27:50<64:38:38, 58.27s/it][2026-05-22 05:34:02,975] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 250])
[2026-05-22 05:34:03,376] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 275])

 20%|██        | 1007/5000 [13:28:36<60:26:47, 54.50s/it]
                                                         
{'loss': '0.4514', 'grad_norm': '12.8', 'learning_rate': '1.63e-05', 'epoch': '0.2014'}

 20%|██        | 1007/5000 [13:28:36<60:26:47, 54.50s/it][2026-05-22 05:34:46,304] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 251])
[2026-05-22 05:34:50,012] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 266])

 20%|██        | 1008/5000 [13:29:24<58:11:56, 52.48s/it]
                                                         
{'loss': '0.4331', 'grad_norm': '11.19', 'learning_rate': '1.629e-05', 'epoch': '0.2016'}

 20%|██        | 1008/5000 [13:29:24<58:11:56, 52.48s/it][2026-05-22 05:35:11,206] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 05:35:11,569] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 20%|██        | 1009/5000 [13:30:11<56:31:30, 50.99s/it]
                                                         
{'loss': '0.2396', 'grad_norm': '6.393', 'learning_rate': '1.629e-05', 'epoch': '0.2018'}

 20%|██        | 1009/5000 [13:30:11<56:31:30, 50.99s/it][2026-05-22 05:35:58,974] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])
[2026-05-22 05:35:59,233] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 20%|██        | 1010/5000 [13:31:00<55:40:37, 50.23s/it]
                                                         
{'loss': '0.3014', 'grad_norm': '8.174', 'learning_rate': '1.629e-05', 'epoch': '0.202'}

 20%|██        | 1010/5000 [13:31:00<55:40:37, 50.23s/it][2026-05-22 05:36:47,672] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 668])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 8])
[2026-05-22 05:36:47,902] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 791])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 12])

 20%|██        | 1011/5000 [13:31:23<46:44:37, 42.19s/it]
                                                         
{'loss': '2.324', 'grad_norm': '6.866', 'learning_rate': '1.628e-05', 'epoch': '0.2022'}

 20%|██        | 1011/5000 [13:31:23<46:44:37, 42.19s/it][2026-05-22 05:37:11,139] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 668])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 6])
[2026-05-22 05:37:11,191] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 668])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 8])

 20%|██        | 1012/5000 [13:31:49<41:27:44, 37.43s/it]
                                                         
{'loss': '2.283', 'grad_norm': '8.66', 'learning_rate': '1.628e-05', 'epoch': '0.2024'}

 20%|██        | 1012/5000 [13:31:49<41:27:44, 37.43s/it][2026-05-22 05:37:36,594] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])
[2026-05-22 05:37:36,677] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 59])

 20%|██        | 1013/5000 [13:32:37<44:58:37, 40.61s/it]
                                                         
{'loss': '0.459', 'grad_norm': '8.958', 'learning_rate': '1.627e-05', 'epoch': '0.2026'}

 20%|██        | 1013/5000 [13:32:37<44:58:37, 40.61s/it][2026-05-22 05:38:24,570] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 63])
[2026-05-22 05:38:25,694] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 87])

 20%|██        | 1014/5000 [13:33:26<47:31:37, 42.92s/it]
                                                         
{'loss': '0.2606', 'grad_norm': '5.629', 'learning_rate': '1.627e-05', 'epoch': '0.2028'}

 20%|██        | 1014/5000 [13:33:26<47:31:37, 42.92s/it][2026-05-22 05:39:13,294] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 05:39:13,391] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 20%|██        | 1015/5000 [13:33:56<43:17:21, 39.11s/it]
                                                         
{'loss': '0.4239', 'grad_norm': '11.45', 'learning_rate': '1.627e-05', 'epoch': '0.203'}

 20%|██        | 1015/5000 [13:33:56<43:17:21, 39.11s/it][2026-05-22 05:39:43,375] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 848])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 05:39:43,381] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 20%|██        | 1016/5000 [13:34:26<40:17:41, 36.41s/it]
                                                         
{'loss': '0.2537', 'grad_norm': '9.739', 'learning_rate': '1.626e-05', 'epoch': '0.2032'}

 20%|██        | 1016/5000 [13:34:26<40:17:41, 36.41s/it][2026-05-22 05:40:14,006] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 682])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 05:40:14,097] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 713])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 111])

 20%|██        | 1017/5000 [13:35:03<40:30:37, 36.61s/it]
                                                         
{'loss': '3.954', 'grad_norm': '10.83', 'learning_rate': '1.626e-05', 'epoch': '0.2034'}

 20%|██        | 1017/5000 [13:35:03<40:30:37, 36.61s/it][2026-05-22 05:40:51,089] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 647])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 05:40:51,148] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 676])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 20%|██        | 1018/5000 [13:35:36<39:17:12, 35.52s/it]
                                                         
{'loss': '3.951', 'grad_norm': '8.783', 'learning_rate': '1.625e-05', 'epoch': '0.2036'}

 20%|██        | 1018/5000 [13:35:36<39:17:12, 35.52s/it][2026-05-22 05:41:28,467] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1016])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 05:41:30,147] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 20%|██        | 1019/5000 [13:36:05<37:08:34, 33.59s/it]
                                                         
{'loss': '0.3471', 'grad_norm': '12.06', 'learning_rate': '1.625e-05', 'epoch': '0.2038'}

 20%|██        | 1019/5000 [13:36:05<37:08:34, 33.59s/it][2026-05-22 05:41:57,360] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1018])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 618])
[2026-05-22 05:41:59,860] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 20%|██        | 1020/5000 [13:36:38<36:43:42, 33.22s/it]
                                                         
{'loss': '0.4075', 'grad_norm': '10.15', 'learning_rate': '1.624e-05', 'epoch': '0.204'}

 20%|██        | 1020/5000 [13:36:38<36:43:42, 33.22s/it][2026-05-22 05:42:27,317] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 140])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-22 05:42:30,061] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 148])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 20%|██        | 1021/5000 [13:37:22<40:28:02, 36.61s/it]
                                                         
{'loss': '0.629', 'grad_norm': '16.93', 'learning_rate': '1.624e-05', 'epoch': '0.2042'}

 20%|██        | 1021/5000 [13:37:22<40:28:02, 36.61s/it][2026-05-22 05:43:11,408] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 140])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 05:43:12,451] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 132])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 20%|██        | 1022/5000 [13:38:06<42:55:53, 38.85s/it]
                                                         
{'loss': '0.4677', 'grad_norm': '14.44', 'learning_rate': '1.624e-05', 'epoch': '0.2044'}

 20%|██        | 1022/5000 [13:38:06<42:55:53, 38.85s/it][2026-05-22 05:43:59,514] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 104])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 05:44:05,538] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 78])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 20%|██        | 1023/5000 [13:38:44<42:32:51, 38.51s/it]
                                                         
{'loss': '2.189', 'grad_norm': '16.52', 'learning_rate': '1.623e-05', 'epoch': '0.2046'}

 20%|██        | 1023/5000 [13:38:44<42:32:51, 38.51s/it][2026-05-22 05:44:38,483] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 100])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 05:44:39,807] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 106])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

 20%|██        | 1024/5000 [13:39:26<43:46:39, 39.64s/it]
                                                         
{'loss': '2.174', 'grad_norm': '16.47', 'learning_rate': '1.623e-05', 'epoch': '0.2048'}

 20%|██        | 1024/5000 [13:39:26<43:46:39, 39.64s/it][2026-05-22 05:45:16,939] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 284])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 05:45:18,167] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])

 20%|██        | 1025/5000 [13:40:37<54:07:37, 49.02s/it]
                                                         
{'loss': '0.9337', 'grad_norm': '18.39', 'learning_rate': '1.622e-05', 'epoch': '0.205'}

 20%|██        | 1025/5000 [13:40:37<54:07:37, 49.02s/it][2026-05-22 05:46:28,738] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 296])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 972])
[2026-05-22 05:46:29,947] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 21%|██        | 1026/5000 [13:41:48<61:14:24, 55.48s/it]
                                                         
{'loss': '1.117', 'grad_norm': '24.32', 'learning_rate': '1.622e-05', 'epoch': '0.2052'}

 21%|██        | 1026/5000 [13:41:48<61:14:24, 55.48s/it][2026-05-22 05:47:35,547] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])
[2026-05-22 05:47:35,741] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 958])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 21%|██        | 1027/5000 [13:42:33<57:57:40, 52.52s/it]
                                                         
{'loss': '0.8', 'grad_norm': '9.67', 'learning_rate': '1.622e-05', 'epoch': '0.2054'}

 21%|██        | 1027/5000 [13:42:33<57:57:40, 52.52s/it][2026-05-22 05:48:21,387] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 588])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 05:48:21,431] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 958])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 21%|██        | 1028/5000 [13:43:19<55:32:26, 50.34s/it]
                                                         
{'loss': '0.6505', 'grad_norm': '11.25', 'learning_rate': '1.621e-05', 'epoch': '0.2056'}

 21%|██        | 1028/5000 [13:43:19<55:32:26, 50.34s/it][2026-05-22 05:49:06,316] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 251])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 808])
[2026-05-22 05:49:08,002] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 261])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 21%|██        | 1029/5000 [13:44:08<55:04:15, 49.93s/it]
                                                         
{'loss': '1.244', 'grad_norm': '9.337', 'learning_rate': '1.621e-05', 'epoch': '0.2058'}

 21%|██        | 1029/5000 [13:44:08<55:04:15, 49.93s/it][2026-05-22 05:49:56,747] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 309])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 05:49:58,837] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 309])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 21%|██        | 1030/5000 [13:44:57<54:46:57, 49.68s/it]
                                                         
{'loss': '1.339', 'grad_norm': '11.8', 'learning_rate': '1.62e-05', 'epoch': '0.206'}

 21%|██        | 1030/5000 [13:44:57<54:46:57, 49.68s/it][2026-05-22 05:50:46,669] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 265])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 05:50:47,895] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 21%|██        | 1031/5000 [13:45:44<53:54:12, 48.89s/it]
                                                         
{'loss': '0.3804', 'grad_norm': '10.88', 'learning_rate': '1.62e-05', 'epoch': '0.2062'}

 21%|██        | 1031/5000 [13:45:44<53:54:12, 48.89s/it][2026-05-22 05:51:34,099] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 05:51:34,948] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 21%|██        | 1032/5000 [13:46:31<53:12:47, 48.28s/it]
                                                         
{'loss': '0.5007', 'grad_norm': '6.277', 'learning_rate': '1.62e-05', 'epoch': '0.2064'}

 21%|██        | 1032/5000 [13:46:31<53:12:47, 48.28s/it][2026-05-22 05:52:17,835] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])
[2026-05-22 05:52:18,731] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 21%|██        | 1033/5000 [13:47:18<52:56:28, 48.04s/it]
                                                         
{'loss': '0.5273', 'grad_norm': '12.41', 'learning_rate': '1.619e-05', 'epoch': '0.2066'}

 21%|██        | 1033/5000 [13:47:18<52:56:28, 48.04s/it][2026-05-22 05:53:06,107] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 05:53:06,274] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 21%|██        | 1034/5000 [13:48:04<52:07:16, 47.31s/it]
                                                         
{'loss': '0.3948', 'grad_norm': '11.8', 'learning_rate': '1.619e-05', 'epoch': '0.2068'}

 21%|██        | 1034/5000 [13:48:04<52:07:16, 47.31s/it][2026-05-22 05:53:50,983] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 232])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 05:53:51,109] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 221])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 21%|██        | 1035/5000 [13:48:51<52:06:40, 47.31s/it]
                                                         
{'loss': '0.8456', 'grad_norm': '11.12', 'learning_rate': '1.618e-05', 'epoch': '0.207'}

 21%|██        | 1035/5000 [13:48:51<52:06:40, 47.31s/it][2026-05-22 05:54:38,404] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 170])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 05:54:38,814] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 507])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 21%|██        | 1036/5000 [13:49:39<52:17:17, 47.49s/it]
                                                         
{'loss': '0.9996', 'grad_norm': '8.685', 'learning_rate': '1.618e-05', 'epoch': '0.2072'}

 21%|██        | 1036/5000 [13:49:39<52:17:17, 47.49s/it][2026-05-22 05:55:30,092] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 241])
[2026-05-22 05:55:30,542] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 251])

 21%|██        | 1037/5000 [13:50:11<47:08:28, 42.82s/it]
                                                         
{'loss': '0.3073', 'grad_norm': '9.325', 'learning_rate': '1.618e-05', 'epoch': '0.2074'}

 21%|██        | 1037/5000 [13:50:11<47:08:28, 42.82s/it][2026-05-22 05:56:02,036] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 198])
[2026-05-22 05:56:02,709] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 129])

 21%|██        | 1038/5000 [13:50:44<43:51:57, 39.86s/it]
                                                         
{'loss': '0.2481', 'grad_norm': '9.923', 'learning_rate': '1.617e-05', 'epoch': '0.2076'}

 21%|██        | 1038/5000 [13:50:44<43:51:57, 39.86s/it][2026-05-22 05:56:46,535] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 237])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 05:56:46,999] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 21%|██        | 1039/5000 [13:51:18<41:58:52, 38.16s/it]
                                                         
{'loss': '0.3961', 'grad_norm': '8.503', 'learning_rate': '1.617e-05', 'epoch': '0.2078'}

 21%|██        | 1039/5000 [13:51:18<41:58:52, 38.16s/it][2026-05-22 05:57:15,708] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 197])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 05:57:21,033] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 501])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 21%|██        | 1040/5000 [13:51:53<41:06:32, 37.37s/it]
                                                         
{'loss': '0.02374', 'grad_norm': '3.021', 'learning_rate': '1.616e-05', 'epoch': '0.208'}

 21%|██        | 1040/5000 [13:51:53<41:06:32, 37.37s/it][2026-05-22 05:57:44,892] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 113])
[2026-05-22 05:57:44,973] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 137])

 21%|██        | 1041/5000 [13:52:20<37:23:43, 34.00s/it]
                                                         
{'loss': '0.2751', 'grad_norm': '6.566', 'learning_rate': '1.616e-05', 'epoch': '0.2082'}

 21%|██        | 1041/5000 [13:52:20<37:23:43, 34.00s/it][2026-05-22 05:58:10,995] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 187])
[2026-05-22 05:58:11,534] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 123])

 21%|██        | 1042/5000 [13:52:49<35:53:48, 32.65s/it]
                                                         
{'loss': '0.2198', 'grad_norm': '4.971', 'learning_rate': '1.616e-05', 'epoch': '0.2084'}

 21%|██        | 1042/5000 [13:52:50<35:53:48, 32.65s/it][2026-05-22 05:58:48,923] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 244])
[2026-05-22 05:58:49,915] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 21%|██        | 1043/5000 [13:53:39<41:34:33, 37.83s/it]
                                                         
{'loss': '0.2808', 'grad_norm': '8.515', 'learning_rate': '1.615e-05', 'epoch': '0.2086'}

 21%|██        | 1043/5000 [13:53:39<41:34:33, 37.83s/it][2026-05-22 05:59:37,627] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 05:59:39,450] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 21%|██        | 1044/5000 [13:54:28<45:09:20, 41.09s/it]
                                                         
{'loss': '0.272', 'grad_norm': '8.78', 'learning_rate': '1.615e-05', 'epoch': '0.2088'}

 21%|██        | 1044/5000 [13:54:28<45:09:20, 41.09s/it][2026-05-22 06:00:24,361] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 307])
[2026-05-22 06:00:24,589] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 280])

 21%|██        | 1045/5000 [13:55:16<47:33:59, 43.30s/it]
                                                         
{'loss': '0.635', 'grad_norm': '9.654', 'learning_rate': '1.614e-05', 'epoch': '0.209'}

 21%|██        | 1045/5000 [13:55:16<47:33:59, 43.30s/it][2026-05-22 06:01:16,243] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 236])
[2026-05-22 06:01:16,306] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 323])

 21%|██        | 1046/5000 [13:56:09<50:49:38, 46.28s/it]
                                                         
{'loss': '0.8126', 'grad_norm': '8.17', 'learning_rate': '1.614e-05', 'epoch': '0.2092'}

 21%|██        | 1046/5000 [13:56:09<50:49:38, 46.28s/it][2026-05-22 06:02:07,320] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 112])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 06:02:08,521] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 71])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

 21%|██        | 1047/5000 [13:56:57<51:25:22, 46.83s/it]
                                                         
{'loss': '0.3074', 'grad_norm': '7.464', 'learning_rate': '1.613e-05', 'epoch': '0.2094'}

 21%|██        | 1047/5000 [13:56:57<51:25:22, 46.83s/it][2026-05-22 06:02:52,492] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 95])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 06:03:00,016] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 103])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 21%|██        | 1048/5000 [13:57:45<51:43:04, 47.11s/it]
                                                         
{'loss': '0.365', 'grad_norm': '10.44', 'learning_rate': '1.613e-05', 'epoch': '0.2096'}

 21%|██        | 1048/5000 [13:57:45<51:43:04, 47.11s/it][2026-05-22 06:03:36,227] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1030])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 06:03:36,348] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 21%|██        | 1049/5000 [13:58:38<53:31:42, 48.77s/it]
                                                         
{'loss': '0.4203', 'grad_norm': '12.22', 'learning_rate': '1.613e-05', 'epoch': '0.2098'}

 21%|██        | 1049/5000 [13:58:38<53:31:42, 48.77s/it][2026-05-22 06:04:28,835] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-22 06:04:29,143] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 21%|██        | 1050/5000 [13:59:30<54:44:44, 49.89s/it]
                                                         
{'loss': '0.117', 'grad_norm': '4.823', 'learning_rate': '1.612e-05', 'epoch': '0.21'}

 21%|██        | 1050/5000 [13:59:30<54:44:44, 49.89s/it][2026-05-22 06:05:37,526] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 209])
[2026-05-22 06:05:37,594] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 74])

 21%|██        | 1051/5000 [14:00:18<54:00:42, 49.24s/it]
                                                         
{'loss': '0.45', 'grad_norm': '9.689', 'learning_rate': '1.612e-05', 'epoch': '0.2102'}

 21%|██        | 1051/5000 [14:00:18<54:00:42, 49.24s/it][2026-05-22 06:06:23,234] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1017])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])
[2026-05-22 06:06:29,387] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1016])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 21%|██        | 1052/5000 [14:01:06<53:34:40, 48.86s/it]
                                                         
{'loss': '0.08581', 'grad_norm': '3.983', 'learning_rate': '1.611e-05', 'epoch': '0.2104'}

 21%|██        | 1052/5000 [14:01:06<53:34:40, 48.86s/it][2026-05-22 06:06:54,179] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 653])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 4])
[2026-05-22 06:06:54,289] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 4])

 21%|██        | 1053/5000 [14:01:53<52:57:40, 48.31s/it]
                                                         
{'loss': '1.033', 'grad_norm': '8.218', 'learning_rate': '1.611e-05', 'epoch': '0.2106'}

 21%|██        | 1053/5000 [14:01:53<52:57:40, 48.31s/it][2026-05-22 06:07:41,227] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 5])
[2026-05-22 06:07:41,232] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 655])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 5])

 21%|██        | 1054/5000 [14:02:40<52:22:12, 47.78s/it]
                                                         
{'loss': '0.7432', 'grad_norm': '8.369', 'learning_rate': '1.611e-05', 'epoch': '0.2108'}

 21%|██        | 1054/5000 [14:02:40<52:22:12, 47.78s/it][2026-05-22 06:08:39,586] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 136])
[2026-05-22 06:08:44,893] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1015])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 124])

 21%|██        | 1055/5000 [14:03:30<53:03:38, 48.42s/it]
                                                         
{'loss': '0.2753', 'grad_norm': '6.238', 'learning_rate': '1.61e-05', 'epoch': '0.211'}

 21%|██        | 1055/5000 [14:03:30<53:03:38, 48.42s/it][2026-05-22 06:09:30,001] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 158])
[2026-05-22 06:09:31,199] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1014])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 181])

 21%|██        | 1056/5000 [14:04:20<53:44:20, 49.05s/it]
                                                         
{'loss': '0.3053', 'grad_norm': '7.661', 'learning_rate': '1.61e-05', 'epoch': '0.2112'}

 21%|██        | 1056/5000 [14:04:20<53:44:20, 49.05s/it][2026-05-22 06:10:06,979] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 483])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-22 06:10:07,116] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 958])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 21%|██        | 1057/5000 [14:05:08<53:16:02, 48.63s/it]
                                                         
{'loss': '0.5405', 'grad_norm': '14.94', 'learning_rate': '1.609e-05', 'epoch': '0.2114'}

 21%|██        | 1057/5000 [14:05:08<53:16:02, 48.63s/it][2026-05-22 06:10:54,744] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 06:10:54,757] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])

 21%|██        | 1058/5000 [14:05:56<53:03:51, 48.46s/it]
                                                         
{'loss': '0.6393', 'grad_norm': '16.57', 'learning_rate': '1.609e-05', 'epoch': '0.2116'}

 21%|██        | 1058/5000 [14:05:56<53:03:51, 48.46s/it][2026-05-22 06:11:44,556] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 06:11:45,617] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 21%|██        | 1059/5000 [14:06:31<48:36:08, 44.40s/it]
                                                         
{'loss': '2.239', 'grad_norm': '15.06', 'learning_rate': '1.609e-05', 'epoch': '0.2118'}

 21%|██        | 1059/5000 [14:06:31<48:36:08, 44.40s/it][2026-05-22 06:12:19,820] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])
[2026-05-22 06:12:20,293] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 493])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 21%|██        | 1060/5000 [14:07:05<45:17:19, 41.38s/it]
                                                         
{'loss': '2.231', 'grad_norm': '16.13', 'learning_rate': '1.608e-05', 'epoch': '0.212'}

 21%|██        | 1060/5000 [14:07:05<45:17:19, 41.38s/it][2026-05-22 06:12:54,680] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 06:12:55,401] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 21%|██        | 1061/5000 [14:07:53<47:26:33, 43.36s/it]
                                                         
{'loss': '0.4539', 'grad_norm': '11.39', 'learning_rate': '1.608e-05', 'epoch': '0.2122'}

 21%|██        | 1061/5000 [14:07:53<47:26:33, 43.36s/it][2026-05-22 06:13:42,811] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-22 06:13:43,946] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 21%|██        | 1062/5000 [14:08:42<49:12:37, 44.99s/it]
                                                         
{'loss': '0.3853', 'grad_norm': '19.96', 'learning_rate': '1.607e-05', 'epoch': '0.2124'}

 21%|██        | 1062/5000 [14:08:42<49:12:37, 44.99s/it][2026-05-22 06:14:41,763] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 06:14:43,859] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 254])

 21%|██▏       | 1063/5000 [14:09:26<48:54:35, 44.72s/it]
                                                         
{'loss': '0.3802', 'grad_norm': '9.797', 'learning_rate': '1.607e-05', 'epoch': '0.2126'}

 21%|██▏       | 1063/5000 [14:09:26<48:54:35, 44.72s/it][2026-05-22 06:15:26,203] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 525])
[2026-05-22 06:15:29,496] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 21%|██▏       | 1064/5000 [14:10:12<49:11:20, 44.99s/it]
                                                         
{'loss': '0.3938', 'grad_norm': '11.4', 'learning_rate': '1.607e-05', 'epoch': '0.2128'}

 21%|██▏       | 1064/5000 [14:10:12<49:11:20, 44.99s/it][2026-05-22 06:16:00,435] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 264])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 06:16:01,010] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 239])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 21%|██▏       | 1065/5000 [14:11:01<50:46:56, 46.46s/it]
                                                         
{'loss': '0.5563', 'grad_norm': '8.494', 'learning_rate': '1.606e-05', 'epoch': '0.213'}

 21%|██▏       | 1065/5000 [14:11:01<50:46:56, 46.46s/it][2026-05-22 06:16:50,961] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 249])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])
[2026-05-22 06:16:53,680] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 21%|██▏       | 1066/5000 [14:11:52<52:08:19, 47.71s/it]
                                                         
{'loss': '0.2187', 'grad_norm': '5.497', 'learning_rate': '1.606e-05', 'epoch': '0.2132'}

 21%|██▏       | 1066/5000 [14:11:52<52:08:19, 47.71s/it][2026-05-22 06:17:40,949] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 8])
[2026-05-22 06:17:41,012] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 258])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 8])

 21%|██▏       | 1067/5000 [14:12:42<52:49:20, 48.35s/it]
                                                         
{'loss': '0.5703', 'grad_norm': '9.716', 'learning_rate': '1.605e-05', 'epoch': '0.2134'}

 21%|██▏       | 1067/5000 [14:12:42<52:49:20, 48.35s/it][2026-05-22 06:18:30,010] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 8])
[2026-05-22 06:18:30,038] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 9])

 21%|██▏       | 1068/5000 [14:13:33<53:41:35, 49.16s/it]
                                                         
{'loss': '0.1627', 'grad_norm': '6.592', 'learning_rate': '1.605e-05', 'epoch': '0.2136'}

 21%|██▏       | 1068/5000 [14:13:33<53:41:35, 49.16s/it][2026-05-22 06:19:21,507] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 218])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 06:19:21,716] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 21%|██▏       | 1069/5000 [14:14:24<54:20:10, 49.76s/it]
                                                         
{'loss': '0.3085', 'grad_norm': '11.02', 'learning_rate': '1.604e-05', 'epoch': '0.2138'}

 21%|██▏       | 1069/5000 [14:14:24<54:20:10, 49.76s/it][2026-05-22 06:20:13,530] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 06:20:14,643] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 216])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 21%|██▏       | 1070/5000 [14:15:19<56:06:33, 51.40s/it]
                                                         
{'loss': '0.3241', 'grad_norm': '5.118', 'learning_rate': '1.604e-05', 'epoch': '0.214'}

 21%|██▏       | 1070/5000 [14:15:19<56:06:33, 51.40s/it][2026-05-22 06:21:09,119] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 169])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 06:21:09,149] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 174])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 21%|██▏       | 1071/5000 [14:16:07<54:48:21, 50.22s/it]
                                                         
{'loss': '1.058', 'grad_norm': '9.159', 'learning_rate': '1.604e-05', 'epoch': '0.2142'}

 21%|██▏       | 1071/5000 [14:16:07<54:48:21, 50.22s/it][2026-05-22 06:21:56,392] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 185])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 06:21:56,610] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 240])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 21%|██▏       | 1072/5000 [14:16:55<54:08:32, 49.62s/it]
                                                         
{'loss': '0.5434', 'grad_norm': '10.65', 'learning_rate': '1.603e-05', 'epoch': '0.2144'}

 21%|██▏       | 1072/5000 [14:16:55<54:08:32, 49.62s/it][2026-05-22 06:22:43,508] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 06:22:44,636] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 229])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 21%|██▏       | 1073/5000 [14:17:20<45:55:16, 42.10s/it]
                                                         
{'loss': '1.368', 'grad_norm': '19.39', 'learning_rate': '1.603e-05', 'epoch': '0.2146'}

 21%|██▏       | 1073/5000 [14:17:20<45:55:16, 42.10s/it][2026-05-22 06:23:08,687] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 169])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 06:23:10,011] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 479])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 980])

 21%|██▏       | 1074/5000 [14:17:45<40:34:58, 37.21s/it]
                                                         
{'loss': '1.471', 'grad_norm': '16.12', 'learning_rate': '1.602e-05', 'epoch': '0.2148'}

 21%|██▏       | 1074/5000 [14:17:45<40:34:58, 37.21s/it][2026-05-22 06:23:32,970] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 969])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 23])
[2026-05-22 06:23:33,075] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 30])

 22%|██▏       | 1075/5000 [14:18:33<43:53:59, 40.26s/it]
                                                         
{'loss': '1.355', 'grad_norm': '12.34', 'learning_rate': '1.602e-05', 'epoch': '0.215'}

 22%|██▏       | 1075/5000 [14:18:33<43:53:59, 40.26s/it][2026-05-22 06:24:20,382] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 958])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 25])
[2026-05-22 06:24:20,385] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 30])

 22%|██▏       | 1076/5000 [14:19:21<46:32:38, 42.70s/it]
                                                         
{'loss': '0.7269', 'grad_norm': '15.72', 'learning_rate': '1.602e-05', 'epoch': '0.2152'}

 22%|██▏       | 1076/5000 [14:19:21<46:32:38, 42.70s/it][2026-05-22 06:25:08,289] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 312])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 80])
[2026-05-22 06:25:08,397] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 620])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 78])

 22%|██▏       | 1077/5000 [14:20:07<47:30:23, 43.60s/it]
                                                         
{'loss': '0.7073', 'grad_norm': '9.702', 'learning_rate': '1.601e-05', 'epoch': '0.2154'}

 22%|██▏       | 1077/5000 [14:20:07<47:30:23, 43.60s/it][2026-05-22 06:25:54,078] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])
[2026-05-22 06:25:55,221] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 54])

 22%|██▏       | 1078/5000 [14:20:54<48:33:20, 44.57s/it]
                                                         
{'loss': '0.5805', 'grad_norm': '10.95', 'learning_rate': '1.601e-05', 'epoch': '0.2156'}

 22%|██▏       | 1078/5000 [14:20:54<48:33:20, 44.57s/it][2026-05-22 06:26:43,707] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 06:26:44,370] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 22%|██▏       | 1079/5000 [14:21:40<49:05:14, 45.07s/it]
                                                         
{'loss': '0.6129', 'grad_norm': '7.377', 'learning_rate': '1.6e-05', 'epoch': '0.2158'}

 22%|██▏       | 1079/5000 [14:21:40<49:05:14, 45.07s/it][2026-05-22 06:27:31,349] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 216])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 06:27:31,430] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 22%|██▏       | 1080/5000 [14:22:32<51:21:35, 47.17s/it]
                                                         
{'loss': '0.5013', 'grad_norm': '9.811', 'learning_rate': '1.6e-05', 'epoch': '0.216'}

 22%|██▏       | 1080/5000 [14:22:32<51:21:35, 47.17s/it][2026-05-22 06:28:19,161] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 06:28:19,588] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 373])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 22%|██▏       | 1081/5000 [14:23:15<50:00:02, 45.93s/it]
                                                         
{'loss': '0.9121', 'grad_norm': '17.66', 'learning_rate': '1.6e-05', 'epoch': '0.2162'}

 22%|██▏       | 1081/5000 [14:23:15<50:00:02, 45.93s/it][2026-05-22 06:29:02,700] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 06:29:02,838] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 691])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 22%|██▏       | 1082/5000 [14:23:58<49:03:26, 45.08s/it]
                                                         
{'loss': '0.9014', 'grad_norm': '17.18', 'learning_rate': '1.599e-05', 'epoch': '0.2164'}

 22%|██▏       | 1082/5000 [14:23:58<49:03:26, 45.08s/it][2026-05-22 06:29:46,866] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 06:29:47,387] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 393])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 958])

 22%|██▏       | 1083/5000 [14:24:30<44:36:11, 40.99s/it]
                                                         
{'loss': '0.2232', 'grad_norm': '9.585', 'learning_rate': '1.599e-05', 'epoch': '0.2166'}

 22%|██▏       | 1083/5000 [14:24:30<44:36:11, 40.99s/it][2026-05-22 06:30:19,091] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 380])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-22 06:30:19,378] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 306])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 22%|██▏       | 1084/5000 [14:25:07<43:26:54, 39.94s/it]
                                                         
{'loss': '0.1357', 'grad_norm': '7.036', 'learning_rate': '1.598e-05', 'epoch': '0.2168'}

 22%|██▏       | 1084/5000 [14:25:07<43:26:54, 39.94s/it][2026-05-22 06:30:58,276] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 103])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 06:30:59,385] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 124])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 22%|██▏       | 1085/5000 [14:25:56<46:19:27, 42.60s/it]
                                                         
{'loss': '0.9033', 'grad_norm': '10.85', 'learning_rate': '1.598e-05', 'epoch': '0.217'}

 22%|██▏       | 1085/5000 [14:25:56<46:19:27, 42.60s/it][2026-05-22 06:31:46,891] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 95])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])
[2026-05-22 06:31:47,570] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 103])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 22%|██▏       | 1086/5000 [14:26:49<49:53:13, 45.89s/it]
                                                         
{'loss': '0.9756', 'grad_norm': '15.11', 'learning_rate': '1.598e-05', 'epoch': '0.2172'}

 22%|██▏       | 1086/5000 [14:26:49<49:53:13, 45.89s/it][2026-05-22 06:32:37,901] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 53])
[2026-05-22 06:32:38,673] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])

 22%|██▏       | 1087/5000 [14:27:19<44:29:43, 40.94s/it]
                                                         
{'loss': '0.4667', 'grad_norm': '10.47', 'learning_rate': '1.597e-05', 'epoch': '0.2174'}

 22%|██▏       | 1087/5000 [14:27:19<44:29:43, 40.94s/it][2026-05-22 06:33:07,381] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])
[2026-05-22 06:33:07,474] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 380])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 256])

 22%|██▏       | 1088/5000 [14:27:57<43:39:53, 40.18s/it]
                                                         
{'loss': '0.3104', 'grad_norm': '9.521', 'learning_rate': '1.597e-05', 'epoch': '0.2176'}

 22%|██▏       | 1088/5000 [14:27:57<43:39:53, 40.18s/it][2026-05-22 06:33:45,880] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 202])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 06:33:47,590] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 202])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 22%|██▏       | 1089/5000 [14:28:47<46:46:28, 43.06s/it]
                                                         
{'loss': '0.5954', 'grad_norm': '10.26', 'learning_rate': '1.596e-05', 'epoch': '0.2178'}

 22%|██▏       | 1089/5000 [14:28:47<46:46:28, 43.06s/it][2026-05-22 06:34:37,188] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 186])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])
[2026-05-22 06:34:37,840] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 130])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

 22%|██▏       | 1090/5000 [14:29:37<48:59:35, 45.11s/it]
                                                         
{'loss': '0.2853', 'grad_norm': '6.334', 'learning_rate': '1.596e-05', 'epoch': '0.218'}

 22%|██▏       | 1090/5000 [14:29:37<48:59:35, 45.11s/it][2026-05-22 06:35:59,271] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 06:36:04,620] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 22%|██▏       | 1091/5000 [14:30:25<49:49:45, 45.89s/it]
                                                         
{'loss': '1.591', 'grad_norm': '9.763', 'learning_rate': '1.596e-05', 'epoch': '0.2182'}

 22%|██▏       | 1091/5000 [14:30:25<49:49:45, 45.89s/it][2026-05-22 06:36:47,576] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])
[2026-05-22 06:36:49,624] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])

 22%|██▏       | 1092/5000 [14:31:24<54:07:37, 49.86s/it]
                                                         
{'loss': '0.5968', 'grad_norm': '12.31', 'learning_rate': '1.595e-05', 'epoch': '0.2184'}

 22%|██▏       | 1092/5000 [14:31:24<54:07:37, 49.86s/it][2026-05-22 06:37:13,046] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 297])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 06:37:16,686] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 261])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

 22%|██▏       | 1093/5000 [14:31:58<48:57:46, 45.12s/it]
                                                         
{'loss': '2.251', 'grad_norm': '22.25', 'learning_rate': '1.595e-05', 'epoch': '0.2186'}

 22%|██▏       | 1093/5000 [14:31:58<48:57:46, 45.12s/it][2026-05-22 06:37:48,094] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 279])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 06:37:48,431] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 234])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 22%|██▏       | 1094/5000 [14:32:27<43:50:13, 40.40s/it]
                                                         
{'loss': '2.545', 'grad_norm': '27.32', 'learning_rate': '1.594e-05', 'epoch': '0.2188'}

 22%|██▏       | 1094/5000 [14:32:27<43:50:13, 40.40s/it][2026-05-22 06:38:15,521] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 958])
[2026-05-22 06:38:17,014] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 22%|██▏       | 1095/5000 [14:33:18<47:08:53, 43.47s/it]
                                                         
{'loss': '1.448', 'grad_norm': '8.289', 'learning_rate': '1.594e-05', 'epoch': '0.219'}

 22%|██▏       | 1095/5000 [14:33:18<47:08:53, 43.47s/it][2026-05-22 06:39:05,860] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 713])
[2026-05-22 06:39:06,284] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 944])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 22%|██▏       | 1096/5000 [14:34:05<48:29:50, 44.72s/it]
                                                         
{'loss': '1.917', 'grad_norm': '12.74', 'learning_rate': '1.593e-05', 'epoch': '0.2192'}

 22%|██▏       | 1096/5000 [14:34:05<48:29:50, 44.72s/it][2026-05-22 06:39:53,121] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 909])
[2026-05-22 06:39:53,742] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 140])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 22%|██▏       | 1097/5000 [14:35:16<56:57:11, 52.53s/it]
                                                         
{'loss': '0.3959', 'grad_norm': '8.819', 'learning_rate': '1.593e-05', 'epoch': '0.2194'}

 22%|██▏       | 1097/5000 [14:35:16<56:57:11, 52.53s/it][2026-05-22 06:41:04,267] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 220])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 975])
[2026-05-22 06:41:05,155] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 132])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 22%|██▏       | 1098/5000 [14:36:27<62:43:08, 57.86s/it]
                                                         
{'loss': '0.505', 'grad_norm': '8.046', 'learning_rate': '1.593e-05', 'epoch': '0.2196'}

 22%|██▏       | 1098/5000 [14:36:27<62:43:08, 57.86s/it][2026-05-22 06:42:14,188] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 294])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 06:42:14,745] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 294])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 22%|██▏       | 1099/5000 [14:37:13<59:07:56, 54.57s/it]
                                                         
{'loss': '0.624', 'grad_norm': '9.942', 'learning_rate': '1.592e-05', 'epoch': '0.2198'}

 22%|██▏       | 1099/5000 [14:37:13<59:07:56, 54.57s/it][2026-05-22 06:43:01,145] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 249])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 06:43:01,741] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 279])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 22%|██▏       | 1100/5000 [14:38:00<56:34:43, 52.23s/it]
                                                         
{'loss': '0.2791', 'grad_norm': '6.99', 'learning_rate': '1.592e-05', 'epoch': '0.22'}

 22%|██▏       | 1100/5000 [14:38:00<56:34:43, 52.23s/it][2026-05-22 06:43:49,706] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 89])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])
[2026-05-22 06:43:49,901] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 97])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 22%|██▏       | 1101/5000 [14:39:09<61:49:10, 57.08s/it]
                                                         
{'loss': '0.6275', 'grad_norm': '14.05', 'learning_rate': '1.591e-05', 'epoch': '0.2202'}

 22%|██▏       | 1101/5000 [14:39:09<61:49:10, 57.08s/it][2026-05-22 06:44:57,449] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 96])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 06:44:59,526] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 80])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 22%|██▏       | 1102/5000 [14:40:15<64:54:18, 59.94s/it]
                                                         
{'loss': '0.8954', 'grad_norm': '14.88', 'learning_rate': '1.591e-05', 'epoch': '0.2204'}

 22%|██▏       | 1102/5000 [14:40:15<64:54:18, 59.94s/it][2026-05-22 06:46:09,062] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 06:46:09,810] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 22%|██▏       | 1103/5000 [14:41:03<60:59:31, 56.34s/it]
                                                         
{'loss': '0.383', 'grad_norm': '8.32', 'learning_rate': '1.591e-05', 'epoch': '0.2206'}

 22%|██▏       | 1103/5000 [14:41:03<60:59:31, 56.34s/it][2026-05-22 06:46:56,744] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 06:46:57,409] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 22%|██▏       | 1104/5000 [14:41:51<58:23:16, 53.95s/it]
                                                         
{'loss': '0.4629', 'grad_norm': '11.23', 'learning_rate': '1.59e-05', 'epoch': '0.2208'}

 22%|██▏       | 1104/5000 [14:41:51<58:23:16, 53.95s/it][2026-05-22 06:47:51,384] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 06:47:54,136] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 22%|██▏       | 1105/5000 [14:42:25<51:49:15, 47.90s/it]
                                                         
{'loss': '0.2175', 'grad_norm': '6.868', 'learning_rate': '1.59e-05', 'epoch': '0.221'}

 22%|██▏       | 1105/5000 [14:42:25<51:49:15, 47.90s/it][2026-05-22 06:48:23,981] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-22 06:48:25,077] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 22%|██▏       | 1106/5000 [14:43:02<48:14:16, 44.60s/it]
                                                         
{'loss': '0.05779', 'grad_norm': '4.158', 'learning_rate': '1.589e-05', 'epoch': '0.2212'}

 22%|██▏       | 1106/5000 [14:43:02<48:14:16, 44.60s/it][2026-05-22 06:48:50,646] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 958])
[2026-05-22 06:48:54,586] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 22%|██▏       | 1107/5000 [14:43:53<50:12:10, 46.42s/it]
                                                         
{'loss': '0.4217', 'grad_norm': '12.21', 'learning_rate': '1.589e-05', 'epoch': '0.2214'}

 22%|██▏       | 1107/5000 [14:43:53<50:12:10, 46.42s/it][2026-05-22 06:49:42,689] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 06:49:44,156] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

 22%|██▏       | 1108/5000 [14:44:40<50:29:10, 46.70s/it]
                                                         
{'loss': '0.371', 'grad_norm': '8.648', 'learning_rate': '1.589e-05', 'epoch': '0.2216'}

 22%|██▏       | 1108/5000 [14:44:40<50:29:10, 46.70s/it][2026-05-22 06:50:28,407] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 958])
[2026-05-22 06:50:30,466] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

 22%|██▏       | 1109/5000 [14:45:51<58:17:35, 53.93s/it]
                                                         
{'loss': '0.6589', 'grad_norm': '10.56', 'learning_rate': '1.588e-05', 'epoch': '0.2218'}

 22%|██▏       | 1109/5000 [14:45:51<58:17:35, 53.93s/it][2026-05-22 06:51:41,270] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])
[2026-05-22 06:51:41,564] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 22%|██▏       | 1110/5000 [14:47:01<63:31:09, 58.78s/it]
                                                         
{'loss': '0.6284', 'grad_norm': '10.84', 'learning_rate': '1.588e-05', 'epoch': '0.222'}

 22%|██▏       | 1110/5000 [14:47:01<63:31:09, 58.78s/it][2026-05-22 06:52:51,013] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])
[2026-05-22 06:52:51,140] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1034])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 112])

 22%|██▏       | 1111/5000 [14:48:11<67:11:01, 62.19s/it]
                                                         
{'loss': '0.6577', 'grad_norm': '10.62', 'learning_rate': '1.587e-05', 'epoch': '0.2222'}

 22%|██▏       | 1111/5000 [14:48:11<67:11:01, 62.19s/it][2026-05-22 06:54:01,306] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 91])
[2026-05-22 06:54:01,576] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 95])

 22%|██▏       | 1112/5000 [14:49:21<69:43:38, 64.56s/it]
                                                         
{'loss': '1.044', 'grad_norm': '9.058', 'learning_rate': '1.587e-05', 'epoch': '0.2224'}

 22%|██▏       | 1112/5000 [14:49:21<69:43:38, 64.56s/it][2026-05-22 06:55:21,375] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 06:55:24,505] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 22%|██▏       | 1113/5000 [14:50:14<65:47:36, 60.94s/it]
                                                         
{'loss': '0.4351', 'grad_norm': '8.695', 'learning_rate': '1.587e-05', 'epoch': '0.2226'}

 22%|██▏       | 1113/5000 [14:50:14<65:47:36, 60.94s/it][2026-05-22 06:56:12,843] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])
[2026-05-22 06:56:15,963] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 22%|██▏       | 1114/5000 [14:51:07<63:23:54, 58.73s/it]
                                                         
{'loss': '0.7542', 'grad_norm': '12.05', 'learning_rate': '1.586e-05', 'epoch': '0.2228'}

 22%|██▏       | 1114/5000 [14:51:07<63:23:54, 58.73s/it][2026-05-22 06:56:56,545] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 944])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])
[2026-05-22 06:56:58,321] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])

 22%|██▏       | 1115/5000 [14:51:57<60:33:13, 56.11s/it]
                                                         
{'loss': '0.5359', 'grad_norm': '8.174', 'learning_rate': '1.586e-05', 'epoch': '0.223'}

 22%|██▏       | 1115/5000 [14:51:57<60:33:13, 56.11s/it][2026-05-22 06:57:46,236] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 286])
[2026-05-22 06:57:46,452] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 282])

 22%|██▏       | 1116/5000 [14:52:47<58:25:14, 54.15s/it]
                                                         
{'loss': '0.3555', 'grad_norm': '7.046', 'learning_rate': '1.585e-05', 'epoch': '0.2232'}

 22%|██▏       | 1116/5000 [14:52:47<58:25:14, 54.15s/it][2026-05-22 06:58:35,775] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 212])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 06:58:38,436] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 993])

 22%|██▏       | 1117/5000 [14:53:35<56:29:05, 52.37s/it]
                                                         
{'loss': '0.1967', 'grad_norm': '7.606', 'learning_rate': '1.585e-05', 'epoch': '0.2234'}

 22%|██▏       | 1117/5000 [14:53:35<56:29:05, 52.37s/it][2026-05-22 06:59:24,986] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 06:59:26,472] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 22%|██▏       | 1118/5000 [14:54:22<54:45:51, 50.79s/it]
                                                         
{'loss': '0.1308', 'grad_norm': '5.035', 'learning_rate': '1.584e-05', 'epoch': '0.2236'}

 22%|██▏       | 1118/5000 [14:54:22<54:45:51, 50.79s/it][2026-05-22 07:00:27,897] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 134])
[2026-05-22 07:00:30,147] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 173])

 22%|██▏       | 1119/5000 [14:55:36<62:06:35, 57.61s/it]
                                                         
{'loss': '0.529', 'grad_norm': '13.79', 'learning_rate': '1.584e-05', 'epoch': '0.2238'}

 22%|██▏       | 1119/5000 [14:55:36<62:06:35, 57.61s/it][2026-05-22 07:01:44,000] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 110])
[2026-05-22 07:01:50,192] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 22%|██▏       | 1120/5000 [14:56:48<66:45:33, 61.94s/it]
                                                         
{'loss': '0.7899', 'grad_norm': '11.25', 'learning_rate': '1.584e-05', 'epoch': '0.224'}

 22%|██▏       | 1120/5000 [14:56:48<66:45:33, 61.94s/it][2026-05-22 07:02:37,227] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 978])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 07:02:37,793] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 984])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 22%|██▏       | 1121/5000 [14:57:35<61:56:42, 57.49s/it]
                                                         
{'loss': '0.5385', 'grad_norm': '11.98', 'learning_rate': '1.583e-05', 'epoch': '0.2242'}

 22%|██▏       | 1121/5000 [14:57:35<61:56:42, 57.49s/it][2026-05-22 07:03:24,548] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 986])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 07:03:25,625] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 979])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 22%|██▏       | 1122/5000 [14:58:23<58:47:53, 54.58s/it]
                                                         
{'loss': '0.3177', 'grad_norm': '11.05', 'learning_rate': '1.583e-05', 'epoch': '0.2244'}

 22%|██▏       | 1122/5000 [14:58:23<58:47:53, 54.58s/it][2026-05-22 07:04:25,487] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 07:04:25,825] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])

 22%|██▏       | 1123/5000 [14:59:20<59:32:55, 55.29s/it]
                                                         
{'loss': '0.7217', 'grad_norm': '11.14', 'learning_rate': '1.582e-05', 'epoch': '0.2246'}

 22%|██▏       | 1123/5000 [14:59:20<59:32:55, 55.29s/it][2026-05-22 07:05:20,824] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 07:05:22,014] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 22%|██▏       | 1124/5000 [15:00:17<60:16:21, 55.98s/it]
                                                         
{'loss': '0.5789', 'grad_norm': '9.659', 'learning_rate': '1.582e-05', 'epoch': '0.2248'}

 22%|██▏       | 1124/5000 [15:00:17<60:16:21, 55.98s/it][2026-05-22 07:06:06,643] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 279])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])
[2026-05-22 07:06:07,742] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 234])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

 22%|██▎       | 1125/5000 [15:01:07<58:23:31, 54.25s/it]
                                                         
{'loss': '0.393', 'grad_norm': '8.097', 'learning_rate': '1.582e-05', 'epoch': '0.225'}

 22%|██▎       | 1125/5000 [15:01:08<58:23:31, 54.25s/it][2026-05-22 07:06:57,319] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 221])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 07:06:58,357] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 265])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 993])

 23%|██▎       | 1126/5000 [15:01:58<57:03:58, 53.03s/it]
                                                         
{'loss': '0.269', 'grad_norm': '7.999', 'learning_rate': '1.581e-05', 'epoch': '0.2252'}

 23%|██▎       | 1126/5000 [15:01:58<57:03:58, 53.03s/it][2026-05-22 07:07:45,326] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 07:07:45,381] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 397])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 23%|██▎       | 1127/5000 [15:02:47<55:52:23, 51.93s/it]
                                                         
{'loss': '0.263', 'grad_norm': '7.551', 'learning_rate': '1.581e-05', 'epoch': '0.2254'}

 23%|██▎       | 1127/5000 [15:02:47<55:52:23, 51.93s/it][2026-05-22 07:08:34,711] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 181])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-22 07:08:35,018] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 477])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 23%|██▎       | 1128/5000 [15:03:37<55:06:32, 51.24s/it]
                                                         
{'loss': '0.08478', 'grad_norm': '3.803', 'learning_rate': '1.58e-05', 'epoch': '0.2256'}

 23%|██▎       | 1128/5000 [15:03:37<55:06:32, 51.24s/it][2026-05-22 07:09:23,067] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 133])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 07:09:23,085] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 125])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 184])

 23%|██▎       | 1129/5000 [15:04:46<61:04:29, 56.80s/it]
                                                         
{'loss': '0.4562', 'grad_norm': '9.037', 'learning_rate': '1.58e-05', 'epoch': '0.2258'}

 23%|██▎       | 1129/5000 [15:04:46<61:04:29, 56.80s/it][2026-05-22 07:10:32,878] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 157])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 111])
[2026-05-22 07:10:33,566] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 125])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 23%|██▎       | 1130/5000 [15:05:56<65:08:37, 60.60s/it]
                                                         
{'loss': '0.576', 'grad_norm': '9.844', 'learning_rate': '1.58e-05', 'epoch': '0.226'}

 23%|██▎       | 1130/5000 [15:05:56<65:08:37, 60.60s/it][2026-05-22 07:11:43,276] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 263])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 07:11:44,704] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 320])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 23%|██▎       | 1131/5000 [15:06:43<60:49:39, 56.60s/it]
                                                         
{'loss': '0.5551', 'grad_norm': '11.1', 'learning_rate': '1.579e-05', 'epoch': '0.2262'}

 23%|██▎       | 1131/5000 [15:06:43<60:49:39, 56.60s/it][2026-05-22 07:12:31,726] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 390])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 07:12:31,749] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 498])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 23%|██▎       | 1132/5000 [15:07:31<57:57:07, 53.94s/it]
                                                         
{'loss': '0.3534', 'grad_norm': '12.2', 'learning_rate': '1.579e-05', 'epoch': '0.2264'}

 23%|██▎       | 1132/5000 [15:07:31<57:57:07, 53.94s/it][2026-05-22 07:13:17,695] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 229])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 07:13:17,695] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

 23%|██▎       | 1133/5000 [15:07:57<48:50:59, 45.48s/it]
                                                         
{'loss': '0.2396', 'grad_norm': '11.21', 'learning_rate': '1.578e-05', 'epoch': '0.2266'}

 23%|██▎       | 1133/5000 [15:07:57<48:50:59, 45.48s/it][2026-05-22 07:13:43,380] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 183])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-22 07:13:44,599] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 462])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 23%|██▎       | 1134/5000 [15:08:25<43:11:00, 40.21s/it]
                                                         
{'loss': '0.2478', 'grad_norm': '11.76', 'learning_rate': '1.578e-05', 'epoch': '0.2268'}

 23%|██▎       | 1134/5000 [15:08:25<43:11:00, 40.21s/it][2026-05-22 07:14:26,447] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 276])
[2026-05-22 07:14:28,099] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])

 23%|██▎       | 1135/5000 [15:08:50<38:29:01, 35.85s/it]
                                                         
{'loss': '0.897', 'grad_norm': '23.14', 'learning_rate': '1.578e-05', 'epoch': '0.227'}

 23%|██▎       | 1135/5000 [15:08:50<38:29:01, 35.85s/it][2026-05-22 07:14:50,908] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 240])
[2026-05-22 07:14:51,342] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1012])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])

 23%|██▎       | 1136/5000 [15:09:17<35:31:48, 33.10s/it]
                                                         
{'loss': '0.5985', 'grad_norm': '19.72', 'learning_rate': '1.577e-05', 'epoch': '0.2272'}

 23%|██▎       | 1136/5000 [15:09:17<35:31:48, 33.10s/it][2026-05-22 07:15:04,931] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 287])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 281])
[2026-05-22 07:15:05,255] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 412])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 23%|██▎       | 1137/5000 [15:09:47<34:39:48, 32.30s/it]
                                                         
{'loss': '0.304', 'grad_norm': '8.402', 'learning_rate': '1.577e-05', 'epoch': '0.2274'}

 23%|██▎       | 1137/5000 [15:09:47<34:39:48, 32.30s/it][2026-05-22 07:15:35,599] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 432])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 07:15:36,142] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 23%|██▎       | 1138/5000 [15:10:21<35:11:55, 32.81s/it]
                                                         
{'loss': '0.02049', 'grad_norm': '2.267', 'learning_rate': '1.576e-05', 'epoch': '0.2276'}

 23%|██▎       | 1138/5000 [15:10:21<35:11:55, 32.81s/it][2026-05-22 07:16:23,411] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-22 07:16:23,687] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 23%|██▎       | 1139/5000 [15:10:50<33:45:45, 31.48s/it]
                                                         
{'loss': '0.2636', 'grad_norm': '9.018', 'learning_rate': '1.576e-05', 'epoch': '0.2278'}

 23%|██▎       | 1139/5000 [15:10:50<33:45:45, 31.48s/it][2026-05-22 07:16:51,985] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1034])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 07:16:53,903] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 23%|██▎       | 1140/5000 [15:11:25<35:03:14, 32.69s/it]
                                                         
{'loss': '0.1938', 'grad_norm': '7.476', 'learning_rate': '1.576e-05', 'epoch': '0.228'}

 23%|██▎       | 1140/5000 [15:11:25<35:03:14, 32.69s/it][2026-05-22 07:17:15,483] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 450])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 07:17:15,492] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 23%|██▎       | 1141/5000 [15:12:15<40:26:02, 37.72s/it]
                                                         
{'loss': '0.4955', 'grad_norm': '8.226', 'learning_rate': '1.575e-05', 'epoch': '0.2282'}

 23%|██▎       | 1141/5000 [15:12:15<40:26:02, 37.72s/it][2026-05-22 07:18:04,462] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 955])
[2026-05-22 07:18:05,600] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 23%|██▎       | 1142/5000 [15:13:04<44:03:13, 41.11s/it]
                                                         
{'loss': '0.2758', 'grad_norm': '7.242', 'learning_rate': '1.575e-05', 'epoch': '0.2284'}

 23%|██▎       | 1142/5000 [15:13:04<44:03:13, 41.11s/it][2026-05-22 07:18:51,117] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])
[2026-05-22 07:18:52,620] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 230])

 23%|██▎       | 1143/5000 [15:13:36<41:05:01, 38.35s/it]
                                                         
{'loss': '0.2909', 'grad_norm': '8.157', 'learning_rate': '1.574e-05', 'epoch': '0.2286'}

 23%|██▎       | 1143/5000 [15:13:36<41:05:01, 38.35s/it][2026-05-22 07:19:24,046] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 266])
[2026-05-22 07:19:24,457] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 308])

 23%|██▎       | 1144/5000 [15:14:09<39:33:34, 36.93s/it]
                                                         
{'loss': '0.02788', 'grad_norm': '2.815', 'learning_rate': '1.574e-05', 'epoch': '0.2288'}

 23%|██▎       | 1144/5000 [15:14:09<39:33:34, 36.93s/it][2026-05-22 07:19:58,509] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 07:19:58,685] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 194])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 23%|██▎       | 1145/5000 [15:15:20<50:29:57, 47.16s/it]
                                                         
{'loss': '0.841', 'grad_norm': '8.284', 'learning_rate': '1.573e-05', 'epoch': '0.229'}

 23%|██▎       | 1145/5000 [15:15:20<50:29:57, 47.16s/it][2026-05-22 07:21:10,075] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 980])
[2026-05-22 07:21:10,383] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 972])

 23%|██▎       | 1146/5000 [15:16:34<58:57:59, 55.08s/it]
                                                         
{'loss': '0.4277', 'grad_norm': '10.06', 'learning_rate': '1.573e-05', 'epoch': '0.2292'}

 23%|██▎       | 1146/5000 [15:16:34<58:57:59, 55.08s/it][2026-05-22 07:22:26,788] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 46])
[2026-05-22 07:22:27,401] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 55])

 23%|██▎       | 1147/5000 [15:17:24<57:14:51, 53.49s/it]
                                                         
{'loss': '0.8977', 'grad_norm': '10.64', 'learning_rate': '1.573e-05', 'epoch': '0.2294'}

 23%|██▎       | 1147/5000 [15:17:24<57:14:51, 53.49s/it][2026-05-22 07:23:16,286] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 45])
[2026-05-22 07:23:16,573] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 54])

 23%|██▎       | 1148/5000 [15:18:17<57:12:31, 53.47s/it]
                                                         
{'loss': '0.3788', 'grad_norm': '9.421', 'learning_rate': '1.572e-05', 'epoch': '0.2296'}

 23%|██▎       | 1148/5000 [15:18:17<57:12:31, 53.47s/it][2026-05-22 07:24:05,079] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 07:24:05,111] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 266])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 23%|██▎       | 1149/5000 [15:18:47<49:31:45, 46.30s/it]
                                                         
{'loss': '0.5376', 'grad_norm': '11.45', 'learning_rate': '1.572e-05', 'epoch': '0.2298'}

 23%|██▎       | 1149/5000 [15:18:47<49:31:45, 46.30s/it][2026-05-22 07:24:34,372] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 313])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 07:24:35,735] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 23%|██▎       | 1150/5000 [15:19:17<44:23:25, 41.51s/it]
                                                         
{'loss': '0.07132', 'grad_norm': '5.324', 'learning_rate': '1.571e-05', 'epoch': '0.23'}

 23%|██▎       | 1150/5000 [15:19:17<44:23:25, 41.51s/it][2026-05-22 07:25:06,326] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 244])
[2026-05-22 07:25:07,452] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 295])

 23%|██▎       | 1151/5000 [15:20:07<47:14:58, 44.19s/it]
                                                         
{'loss': '1.14', 'grad_norm': '9.235', 'learning_rate': '1.571e-05', 'epoch': '0.2302'}

 23%|██▎       | 1151/5000 [15:20:07<47:14:58, 44.19s/it][2026-05-22 07:25:57,612] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 984])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 284])
[2026-05-22 07:25:57,790] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 288])

 23%|██▎       | 1152/5000 [15:20:59<49:38:05, 46.44s/it]
                                                         
{'loss': '0.9342', 'grad_norm': '8.762', 'learning_rate': '1.571e-05', 'epoch': '0.2304'}

 23%|██▎       | 1152/5000 [15:20:59<49:38:05, 46.44s/it][2026-05-22 07:26:47,869] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 108])
[2026-05-22 07:26:48,396] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 975])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])

 23%|██▎       | 1153/5000 [15:21:46<49:48:31, 46.61s/it]
                                                         
{'loss': '0.08211', 'grad_norm': '5.706', 'learning_rate': '1.57e-05', 'epoch': '0.2306'}

 23%|██▎       | 1153/5000 [15:21:46<49:48:31, 46.61s/it][2026-05-22 07:27:34,959] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 973])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 115])
[2026-05-22 07:27:35,639] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 976])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 41])

 23%|██▎       | 1154/5000 [15:22:33<50:01:35, 46.83s/it]
                                                         
{'loss': '0.1752', 'grad_norm': '6.187', 'learning_rate': '1.57e-05', 'epoch': '0.2308'}

 23%|██▎       | 1154/5000 [15:22:33<50:01:35, 46.83s/it][2026-05-22 07:28:21,164] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 664])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 235])
[2026-05-22 07:28:22,342] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 640])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 294])

 23%|██▎       | 1155/5000 [15:23:04<44:46:01, 41.91s/it]
                                                         
{'loss': '0.4295', 'grad_norm': '8.288', 'learning_rate': '1.569e-05', 'epoch': '0.231'}

 23%|██▎       | 1155/5000 [15:23:04<44:46:01, 41.91s/it][2026-05-22 07:28:52,326] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 637])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 249])
[2026-05-22 07:28:52,763] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 642])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 222])

 23%|██▎       | 1156/5000 [15:23:38<42:09:17, 39.48s/it]
                                                         
{'loss': '0.01876', 'grad_norm': '2.369', 'learning_rate': '1.569e-05', 'epoch': '0.2312'}

 23%|██▎       | 1156/5000 [15:23:38<42:09:17, 39.48s/it][2026-05-22 07:29:25,220] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 629])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 59])
[2026-05-22 07:29:25,857] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 630])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])

 23%|██▎       | 1157/5000 [15:24:24<44:29:47, 41.68s/it]
                                                         
{'loss': '0.6323', 'grad_norm': '10.67', 'learning_rate': '1.569e-05', 'epoch': '0.2314'}

 23%|██▎       | 1157/5000 [15:24:24<44:29:47, 41.68s/it][2026-05-22 07:30:12,181] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 631])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 73])
[2026-05-22 07:30:13,531] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 679])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 139])

 23%|██▎       | 1158/5000 [15:25:12<46:15:26, 43.34s/it]
                                                         
{'loss': '0.05858', 'grad_norm': '3.163', 'learning_rate': '1.568e-05', 'epoch': '0.2316'}

 23%|██▎       | 1158/5000 [15:25:12<46:15:26, 43.34s/it][2026-05-22 07:31:12,714] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 246])
[2026-05-22 07:31:14,931] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 252])

 23%|██▎       | 1159/5000 [15:25:57<47:00:03, 44.05s/it]
                                                         
{'loss': '0.6279', 'grad_norm': '25.2', 'learning_rate': '1.568e-05', 'epoch': '0.2318'}

 23%|██▎       | 1159/5000 [15:25:57<47:00:03, 44.05s/it][2026-05-22 07:31:57,062] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 250])
[2026-05-22 07:32:00,621] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 246])

 23%|██▎       | 1160/5000 [15:26:43<47:29:17, 44.52s/it]
                                                         
{'loss': '0.564', 'grad_norm': '17.49', 'learning_rate': '1.567e-05', 'epoch': '0.232'}

 23%|██▎       | 1160/5000 [15:26:43<47:29:17, 44.52s/it][2026-05-22 07:32:31,120] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 07:32:31,129] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 248])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])

 23%|██▎       | 1161/5000 [15:27:18<44:20:17, 41.58s/it]
                                                         
{'loss': '0.9669', 'grad_norm': '9.188', 'learning_rate': '1.567e-05', 'epoch': '0.2322'}

 23%|██▎       | 1161/5000 [15:27:18<44:20:17, 41.58s/it][2026-05-22 07:33:05,775] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 431])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 07:33:06,594] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 23%|██▎       | 1162/5000 [15:27:52<41:58:38, 39.37s/it]
                                                         
{'loss': '0.8179', 'grad_norm': '8.705', 'learning_rate': '1.567e-05', 'epoch': '0.2324'}

 23%|██▎       | 1162/5000 [15:27:52<41:58:38, 39.37s/it][2026-05-22 07:34:05,830] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 142])
[2026-05-22 07:34:06,461] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

 23%|██▎       | 1163/5000 [15:28:24<39:27:33, 37.02s/it]
                                                         
{'loss': '0.8471', 'grad_norm': '9.248', 'learning_rate': '1.566e-05', 'epoch': '0.2326'}

 23%|██▎       | 1163/5000 [15:28:24<39:27:33, 37.02s/it][2026-05-22 07:34:41,784] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 163])
[2026-05-22 07:34:43,968] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])

 23%|██▎       | 1164/5000 [15:29:03<40:05:57, 37.63s/it]
                                                         
{'loss': '0.7964', 'grad_norm': '7.668', 'learning_rate': '1.566e-05', 'epoch': '0.2328'}

 23%|██▎       | 1164/5000 [15:29:03<40:05:57, 37.63s/it][2026-05-22 07:35:03,323] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 07:35:03,899] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 23%|██▎       | 1165/5000 [15:29:57<45:36:16, 42.81s/it]
                                                         
{'loss': '0.3534', 'grad_norm': '12.12', 'learning_rate': '1.565e-05', 'epoch': '0.233'}

 23%|██▎       | 1165/5000 [15:29:57<45:36:16, 42.81s/it][2026-05-22 07:35:58,771] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])
[2026-05-22 07:36:00,704] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

 23%|██▎       | 1166/5000 [15:30:48<48:11:50, 45.26s/it]
                                                         
{'loss': '0.1615', 'grad_norm': '9.379', 'learning_rate': '1.565e-05', 'epoch': '0.2332'}

 23%|██▎       | 1166/5000 [15:30:48<48:11:50, 45.26s/it][2026-05-22 07:36:44,338] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 07:36:47,298] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 23%|██▎       | 1167/5000 [15:31:22<44:25:27, 41.72s/it]
                                                         
{'loss': '0.3954', 'grad_norm': '7.754', 'learning_rate': '1.564e-05', 'epoch': '0.2334'}

 23%|██▎       | 1167/5000 [15:31:22<44:25:27, 41.72s/it][2026-05-22 07:37:16,864] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 07:37:17,035] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 23%|██▎       | 1168/5000 [15:32:00<43:16:34, 40.66s/it]
                                                         
{'loss': '0.03081', 'grad_norm': '3.999', 'learning_rate': '1.564e-05', 'epoch': '0.2336'}

 23%|██▎       | 1168/5000 [15:32:00<43:16:34, 40.66s/it][2026-05-22 07:37:58,792] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 07:37:59,276] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 23%|██▎       | 1169/5000 [15:32:54<47:36:59, 44.75s/it]
                                                         
{'loss': '0.5035', 'grad_norm': '10.4', 'learning_rate': '1.564e-05', 'epoch': '0.2338'}

 23%|██▎       | 1169/5000 [15:32:54<47:36:59, 44.75s/it][2026-05-22 07:38:54,521] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 07:38:56,210] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])

 23%|██▎       | 1170/5000 [15:33:47<50:05:26, 47.08s/it]
                                                         
{'loss': '0.3476', 'grad_norm': '11.97', 'learning_rate': '1.563e-05', 'epoch': '0.234'}

 23%|██▎       | 1170/5000 [15:33:47<50:05:26, 47.08s/it][2026-05-22 07:39:39,299] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 188])
[2026-05-22 07:39:40,024] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 154])

 23%|██▎       | 1171/5000 [15:34:57<57:31:49, 54.09s/it]
                                                         
{'loss': '1.192', 'grad_norm': '9.809', 'learning_rate': '1.563e-05', 'epoch': '0.2342'}

 23%|██▎       | 1171/5000 [15:34:57<57:31:49, 54.09s/it][2026-05-22 07:40:49,325] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 93])
[2026-05-22 07:40:50,545] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 96])

 23%|██▎       | 1172/5000 [15:36:10<63:17:47, 59.53s/it]
                                                         
{'loss': '0.9639', 'grad_norm': '9.306', 'learning_rate': '1.562e-05', 'epoch': '0.2344'}

 23%|██▎       | 1172/5000 [15:36:10<63:17:47, 59.53s/it][2026-05-22 07:42:06,122] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])
[2026-05-22 07:42:07,424] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 155])

 23%|██▎       | 1173/5000 [15:37:21<67:01:18, 63.05s/it]
                                                         
{'loss': '1.108', 'grad_norm': '9.191', 'learning_rate': '1.562e-05', 'epoch': '0.2346'}

 23%|██▎       | 1173/5000 [15:37:21<67:01:18, 63.05s/it][2026-05-22 07:43:18,085] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 96])
[2026-05-22 07:43:18,141] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 147])

 23%|██▎       | 1174/5000 [15:38:31<69:21:36, 65.26s/it]
                                                         
{'loss': '0.9017', 'grad_norm': '9.29', 'learning_rate': '1.562e-05', 'epoch': '0.2348'}

 23%|██▎       | 1174/5000 [15:38:31<69:21:36, 65.26s/it][2026-05-22 07:44:31,160] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 112])
[2026-05-22 07:44:34,957] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1024])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 131])

 24%|██▎       | 1175/5000 [15:39:43<71:27:41, 67.26s/it]
                                                         
{'loss': '1.072', 'grad_norm': '8.057', 'learning_rate': '1.561e-05', 'epoch': '0.235'}

 24%|██▎       | 1175/5000 [15:39:43<71:27:41, 67.26s/it][2026-05-22 07:45:44,222] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 186])
[2026-05-22 07:45:46,764] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 180])

 24%|██▎       | 1176/5000 [15:40:55<72:58:47, 68.70s/it]
                                                         
{'loss': '0.7138', 'grad_norm': '8.495', 'learning_rate': '1.561e-05', 'epoch': '0.2352'}

 24%|██▎       | 1176/5000 [15:40:55<72:58:47, 68.70s/it][2026-05-22 07:46:51,610] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 178])
[2026-05-22 07:46:51,973] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 109])

 24%|██▎       | 1177/5000 [15:41:42<65:56:53, 62.10s/it]
                                                         
{'loss': '0.1648', 'grad_norm': '7.111', 'learning_rate': '1.56e-05', 'epoch': '0.2354'}

 24%|██▎       | 1177/5000 [15:41:42<65:56:53, 62.10s/it][2026-05-22 07:47:38,669] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 186])
[2026-05-22 07:47:41,260] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 138])

 24%|██▎       | 1178/5000 [15:42:28<60:53:26, 57.35s/it]
                                                         
{'loss': '0.1109', 'grad_norm': '6.914', 'learning_rate': '1.56e-05', 'epoch': '0.2356'}

 24%|██▎       | 1178/5000 [15:42:28<60:53:26, 57.35s/it][2026-05-22 07:48:16,840] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 232])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 07:48:17,290] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 258])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 24%|██▎       | 1179/5000 [15:43:15<57:24:46, 54.09s/it]
                                                         
{'loss': '0.7574', 'grad_norm': '20.49', 'learning_rate': '1.56e-05', 'epoch': '0.2358'}

 24%|██▎       | 1179/5000 [15:43:15<57:24:46, 54.09s/it][2026-05-22 07:49:03,131] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 958])
[2026-05-22 07:49:03,223] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 24%|██▎       | 1180/5000 [15:44:01<54:55:54, 51.77s/it]
                                                         
{'loss': '0.7159', 'grad_norm': '19.9', 'learning_rate': '1.559e-05', 'epoch': '0.236'}

 24%|██▎       | 1180/5000 [15:44:01<54:55:54, 51.77s/it][2026-05-22 07:49:56,656] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 07:49:56,752] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 24%|██▎       | 1181/5000 [15:44:52<54:35:33, 51.46s/it]
                                                         
{'loss': '0.5127', 'grad_norm': '8.863', 'learning_rate': '1.559e-05', 'epoch': '0.2362'}

 24%|██▎       | 1181/5000 [15:44:52<54:35:33, 51.46s/it][2026-05-22 07:50:46,596] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1015])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 07:50:48,576] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 24%|██▎       | 1182/5000 [15:45:42<54:02:31, 50.96s/it]
                                                         
{'loss': '0.5104', 'grad_norm': '10.2', 'learning_rate': '1.558e-05', 'epoch': '0.2364'}

 24%|██▎       | 1182/5000 [15:45:42<54:02:31, 50.96s/it][2026-05-22 07:51:34,803] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 416])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 61])
[2026-05-22 07:51:39,052] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 70])

 24%|██▎       | 1183/5000 [15:46:28<52:27:22, 49.47s/it]
                                                         
{'loss': '0.9654', 'grad_norm': '18.68', 'learning_rate': '1.558e-05', 'epoch': '0.2366'}

 24%|██▎       | 1183/5000 [15:46:28<52:27:22, 49.47s/it][2026-05-22 07:52:21,345] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 199])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 07:52:24,365] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 480])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 69])

 24%|██▎       | 1184/5000 [15:47:16<52:00:54, 49.07s/it]
                                                         
{'loss': '0.9805', 'grad_norm': '13.72', 'learning_rate': '1.558e-05', 'epoch': '0.2368'}

 24%|██▎       | 1184/5000 [15:47:16<52:00:54, 49.07s/it][2026-05-22 07:53:06,663] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 24])
[2026-05-22 07:53:07,418] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 976])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 29])

 24%|██▎       | 1185/5000 [15:48:05<52:14:08, 49.29s/it]
                                                         
{'loss': '0.3766', 'grad_norm': '11.42', 'learning_rate': '1.557e-05', 'epoch': '0.237'}

 24%|██▎       | 1185/5000 [15:48:06<52:14:08, 49.29s/it][2026-05-22 07:53:56,603] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 34])
[2026-05-22 07:53:57,464] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 20])

 24%|██▎       | 1186/5000 [15:48:55<52:12:00, 49.27s/it]
                                                         
{'loss': '0.2997', 'grad_norm': '11.85', 'learning_rate': '1.557e-05', 'epoch': '0.2372'}

 24%|██▎       | 1186/5000 [15:48:55<52:12:00, 49.27s/it][2026-05-22 07:54:42,067] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 279])
[2026-05-22 07:54:43,414] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 433])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 24%|██▎       | 1187/5000 [15:50:05<58:49:55, 55.55s/it]
                                                         
{'loss': '0.3935', 'grad_norm': '10.54', 'learning_rate': '1.556e-05', 'epoch': '0.2374'}

 24%|██▎       | 1187/5000 [15:50:05<58:49:55, 55.55s/it][2026-05-22 07:55:53,197] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 07:55:53,838] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 492])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])

 24%|██▍       | 1188/5000 [15:51:15<63:17:34, 59.77s/it]
                                                         
{'loss': '0.49', 'grad_norm': '8.719', 'learning_rate': '1.556e-05', 'epoch': '0.2376'}

 24%|██▍       | 1188/5000 [15:51:15<63:17:34, 59.77s/it][2026-05-22 07:57:02,374] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 07:57:02,754] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])

 24%|██▍       | 1189/5000 [15:51:40<52:30:36, 49.60s/it]
                                                         
{'loss': '0.1345', 'grad_norm': '6.351', 'learning_rate': '1.556e-05', 'epoch': '0.2378'}

 24%|██▍       | 1189/5000 [15:51:40<52:30:36, 49.60s/it][2026-05-22 07:57:28,265] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 07:57:29,040] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 398])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 24%|██▍       | 1190/5000 [15:52:09<45:45:10, 43.23s/it]
                                                         
{'loss': '0.169', 'grad_norm': '7.202', 'learning_rate': '1.555e-05', 'epoch': '0.238'}

 24%|██▍       | 1190/5000 [15:52:09<45:45:10, 43.23s/it][2026-05-22 07:57:55,215] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 137])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-22 07:57:55,813] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 173])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

 24%|██▍       | 1191/5000 [15:52:54<46:24:42, 43.87s/it]
                                                         
{'loss': '0.1899', 'grad_norm': '12.87', 'learning_rate': '1.555e-05', 'epoch': '0.2382'}

 24%|██▍       | 1191/5000 [15:52:54<46:24:42, 43.87s/it][2026-05-22 07:58:40,605] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 141])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 07:58:41,169] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 173])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 125])

 24%|██▍       | 1192/5000 [15:53:40<46:59:35, 44.43s/it]
                                                         
{'loss': '0.2432', 'grad_norm': '9.913', 'learning_rate': '1.554e-05', 'epoch': '0.2384'}

 24%|██▍       | 1192/5000 [15:53:40<46:59:35, 44.43s/it][2026-05-22 07:59:28,831] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 111])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 07:59:29,233] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 101])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 24%|██▍       | 1193/5000 [15:54:11<42:49:24, 40.50s/it]
                                                         
{'loss': '0.5701', 'grad_norm': '9.65', 'learning_rate': '1.554e-05', 'epoch': '0.2386'}

 24%|██▍       | 1193/5000 [15:54:11<42:49:24, 40.50s/it][2026-05-22 08:00:00,463] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 105])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 08:00:01,587] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 102])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 24%|██▍       | 1194/5000 [15:54:46<41:03:35, 38.84s/it]
                                                         
{'loss': '0.03773', 'grad_norm': '4.103', 'learning_rate': '1.553e-05', 'epoch': '0.2388'}

 24%|██▍       | 1194/5000 [15:54:46<41:03:35, 38.84s/it]
 24%|██▍       | 1195/5000 [15:55:17<38:25:59, 36.36s/it]
                                                         
{'loss': '1.341', 'grad_norm': '21.68', 'learning_rate': '1.553e-05', 'epoch': '0.239'}

 24%|██▍       | 1195/5000 [15:55:17<38:25:59, 36.36s/it][2026-05-22 08:01:10,944] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1012])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 155])
[2026-05-22 08:01:11,678] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 391])

 24%|██▍       | 1196/5000 [15:55:51<37:49:07, 35.79s/it]
                                                         
{'loss': '1.805', 'grad_norm': '31.13', 'learning_rate': '1.553e-05', 'epoch': '0.2392'}

 24%|██▍       | 1196/5000 [15:55:51<37:49:07, 35.79s/it][2026-05-22 08:01:38,123] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 154])
[2026-05-22 08:01:38,617] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 08:01:39,364] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 08:01:43,452] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 226])

 24%|██▍       | 1197/5000 [15:56:19<35:20:56, 33.46s/it]
                                                         
{'loss': '0.6912', 'grad_norm': '20.08', 'learning_rate': '1.552e-05', 'epoch': '0.2394'}

 24%|██▍       | 1197/5000 [15:56:19<35:20:56, 33.46s/it][2026-05-22 08:02:06,582] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 694])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 08:02:07,220] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 694])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 24%|██▍       | 1198/5000 [15:56:42<32:01:38, 30.33s/it]
                                                         
{'loss': '0.5989', 'grad_norm': '20.2', 'learning_rate': '1.552e-05', 'epoch': '0.2396'}

 24%|██▍       | 1198/5000 [15:56:42<32:01:38, 30.33s/it][2026-05-22 08:02:31,917] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 877])
[2026-05-22 08:02:33,173] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 24%|██▍       | 1199/5000 [15:57:28<36:54:37, 34.96s/it]
                                                         
{'loss': '1.355', 'grad_norm': '12.95', 'learning_rate': '1.551e-05', 'epoch': '0.2398'}

 24%|██▍       | 1199/5000 [15:57:28<36:54:37, 34.96s/it][2026-05-22 08:03:18,449] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 171])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])
[2026-05-22 08:03:18,537] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 24%|██▍       | 1200/5000 [15:58:15<40:36:22, 38.47s/it]
                                                         
{'loss': '0.9927', 'grad_norm': '13.34', 'learning_rate': '1.551e-05', 'epoch': '0.24'}

 24%|██▍       | 1200/5000 [15:58:15<40:36:22, 38.47s/it][2026-05-22 08:04:03,536] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 08:04:04,178] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 24%|██▍       | 1201/5000 [15:59:06<44:33:48, 42.23s/it]
                                                         
{'loss': '0.248', 'grad_norm': '6.887', 'learning_rate': '1.551e-05', 'epoch': '0.2402'}

 24%|██▍       | 1201/5000 [15:59:06<44:33:48, 42.23s/it][2026-05-22 08:04:55,304] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 179])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 08:04:56,834] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 528])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 24%|██▍       | 1202/5000 [15:59:56<47:07:27, 44.67s/it]
                                                         
{'loss': '0.3501', 'grad_norm': '6.864', 'learning_rate': '1.55e-05', 'epoch': '0.2404'}

 24%|██▍       | 1202/5000 [15:59:56<47:07:27, 44.67s/it][2026-05-22 08:05:48,305] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])
[2026-05-22 08:05:48,948] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 24%|██▍       | 1203/5000 [16:00:31<43:59:30, 41.71s/it]
                                                         
{'loss': '3.666', 'grad_norm': '3.789', 'learning_rate': '1.55e-05', 'epoch': '0.2406'}

 24%|██▍       | 1203/5000 [16:00:31<43:59:30, 41.71s/it][2026-05-22 08:06:23,817] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 984])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 08:06:26,086] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 24%|██▍       | 1204/5000 [16:01:08<42:27:12, 40.26s/it]
                                                         
{'loss': '3.683', 'grad_norm': '4.366', 'learning_rate': '1.549e-05', 'epoch': '0.2408'}

 24%|██▍       | 1204/5000 [16:01:08<42:27:12, 40.26s/it][2026-05-22 08:06:56,999] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 08:06:57,519] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 24%|██▍       | 1205/5000 [16:01:55<44:42:15, 42.41s/it]
                                                         
{'loss': '0.6532', 'grad_norm': '7.911', 'learning_rate': '1.549e-05', 'epoch': '0.241'}

 24%|██▍       | 1205/5000 [16:01:55<44:42:15, 42.41s/it][2026-05-22 08:07:43,827] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 08:07:45,040] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 24%|██▍       | 1206/5000 [16:02:46<47:13:55, 44.82s/it]
                                                         
{'loss': '0.9215', 'grad_norm': '8.578', 'learning_rate': '1.549e-05', 'epoch': '0.2412'}

 24%|██▍       | 1206/5000 [16:02:46<47:13:55, 44.82s/it][2026-05-22 08:08:44,931] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 295])
[2026-05-22 08:08:46,637] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])

 24%|██▍       | 1207/5000 [16:03:37<49:23:15, 46.87s/it]
                                                         
{'loss': '0.9444', 'grad_norm': '8.531', 'learning_rate': '1.548e-05', 'epoch': '0.2414'}

 24%|██▍       | 1207/5000 [16:03:37<49:23:15, 46.87s/it][2026-05-22 08:09:37,974] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 283])
[2026-05-22 08:09:38,559] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 254])

 24%|██▍       | 1208/5000 [16:04:33<52:17:28, 49.64s/it]
                                                         
{'loss': '0.6659', 'grad_norm': '8.346', 'learning_rate': '1.548e-05', 'epoch': '0.2416'}

 24%|██▍       | 1208/5000 [16:04:33<52:17:28, 49.64s/it][2026-05-22 08:10:21,290] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 257])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 283])
[2026-05-22 08:10:21,414] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 24%|██▍       | 1209/5000 [16:05:42<58:23:27, 55.45s/it]
                                                         
{'loss': '1.015', 'grad_norm': '12.19', 'learning_rate': '1.547e-05', 'epoch': '0.2418'}

 24%|██▍       | 1209/5000 [16:05:42<58:23:27, 55.45s/it][2026-05-22 08:11:30,726] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 08:11:31,191] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 24%|██▍       | 1210/5000 [16:06:51<62:41:18, 59.55s/it]
                                                         
{'loss': '1.263', 'grad_norm': '13.67', 'learning_rate': '1.547e-05', 'epoch': '0.242'}

 24%|██▍       | 1210/5000 [16:06:51<62:41:18, 59.55s/it][2026-05-22 08:12:38,813] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 416])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 08:12:39,956] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 295])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 24%|██▍       | 1211/5000 [16:07:37<58:10:35, 55.27s/it]
                                                         
{'loss': '0.4525', 'grad_norm': '23.17', 'learning_rate': '1.547e-05', 'epoch': '0.2422'}

 24%|██▍       | 1211/5000 [16:07:37<58:10:35, 55.27s/it][2026-05-22 08:13:24,995] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 538])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 08:13:25,232] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 269])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 24%|██▍       | 1212/5000 [16:08:22<55:08:28, 52.40s/it]
                                                         
{'loss': '0.421', 'grad_norm': '12.75', 'learning_rate': '1.546e-05', 'epoch': '0.2424'}

 24%|██▍       | 1212/5000 [16:08:22<55:08:28, 52.40s/it][2026-05-22 08:14:09,847] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 415])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 67])
[2026-05-22 08:14:10,402] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])

 24%|██▍       | 1213/5000 [16:09:11<53:46:45, 51.12s/it]
                                                         
{'loss': '0.3198', 'grad_norm': '7.181', 'learning_rate': '1.546e-05', 'epoch': '0.2426'}

 24%|██▍       | 1213/5000 [16:09:11<53:46:45, 51.12s/it][2026-05-22 08:14:57,854] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 182])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 74])
[2026-05-22 08:15:00,029] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 56])

 24%|██▍       | 1214/5000 [16:10:00<53:07:44, 50.52s/it]
                                                         
{'loss': '0.2171', 'grad_norm': '5.308', 'learning_rate': '1.545e-05', 'epoch': '0.2428'}

 24%|██▍       | 1214/5000 [16:10:00<53:07:44, 50.52s/it][2026-05-22 08:15:47,031] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 52])
[2026-05-22 08:15:47,701] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 67])

 24%|██▍       | 1215/5000 [16:10:33<47:37:27, 45.30s/it]
                                                         
{'loss': '0.3987', 'grad_norm': '9.355', 'learning_rate': '1.545e-05', 'epoch': '0.243'}

 24%|██▍       | 1215/5000 [16:10:33<47:37:27, 45.30s/it][2026-05-22 08:16:20,820] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 62])
[2026-05-22 08:16:21,166] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 498])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 98])

 24%|██▍       | 1216/5000 [16:11:06<43:55:54, 41.80s/it]
                                                         
{'loss': '0.07024', 'grad_norm': '5.395', 'learning_rate': '1.544e-05', 'epoch': '0.2432'}

 24%|██▍       | 1216/5000 [16:11:06<43:55:54, 41.80s/it][2026-05-22 08:16:53,416] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 8])
[2026-05-22 08:16:54,052] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 11])

 24%|██▍       | 1217/5000 [16:11:37<40:23:14, 38.43s/it]
                                                         
{'loss': '0.3254', 'grad_norm': '9.241', 'learning_rate': '1.544e-05', 'epoch': '0.2434'}

 24%|██▍       | 1217/5000 [16:11:37<40:23:14, 38.43s/it][2026-05-22 08:17:23,980] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 10])
[2026-05-22 08:17:24,018] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 7])

 24%|██▍       | 1218/5000 [16:12:11<39:01:39, 37.15s/it]
                                                         
{'loss': '0.03031', 'grad_norm': '4.042', 'learning_rate': '1.544e-05', 'epoch': '0.2436'}

 24%|██▍       | 1218/5000 [16:12:11<39:01:39, 37.15s/it][2026-05-22 08:18:10,065] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 08:18:10,844] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 24%|██▍       | 1219/5000 [16:12:41<36:44:12, 34.98s/it]
                                                         
{'loss': '0.122', 'grad_norm': '6.824', 'learning_rate': '1.543e-05', 'epoch': '0.2438'}

 24%|██▍       | 1219/5000 [16:12:41<36:44:12, 34.98s/it][2026-05-22 08:18:40,281] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 08:18:42,724] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 24%|██▍       | 1220/5000 [16:13:14<36:12:22, 34.48s/it]
                                                         
{'loss': '0.1194', 'grad_norm': '7.719', 'learning_rate': '1.543e-05', 'epoch': '0.244'}

 24%|██▍       | 1220/5000 [16:13:14<36:12:22, 34.48s/it][2026-05-22 08:19:08,179] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 976])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 08:19:08,530] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 24%|██▍       | 1221/5000 [16:13:46<35:22:14, 33.70s/it]
                                                         
{'loss': '0.2265', 'grad_norm': '13.7', 'learning_rate': '1.542e-05', 'epoch': '0.2442'}

 24%|██▍       | 1221/5000 [16:13:46<35:22:14, 33.70s/it][2026-05-22 08:19:39,561] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-22 08:19:41,411] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 978])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

 24%|██▍       | 1222/5000 [16:14:19<35:06:51, 33.46s/it]
                                                         
{'loss': '0.1695', 'grad_norm': '9.995', 'learning_rate': '1.542e-05', 'epoch': '0.2444'}

 24%|██▍       | 1222/5000 [16:14:19<35:06:51, 33.46s/it][2026-05-22 08:20:07,001] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 290])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])
[2026-05-22 08:20:08,623] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 317])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 24%|██▍       | 1223/5000 [16:14:47<33:20:21, 31.78s/it]
                                                         
{'loss': '1.047', 'grad_norm': '17.76', 'learning_rate': '1.542e-05', 'epoch': '0.2446'}

 24%|██▍       | 1223/5000 [16:14:47<33:20:21, 31.78s/it][2026-05-22 08:20:36,971] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 246])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])
[2026-05-22 08:20:37,479] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 333])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])

 24%|██▍       | 1224/5000 [16:15:15<32:12:29, 30.71s/it]
                                                         
{'loss': '1.157', 'grad_norm': '24.35', 'learning_rate': '1.541e-05', 'epoch': '0.2448'}

 24%|██▍       | 1224/5000 [16:15:15<32:12:29, 30.71s/it][2026-05-22 08:21:06,137] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 106])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 980])
[2026-05-22 08:21:06,486] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 104])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 24%|██▍       | 1225/5000 [16:16:29<45:47:29, 43.67s/it]
                                                         
{'loss': '0.2537', 'grad_norm': '7.297', 'learning_rate': '1.541e-05', 'epoch': '0.245'}

 24%|██▍       | 1225/5000 [16:16:29<45:47:29, 43.67s/it][2026-05-22 08:22:20,247] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 98])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 08:22:21,016] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 91])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 25%|██▍       | 1226/5000 [16:17:42<54:53:05, 52.35s/it]
                                                         
{'loss': '0.732', 'grad_norm': '7.743', 'learning_rate': '1.54e-05', 'epoch': '0.2452'}

 25%|██▍       | 1226/5000 [16:17:42<54:53:05, 52.35s/it][2026-05-22 08:23:29,129] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 186])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])
[2026-05-22 08:23:30,281] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 179])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 215])

 25%|██▍       | 1227/5000 [16:18:52<60:23:15, 57.62s/it]
                                                         
{'loss': '0.4251', 'grad_norm': '16.84', 'learning_rate': '1.54e-05', 'epoch': '0.2454'}

 25%|██▍       | 1227/5000 [16:18:52<60:23:15, 57.62s/it][2026-05-22 08:24:39,907] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 219])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-22 08:24:40,063] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 169])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])

 25%|██▍       | 1228/5000 [16:20:01<64:03:57, 61.14s/it]
                                                         
{'loss': '0.6737', 'grad_norm': '9.644', 'learning_rate': '1.54e-05', 'epoch': '0.2456'}

 25%|██▍       | 1228/5000 [16:20:01<64:03:57, 61.14s/it][2026-05-22 08:25:48,152] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 281])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 08:25:48,806] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 238])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 25%|██▍       | 1229/5000 [16:20:46<59:00:57, 56.34s/it]
                                                         
{'loss': '0.4503', 'grad_norm': '8.833', 'learning_rate': '1.539e-05', 'epoch': '0.2458'}

 25%|██▍       | 1229/5000 [16:20:46<59:00:57, 56.34s/it][2026-05-22 08:26:33,211] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 245])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 08:26:33,262] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 205])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 25%|██▍       | 1230/5000 [16:21:34<56:25:26, 53.88s/it]
                                                         
{'loss': '0.6737', 'grad_norm': '10.09', 'learning_rate': '1.539e-05', 'epoch': '0.246'}

 25%|██▍       | 1230/5000 [16:21:34<56:25:26, 53.88s/it][2026-05-22 08:27:23,230] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-22 08:27:23,998] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 295])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 25%|██▍       | 1231/5000 [16:22:23<54:37:48, 52.18s/it]
                                                         
{'loss': '1.339', 'grad_norm': '16.18', 'learning_rate': '1.538e-05', 'epoch': '0.2462'}

 25%|██▍       | 1231/5000 [16:22:23<54:37:48, 52.18s/it][2026-05-22 08:28:11,730] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 282])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 938])
[2026-05-22 08:28:13,068] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 303])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 25%|██▍       | 1232/5000 [16:23:11<53:23:33, 51.01s/it]
                                                         
{'loss': '1.018', 'grad_norm': '10.13', 'learning_rate': '1.538e-05', 'epoch': '0.2464'}

 25%|██▍       | 1232/5000 [16:23:11<53:23:33, 51.01s/it][2026-05-22 08:29:04,917] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 08:29:05,213] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 211])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

 25%|██▍       | 1233/5000 [16:23:37<45:37:06, 43.60s/it]
                                                         
{'loss': '0.409', 'grad_norm': '9.137', 'learning_rate': '1.538e-05', 'epoch': '0.2466'}

 25%|██▍       | 1233/5000 [16:23:37<45:37:06, 43.60s/it][2026-05-22 08:29:30,415] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 08:29:32,034] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 25%|██▍       | 1234/5000 [16:24:04<40:14:29, 38.47s/it]
                                                         
{'loss': '0.421', 'grad_norm': '8.856', 'learning_rate': '1.537e-05', 'epoch': '0.2468'}

 25%|██▍       | 1234/5000 [16:24:04<40:14:29, 38.47s/it][2026-05-22 08:30:05,024] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 285])
[2026-05-22 08:30:05,898] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 368])

 25%|██▍       | 1235/5000 [16:24:38<38:58:11, 37.26s/it]
                                                         
{'loss': '0.2689', 'grad_norm': '9.63', 'learning_rate': '1.537e-05', 'epoch': '0.247'}

 25%|██▍       | 1235/5000 [16:24:38<38:58:11, 37.26s/it][2026-05-22 08:30:39,584] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 155])
[2026-05-22 08:30:40,468] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 206])

 25%|██▍       | 1236/5000 [16:25:13<38:11:10, 36.52s/it]
                                                         
{'loss': '0.3474', 'grad_norm': '10.32', 'learning_rate': '1.536e-05', 'epoch': '0.2472'}

 25%|██▍       | 1236/5000 [16:25:13<38:11:10, 36.52s/it][2026-05-22 08:31:15,251] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 330])
[2026-05-22 08:31:18,082] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 132])

 25%|██▍       | 1237/5000 [16:26:04<42:42:15, 40.85s/it]
                                                         
{'loss': '0.5112', 'grad_norm': '7.094', 'learning_rate': '1.536e-05', 'epoch': '0.2474'}

 25%|██▍       | 1237/5000 [16:26:04<42:42:15, 40.85s/it][2026-05-22 08:32:05,493] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1022])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 153])
[2026-05-22 08:32:07,460] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 295])

 25%|██▍       | 1238/5000 [16:26:56<46:11:36, 44.20s/it]
                                                         
{'loss': '0.3144', 'grad_norm': '7.806', 'learning_rate': '1.536e-05', 'epoch': '0.2476'}

 25%|██▍       | 1238/5000 [16:26:56<46:11:36, 44.20s/it][2026-05-22 08:32:51,942] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 279])
[2026-05-22 08:32:54,399] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 275])

 25%|██▍       | 1239/5000 [16:27:47<48:24:05, 46.33s/it]
                                                         
{'loss': '0.8022', 'grad_norm': '9.908', 'learning_rate': '1.535e-05', 'epoch': '0.2478'}

 25%|██▍       | 1239/5000 [16:27:47<48:24:05, 46.33s/it][2026-05-22 08:33:45,407] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])
[2026-05-22 08:33:46,461] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])

 25%|██▍       | 1240/5000 [16:28:38<49:55:36, 47.80s/it]
                                                         
{'loss': '0.176', 'grad_norm': '6.179', 'learning_rate': '1.535e-05', 'epoch': '0.248'}

 25%|██▍       | 1240/5000 [16:28:38<49:55:36, 47.80s/it][2026-05-22 08:34:25,364] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 466])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 230])
[2026-05-22 08:34:25,779] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 254])

 25%|██▍       | 1241/5000 [16:29:29<50:51:08, 48.70s/it]
                                                         
{'loss': '0.3495', 'grad_norm': '7.302', 'learning_rate': '1.534e-05', 'epoch': '0.2482'}

 25%|██▍       | 1241/5000 [16:29:29<50:51:08, 48.70s/it][2026-05-22 08:35:16,191] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])
[2026-05-22 08:35:16,261] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])

 25%|██▍       | 1242/5000 [16:30:19<51:08:19, 48.99s/it]
                                                         
{'loss': '0.5208', 'grad_norm': '10.51', 'learning_rate': '1.534e-05', 'epoch': '0.2484'}

 25%|██▍       | 1242/5000 [16:30:19<51:08:19, 48.99s/it][2026-05-22 08:36:06,594] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 554])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 80])
[2026-05-22 08:36:07,234] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 964])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 96])

 25%|██▍       | 1243/5000 [16:31:10<51:44:42, 49.58s/it]
                                                         
{'loss': '0.3615', 'grad_norm': '9.384', 'learning_rate': '1.533e-05', 'epoch': '0.2486'}

 25%|██▍       | 1243/5000 [16:31:10<51:44:42, 49.58s/it][2026-05-22 08:36:58,018] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 812])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 62])
[2026-05-22 08:36:59,160] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 971])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 125])

 25%|██▍       | 1244/5000 [16:32:01<52:07:29, 49.96s/it]
                                                         
{'loss': '0.4866', 'grad_norm': '9.76', 'learning_rate': '1.533e-05', 'epoch': '0.2488'}

 25%|██▍       | 1244/5000 [16:32:01<52:07:29, 49.96s/it][2026-05-22 08:37:48,775] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 970])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 08:37:48,886] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 551])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 25%|██▍       | 1245/5000 [16:32:31<45:58:37, 44.08s/it]
                                                         
{'loss': '0.4899', 'grad_norm': '9.849', 'learning_rate': '1.533e-05', 'epoch': '0.249'}

 25%|██▍       | 1245/5000 [16:32:31<45:58:37, 44.08s/it][2026-05-22 08:38:19,199] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 981])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 08:38:19,315] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 926])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 25%|██▍       | 1246/5000 [16:33:02<41:43:32, 40.01s/it]
                                                         
{'loss': '0.09233', 'grad_norm': '5.998', 'learning_rate': '1.532e-05', 'epoch': '0.2492'}

 25%|██▍       | 1246/5000 [16:33:02<41:43:32, 40.01s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 08:38:53,345] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 174])
[2026-05-22 08:38:54,127] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 194])

 25%|██▍       | 1247/5000 [16:33:32<38:48:59, 37.23s/it]
                                                         
{'loss': '0.9444', 'grad_norm': '21.51', 'learning_rate': '1.532e-05', 'epoch': '0.2494'}

 25%|██▍       | 1247/5000 [16:33:32<38:48:59, 37.23s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 08:39:24,124] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 986])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 221])
[2026-05-22 08:39:24,233] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 272])

 25%|██▍       | 1248/5000 [16:34:03<36:52:25, 35.38s/it]
                                                         
{'loss': '0.5191', 'grad_norm': '17.32', 'learning_rate': '1.531e-05', 'epoch': '0.2496'}

 25%|██▍       | 1248/5000 [16:34:03<36:52:25, 35.38s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 08:39:50,240] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 290])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 207])
[2026-05-22 08:39:50,844] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 249])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 25%|██▍       | 1249/5000 [16:34:47<39:25:32, 37.84s/it]
                                                         
{'loss': '0.3236', 'grad_norm': '8.225', 'learning_rate': '1.531e-05', 'epoch': '0.2498'}

 25%|██▍       | 1249/5000 [16:34:47<39:25:32, 37.84s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 08:40:34,447] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 253])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 08:40:34,620] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 280])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 25%|██▌       | 1250/5000 [16:35:31<41:25:04, 39.76s/it]
                                                         
{'loss': '0.1323', 'grad_norm': '6.962', 'learning_rate': '1.531e-05', 'epoch': '0.25'}

 25%|██▌       | 1250/5000 [16:35:31<41:25:04, 39.76s/it][2026-05-22 08:41:16,303] INFO [src.utils.basic_utils:19] Saving model to /workspace/VLM2Vec/outputs/Qwen3.5-08b-continued-mixed-alltasks/checkpoint-1250
[2026-05-22 08:41:16,475] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-22 08:41:16,487] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-22 08:41:20,641] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-22 08:41:21,022] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 25%|██▌       | 1251/5000 [16:36:21<44:34:23, 42.80s/it]
                                                         
{'loss': '0.371', 'grad_norm': '9.303', 'learning_rate': '1.53e-05', 'epoch': '0.2502'}

 25%|██▌       | 1251/5000 [16:36:21<44:34:23, 42.80s/it][2026-05-22 08:42:10,455] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-22 08:42:10,532] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 25%|██▌       | 1252/5000 [16:37:11<46:48:58, 44.97s/it]
                                                         
{'loss': '0.1243', 'grad_norm': '14.64', 'learning_rate': '1.53e-05', 'epoch': '0.2504'}

 25%|██▌       | 1252/5000 [16:37:11<46:48:58, 44.97s/it][2026-05-22 08:43:14,237] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])
[2026-05-22 08:43:16,432] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 184])

 25%|██▌       | 1253/5000 [16:38:00<48:01:23, 46.14s/it]
                                                         
{'loss': '0.198', 'grad_norm': '13.53', 'learning_rate': '1.529e-05', 'epoch': '0.2506'}

 25%|██▌       | 1253/5000 [16:38:00<48:01:23, 46.14s/it][2026-05-22 08:44:01,977] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 08:44:05,602] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 158])

 25%|██▌       | 1254/5000 [16:38:50<49:20:32, 47.42s/it]
                                                         
{'loss': '0.1937', 'grad_norm': '10.61', 'learning_rate': '1.529e-05', 'epoch': '0.2508'}

 25%|██▌       | 1254/5000 [16:38:50<49:20:32, 47.42s/it][2026-05-22 08:44:47,912] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 196])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])
[2026-05-22 08:44:51,233] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 156])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 25%|██▌       | 1255/5000 [16:39:19<43:36:47, 41.92s/it]
                                                         
{'loss': '0.2507', 'grad_norm': '7.564', 'learning_rate': '1.529e-05', 'epoch': '0.251'}

 25%|██▌       | 1255/5000 [16:39:19<43:36:47, 41.92s/it][2026-05-22 08:45:18,900] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 164])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 08:45:20,676] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 188])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 25%|██▌       | 1256/5000 [16:39:49<39:52:44, 38.35s/it]
                                                         
{'loss': '0.1107', 'grad_norm': '6.853', 'learning_rate': '1.528e-05', 'epoch': '0.2512'}

 25%|██▌       | 1256/5000 [16:39:49<39:52:44, 38.35s/it][2026-05-22 08:45:38,264] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 290])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 958])
[2026-05-22 08:45:38,920] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 25%|██▌       | 1257/5000 [16:40:35<42:10:58, 40.57s/it]
                                                         
{'loss': '0.6058', 'grad_norm': '20.3', 'learning_rate': '1.528e-05', 'epoch': '0.2514'}

 25%|██▌       | 1257/5000 [16:40:35<42:10:58, 40.57s/it][2026-05-22 08:46:25,117] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 299])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 08:46:25,390] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 285])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 25%|██▌       | 1258/5000 [16:41:23<44:27:42, 42.77s/it]
                                                         
{'loss': '0.4575', 'grad_norm': '15.32', 'learning_rate': '1.527e-05', 'epoch': '0.2516'}

 25%|██▌       | 1258/5000 [16:41:23<44:27:42, 42.77s/it][2026-05-22 08:47:12,460] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 08:47:13,616] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 25%|██▌       | 1259/5000 [16:42:13<46:44:54, 44.99s/it]
                                                         
{'loss': '0.4138', 'grad_norm': '9.376', 'learning_rate': '1.527e-05', 'epoch': '0.2518'}

 25%|██▌       | 1259/5000 [16:42:13<46:44:54, 44.99s/it][2026-05-22 08:48:03,837] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 08:48:03,972] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 25%|██▌       | 1260/5000 [16:43:03<48:11:02, 46.38s/it]
                                                         
{'loss': '0.4268', 'grad_norm': '13.64', 'learning_rate': '1.527e-05', 'epoch': '0.252'}

 25%|██▌       | 1260/5000 [16:43:03<48:11:02, 46.38s/it][2026-05-22 08:49:18,258] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 08:49:18,687] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 25%|██▌       | 1261/5000 [16:43:48<47:38:43, 45.87s/it]
                                                         
{'loss': '0.525', 'grad_norm': '10.03', 'learning_rate': '1.526e-05', 'epoch': '0.2522'}

 25%|██▌       | 1261/5000 [16:43:48<47:38:43, 45.87s/it][2026-05-22 08:49:58,539] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 08:50:03,756] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 25%|██▌       | 1262/5000 [16:44:34<47:56:02, 46.16s/it]
                                                         
{'loss': '0.7168', 'grad_norm': '10.91', 'learning_rate': '1.526e-05', 'epoch': '0.2524'}

 25%|██▌       | 1262/5000 [16:44:34<47:56:02, 46.16s/it][2026-05-22 08:50:21,121] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 148])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 23])
[2026-05-22 08:50:22,055] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 164])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 27])

 25%|██▌       | 1263/5000 [16:45:19<47:23:33, 45.66s/it]
                                                         
{'loss': '0.4509', 'grad_norm': '7.24', 'learning_rate': '1.525e-05', 'epoch': '0.2526'}

 25%|██▌       | 1263/5000 [16:45:19<47:23:33, 45.66s/it][2026-05-22 08:51:07,365] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 180])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 29])
[2026-05-22 08:51:07,585] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 132])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 30])

 25%|██▌       | 1264/5000 [16:46:06<47:41:04, 45.95s/it]
                                                         
{'loss': '0.2486', 'grad_norm': '7.31', 'learning_rate': '1.525e-05', 'epoch': '0.2528'}

 25%|██▌       | 1264/5000 [16:46:06<47:41:04, 45.95s/it][2026-05-22 08:51:52,620] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 271])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 08:51:53,227] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 245])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 25%|██▌       | 1265/5000 [16:46:55<48:37:08, 46.86s/it]
                                                         
{'loss': '0.4038', 'grad_norm': '8.476', 'learning_rate': '1.524e-05', 'epoch': '0.253'}

 25%|██▌       | 1265/5000 [16:46:55<48:37:08, 46.86s/it][2026-05-22 08:52:41,602] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 220])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 08:52:41,649] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 217])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 25%|██▌       | 1266/5000 [16:47:45<49:43:28, 47.94s/it]
                                                         
{'loss': '0.1444', 'grad_norm': '5.272', 'learning_rate': '1.524e-05', 'epoch': '0.2532'}

 25%|██▌       | 1266/5000 [16:47:45<49:43:28, 47.94s/it][2026-05-22 08:53:32,366] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])
[2026-05-22 08:53:33,268] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 286])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 25%|██▌       | 1267/5000 [16:48:55<56:30:45, 54.50s/it]
                                                         
{'loss': '0.4058', 'grad_norm': '11.92', 'learning_rate': '1.524e-05', 'epoch': '0.2534'}

 25%|██▌       | 1267/5000 [16:48:55<56:30:45, 54.50s/it][2026-05-22 08:54:43,156] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 241])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 08:54:44,028] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 484])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 25%|██▌       | 1268/5000 [16:50:05<61:17:30, 59.12s/it]
                                                         
{'loss': '0.7641', 'grad_norm': '7.543', 'learning_rate': '1.523e-05', 'epoch': '0.2536'}

 25%|██▌       | 1268/5000 [16:50:05<61:17:30, 59.12s/it][2026-05-22 08:55:52,081] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 291])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])
[2026-05-22 08:55:53,183] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 25%|██▌       | 1269/5000 [16:50:27<49:55:43, 48.18s/it]
                                                         
{'loss': '0.7065', 'grad_norm': '13.29', 'learning_rate': '1.523e-05', 'epoch': '0.2538'}

 25%|██▌       | 1269/5000 [16:50:27<49:55:43, 48.18s/it][2026-05-22 08:56:15,649] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 08:56:15,797] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 433])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 25%|██▌       | 1270/5000 [16:50:51<42:11:40, 40.72s/it]
                                                         
{'loss': '0.7602', 'grad_norm': '13.1', 'learning_rate': '1.522e-05', 'epoch': '0.254'}

 25%|██▌       | 1270/5000 [16:50:51<42:11:40, 40.72s/it][2026-05-22 08:56:40,912] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 90])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 958])
[2026-05-22 08:56:43,747] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 110])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 25%|██▌       | 1271/5000 [16:51:21<39:04:02, 37.72s/it]
                                                         
{'loss': '0.2682', 'grad_norm': '9.942', 'learning_rate': '1.522e-05', 'epoch': '0.2542'}

 25%|██▌       | 1271/5000 [16:51:21<39:04:02, 37.72s/it][2026-05-22 08:57:12,914] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 105])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 08:57:13,980] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 103])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 25%|██▌       | 1272/5000 [16:51:51<36:29:12, 35.23s/it]
                                                         
{'loss': '0.4089', 'grad_norm': '12.95', 'learning_rate': '1.522e-05', 'epoch': '0.2544'}

 25%|██▌       | 1272/5000 [16:51:51<36:29:12, 35.23s/it][2026-05-22 08:57:41,687] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 102])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 08:57:42,046] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 114])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 25%|██▌       | 1273/5000 [16:52:22<35:20:40, 34.14s/it]
                                                         
{'loss': '0.1643', 'grad_norm': '6.245', 'learning_rate': '1.521e-05', 'epoch': '0.2546'}

 25%|██▌       | 1273/5000 [16:52:22<35:20:40, 34.14s/it][2026-05-22 08:58:14,524] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 105])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-22 08:58:14,786] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 100])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 25%|██▌       | 1274/5000 [16:52:56<35:13:58, 34.04s/it]
                                                         
{'loss': '0.06836', 'grad_norm': '5.341', 'learning_rate': '1.521e-05', 'epoch': '0.2548'}

 25%|██▌       | 1274/5000 [16:52:56<35:13:58, 34.04s/it][2026-05-22 08:58:50,285] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1018])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 08:58:50,928] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 26%|██▌       | 1275/5000 [16:53:31<35:25:01, 34.23s/it]
                                                         
{'loss': '0.2782', 'grad_norm': '7.73', 'learning_rate': '1.52e-05', 'epoch': '0.255'}

 26%|██▌       | 1275/5000 [16:53:31<35:25:01, 34.23s/it][2026-05-22 08:59:24,865] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 08:59:25,451] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 26%|██▌       | 1276/5000 [16:54:05<35:31:22, 34.34s/it]
                                                         
{'loss': '0.05063', 'grad_norm': '4.241', 'learning_rate': '1.52e-05', 'epoch': '0.2552'}

 26%|██▌       | 1276/5000 [16:54:05<35:31:22, 34.34s/it][2026-05-22 09:00:01,139] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 115])
[2026-05-22 09:00:02,519] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])

 26%|██▌       | 1277/5000 [16:54:54<39:58:53, 38.66s/it]
                                                         
{'loss': '0.5075', 'grad_norm': '12.09', 'learning_rate': '1.52e-05', 'epoch': '0.2554'}

 26%|██▌       | 1277/5000 [16:54:54<39:58:53, 38.66s/it][2026-05-22 09:00:51,782] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 113])
[2026-05-22 09:00:52,438] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 162])

 26%|██▌       | 1278/5000 [16:55:43<43:01:08, 41.61s/it]
                                                         
{'loss': '1.613', 'grad_norm': '9.128', 'learning_rate': '1.519e-05', 'epoch': '0.2556'}

 26%|██▌       | 1278/5000 [16:55:43<43:01:08, 41.61s/it][2026-05-22 09:01:30,394] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 53])
[2026-05-22 09:01:30,496] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 89])

 26%|██▌       | 1279/5000 [16:56:30<44:39:45, 43.21s/it]
                                                         
{'loss': '1.121', 'grad_norm': '12.11', 'learning_rate': '1.519e-05', 'epoch': '0.2558'}

 26%|██▌       | 1279/5000 [16:56:30<44:39:45, 43.21s/it][2026-05-22 09:02:17,833] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 50])
[2026-05-22 09:02:18,876] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 788])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 73])

 26%|██▌       | 1280/5000 [16:57:18<46:16:18, 44.78s/it]
                                                         
{'loss': '1.121', 'grad_norm': '10.89', 'learning_rate': '1.518e-05', 'epoch': '0.256'}

 26%|██▌       | 1280/5000 [16:57:18<46:16:18, 44.78s/it][2026-05-22 09:03:06,839] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 229])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 09:03:09,927] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 987])

 26%|██▌       | 1281/5000 [16:58:29<54:24:39, 52.67s/it]
                                                         
{'loss': '0.4684', 'grad_norm': '11.5', 'learning_rate': '1.518e-05', 'epoch': '0.2562'}

 26%|██▌       | 1281/5000 [16:58:29<54:24:39, 52.67s/it][2026-05-22 09:04:18,000] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 167])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 09:04:18,062] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 431])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 26%|██▌       | 1282/5000 [16:59:38<59:27:48, 57.58s/it]
                                                         
{'loss': '0.7501', 'grad_norm': '10.02', 'learning_rate': '1.518e-05', 'epoch': '0.2564'}

 26%|██▌       | 1282/5000 [16:59:38<59:27:48, 57.58s/it][2026-05-22 09:05:27,105] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 417])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])
[2026-05-22 09:05:30,909] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 229])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])

 26%|██▌       | 1283/5000 [17:00:23<55:38:25, 53.89s/it]
                                                         
{'loss': '0.8068', 'grad_norm': '13.12', 'learning_rate': '1.517e-05', 'epoch': '0.2566'}

 26%|██▌       | 1283/5000 [17:00:24<55:38:25, 53.89s/it][2026-05-22 09:06:13,256] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 498])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 09:06:13,511] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 185])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 26%|██▌       | 1284/5000 [17:01:10<53:20:48, 51.68s/it]
                                                         
{'loss': '0.9744', 'grad_norm': '14.13', 'learning_rate': '1.517e-05', 'epoch': '0.2568'}

 26%|██▌       | 1284/5000 [17:01:10<53:20:48, 51.68s/it][2026-05-22 09:06:58,651] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 158])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 09:07:01,255] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 151])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 26%|██▌       | 1285/5000 [17:01:40<46:27:28, 45.02s/it]
                                                         
{'loss': '0.8493', 'grad_norm': '14.02', 'learning_rate': '1.516e-05', 'epoch': '0.257'}

 26%|██▌       | 1285/5000 [17:01:40<46:27:28, 45.02s/it][2026-05-22 09:07:27,860] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 160])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 09:07:28,455] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 217])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 26%|██▌       | 1286/5000 [17:02:18<44:33:36, 43.19s/it]
                                                         
{'loss': '0.8444', 'grad_norm': '17.33', 'learning_rate': '1.516e-05', 'epoch': '0.2572'}

 26%|██▌       | 1286/5000 [17:02:18<44:33:36, 43.19s/it][2026-05-22 09:08:11,371] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 141])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 84])
[2026-05-22 09:08:11,645] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 157])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 136])

 26%|██▌       | 1287/5000 [17:03:06<45:46:13, 44.38s/it]
                                                         
{'loss': '0.746', 'grad_norm': '8.146', 'learning_rate': '1.516e-05', 'epoch': '0.2574'}

 26%|██▌       | 1287/5000 [17:03:06<45:46:13, 44.38s/it][2026-05-22 09:08:57,951] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 197])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 09:08:58,321] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 165])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 178])

 26%|██▌       | 1288/5000 [17:03:54<47:07:53, 45.71s/it]
                                                         
{'loss': '0.893', 'grad_norm': '9.656', 'learning_rate': '1.515e-05', 'epoch': '0.2576'}

 26%|██▌       | 1288/5000 [17:03:54<47:07:53, 45.71s/it][2026-05-22 09:09:42,359] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 645])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 09:09:42,411] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 661])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 26%|██▌       | 1289/5000 [17:04:22<41:30:27, 40.27s/it]
                                                         
{'loss': '0.161', 'grad_norm': '7.888', 'learning_rate': '1.515e-05', 'epoch': '0.2578'}

 26%|██▌       | 1289/5000 [17:04:22<41:30:27, 40.27s/it][2026-05-22 09:10:10,037] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 661])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 09:10:10,097] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 653])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 26%|██▌       | 1290/5000 [17:04:51<38:06:19, 36.98s/it]
                                                         
{'loss': '0.1323', 'grad_norm': '6.188', 'learning_rate': '1.514e-05', 'epoch': '0.258'}

 26%|██▌       | 1290/5000 [17:04:51<38:06:19, 36.98s/it][2026-05-22 09:10:39,851] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 775])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])
[2026-05-22 09:10:39,855] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])

 26%|██▌       | 1291/5000 [17:05:39<41:28:59, 40.26s/it]
                                                         
{'loss': '0.5125', 'grad_norm': '10.93', 'learning_rate': '1.514e-05', 'epoch': '0.2582'}

 26%|██▌       | 1291/5000 [17:05:39<41:28:59, 40.26s/it][2026-05-22 09:11:27,653] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 971])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 84])
[2026-05-22 09:11:28,181] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 969])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 54])

 26%|██▌       | 1292/5000 [17:06:26<43:34:29, 42.31s/it]
                                                         
{'loss': '0.7526', 'grad_norm': '8.066', 'learning_rate': '1.513e-05', 'epoch': '0.2584'}

 26%|██▌       | 1292/5000 [17:06:26<43:34:29, 42.31s/it][2026-05-22 09:12:17,703] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 200])
[2026-05-22 09:12:17,732] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 126])

 26%|██▌       | 1293/5000 [17:06:54<39:02:34, 37.92s/it]
                                                         
{'loss': '0.7566', 'grad_norm': '20.99', 'learning_rate': '1.513e-05', 'epoch': '0.2586'}

 26%|██▌       | 1293/5000 [17:06:54<39:02:34, 37.92s/it][2026-05-22 09:12:46,079] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1014])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 130])
[2026-05-22 09:12:46,236] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1014])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 121])

 26%|██▌       | 1294/5000 [17:07:23<36:11:59, 35.16s/it]
                                                         
{'loss': '0.5882', 'grad_norm': '18.26', 'learning_rate': '1.513e-05', 'epoch': '0.2588'}

 26%|██▌       | 1294/5000 [17:07:23<36:11:59, 35.16s/it][2026-05-22 09:13:22,693] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 09:13:24,556] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])

 26%|██▌       | 1295/5000 [17:07:58<36:14:18, 35.21s/it]
                                                         
{'loss': '2.224', 'grad_norm': '15.24', 'learning_rate': '1.512e-05', 'epoch': '0.259'}

 26%|██▌       | 1295/5000 [17:07:58<36:14:18, 35.21s/it][2026-05-22 09:13:58,395] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])
[2026-05-22 09:13:59,974] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 26%|██▌       | 1296/5000 [17:08:37<37:19:41, 36.28s/it]
                                                         
{'loss': '2.202', 'grad_norm': '15.43', 'learning_rate': '1.512e-05', 'epoch': '0.2592'}

 26%|██▌       | 1296/5000 [17:08:37<37:19:41, 36.28s/it][2026-05-22 09:14:36,782] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 242])
[2026-05-22 09:14:39,169] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 182])

 26%|██▌       | 1297/5000 [17:09:25<40:56:44, 39.81s/it]
                                                         
{'loss': '0.7313', 'grad_norm': '15.54', 'learning_rate': '1.511e-05', 'epoch': '0.2594'}

 26%|██▌       | 1297/5000 [17:09:25<40:56:44, 39.81s/it][2026-05-22 09:15:27,238] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 149])
[2026-05-22 09:15:27,304] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 195])

 26%|██▌       | 1298/5000 [17:10:13<43:34:57, 42.38s/it]
                                                         
{'loss': '0.4244', 'grad_norm': '13.83', 'learning_rate': '1.511e-05', 'epoch': '0.2596'}

 26%|██▌       | 1298/5000 [17:10:13<43:34:57, 42.38s/it][2026-05-22 09:16:12,695] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 90])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 09:16:13,575] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 128])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 26%|██▌       | 1299/5000 [17:11:03<45:50:00, 44.58s/it]
                                                         
{'loss': '0.2007', 'grad_norm': '10.95', 'learning_rate': '1.511e-05', 'epoch': '0.2598'}

 26%|██▌       | 1299/5000 [17:11:03<45:50:00, 44.58s/it][2026-05-22 09:17:01,497] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 136])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 09:17:02,449] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 134])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

 26%|██▌       | 1300/5000 [17:11:52<47:16:35, 46.00s/it]
                                                         
{'loss': '0.2436', 'grad_norm': '4.908', 'learning_rate': '1.51e-05', 'epoch': '0.26'}

 26%|██▌       | 1300/5000 [17:11:52<47:16:35, 46.00s/it][2026-05-22 09:17:39,142] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 229])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 09:17:39,267] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 415])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 26%|██▌       | 1301/5000 [17:13:04<55:04:03, 53.59s/it]
                                                         
{'loss': '0.7243', 'grad_norm': '11.26', 'learning_rate': '1.51e-05', 'epoch': '0.2602'}

 26%|██▌       | 1301/5000 [17:13:04<55:04:03, 53.59s/it][2026-05-22 09:18:50,498] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 171])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 111])
[2026-05-22 09:18:51,634] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 430])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 26%|██▌       | 1302/5000 [17:14:14<60:11:51, 58.60s/it]
                                                         
{'loss': '0.5433', 'grad_norm': '9.531', 'learning_rate': '1.509e-05', 'epoch': '0.2604'}

 26%|██▌       | 1302/5000 [17:14:14<60:11:51, 58.60s/it][2026-05-22 09:20:01,852] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 286])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])
[2026-05-22 09:20:02,483] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 26%|██▌       | 1303/5000 [17:15:03<57:21:54, 55.86s/it]
                                                         
{'loss': '0.4774', 'grad_norm': '10.34', 'learning_rate': '1.509e-05', 'epoch': '0.2606'}

 26%|██▌       | 1303/5000 [17:15:03<57:21:54, 55.86s/it][2026-05-22 09:20:52,074] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 09:20:52,172] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 507])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 26%|██▌       | 1304/5000 [17:15:53<55:28:40, 54.04s/it]
                                                         
{'loss': '0.5568', 'grad_norm': '11.52', 'learning_rate': '1.509e-05', 'epoch': '0.2608'}

 26%|██▌       | 1304/5000 [17:15:53<55:28:40, 54.04s/it][2026-05-22 09:21:45,053] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 283])
[2026-05-22 09:21:45,178] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 280])

 26%|██▌       | 1305/5000 [17:16:42<53:46:25, 52.39s/it]
                                                         
{'loss': '1.531', 'grad_norm': '15.11', 'learning_rate': '1.508e-05', 'epoch': '0.261'}

 26%|██▌       | 1305/5000 [17:16:42<53:46:25, 52.39s/it][2026-05-22 09:22:33,137] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 290])
[2026-05-22 09:22:33,397] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])

 26%|██▌       | 1306/5000 [17:17:31<52:49:51, 51.49s/it]
                                                         
{'loss': '1.148', 'grad_norm': '14.12', 'learning_rate': '1.508e-05', 'epoch': '0.2612'}

 26%|██▌       | 1306/5000 [17:17:31<52:49:51, 51.49s/it][2026-05-22 09:23:27,546] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 09:23:27,763] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 314])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])

 26%|██▌       | 1307/5000 [17:17:57<45:02:39, 43.91s/it]
                                                         
{'loss': '0.2341', 'grad_norm': '10.08', 'learning_rate': '1.507e-05', 'epoch': '0.2614'}

 26%|██▌       | 1307/5000 [17:17:57<45:02:39, 43.91s/it][2026-05-22 09:23:54,606] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 220])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 09:23:56,963] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 430])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])

 26%|██▌       | 1308/5000 [17:18:24<39:44:45, 38.76s/it]
                                                         
{'loss': '0.1358', 'grad_norm': '8.792', 'learning_rate': '1.507e-05', 'epoch': '0.2616'}

 26%|██▌       | 1308/5000 [17:18:24<39:44:45, 38.76s/it][2026-05-22 09:24:15,186] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 152])
[2026-05-22 09:24:15,383] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 197])

 26%|██▌       | 1309/5000 [17:18:56<37:32:49, 36.62s/it]
                                                         
{'loss': '0.747', 'grad_norm': '10.89', 'learning_rate': '1.507e-05', 'epoch': '0.2618'}

 26%|██▌       | 1309/5000 [17:18:56<37:32:49, 36.62s/it][2026-05-22 09:24:46,894] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 225])
[2026-05-22 09:24:47,701] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 164])

 26%|██▌       | 1310/5000 [17:19:31<37:02:03, 36.13s/it]
                                                         
{'loss': '0.03098', 'grad_norm': '4.143', 'learning_rate': '1.506e-05', 'epoch': '0.262'}

 26%|██▌       | 1310/5000 [17:19:31<37:02:03, 36.13s/it][2026-05-22 09:25:26,909] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 133])
[2026-05-22 09:25:30,670] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 237])

 26%|██▌       | 1311/5000 [17:20:20<40:57:34, 39.97s/it]
                                                         
{'loss': '0.3082', 'grad_norm': '8.848', 'learning_rate': '1.506e-05', 'epoch': '0.2622'}

 26%|██▌       | 1311/5000 [17:20:20<40:57:34, 39.97s/it][2026-05-22 09:26:18,424] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 09:26:19,101] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 79])

 26%|██▌       | 1312/5000 [17:21:08<43:34:22, 42.53s/it]
                                                         
{'loss': '0.2322', 'grad_norm': '8.151', 'learning_rate': '1.505e-05', 'epoch': '0.2624'}

 26%|██▌       | 1312/5000 [17:21:08<43:34:22, 42.53s/it][2026-05-22 09:27:07,335] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 249])
[2026-05-22 09:27:07,733] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])

 26%|██▋       | 1313/5000 [17:21:39<40:01:01, 39.07s/it]
                                                         
{'loss': '0.3132', 'grad_norm': '10.77', 'learning_rate': '1.505e-05', 'epoch': '0.2626'}

 26%|██▋       | 1313/5000 [17:21:39<40:01:01, 39.07s/it][2026-05-22 09:27:38,644] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 281])
[2026-05-22 09:27:40,481] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 241])

 26%|██▋       | 1314/5000 [17:22:10<37:37:12, 36.74s/it]
                                                         
{'loss': '0.2228', 'grad_norm': '9.695', 'learning_rate': '1.504e-05', 'epoch': '0.2628'}

 26%|██▋       | 1314/5000 [17:22:10<37:37:12, 36.74s/it][2026-05-22 09:28:01,288] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 09:28:01,884] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 26%|██▋       | 1315/5000 [17:23:00<41:39:58, 40.71s/it]
                                                         
{'loss': '0.2085', 'grad_norm': '7.009', 'learning_rate': '1.504e-05', 'epoch': '0.263'}

 26%|██▋       | 1315/5000 [17:23:00<41:39:58, 40.71s/it][2026-05-22 09:28:49,226] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 816])
[2026-05-22 09:28:49,725] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 493])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 813])

 26%|██▋       | 1316/5000 [17:23:49<44:00:20, 43.00s/it]
                                                         
{'loss': '0.1899', 'grad_norm': '9.529', 'learning_rate': '1.504e-05', 'epoch': '0.2632'}

 26%|██▋       | 1316/5000 [17:23:49<44:00:20, 43.00s/it][2026-05-22 09:29:35,993] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 246])
[2026-05-22 09:29:36,809] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 472])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 294])

 26%|██▋       | 1317/5000 [17:24:35<44:54:46, 43.90s/it]
                                                         
{'loss': '2.393', 'grad_norm': '7.742', 'learning_rate': '1.503e-05', 'epoch': '0.2634'}

 26%|██▋       | 1317/5000 [17:24:35<44:54:46, 43.90s/it][2026-05-22 09:30:22,748] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 326])
[2026-05-22 09:30:22,794] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 303])

 26%|██▋       | 1318/5000 [17:25:20<45:17:42, 44.29s/it]
                                                         
{'loss': '2.425', 'grad_norm': '10.18', 'learning_rate': '1.503e-05', 'epoch': '0.2636'}

 26%|██▋       | 1318/5000 [17:25:20<45:17:42, 44.29s/it][2026-05-22 09:31:07,077] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 09:31:07,096] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 330])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])

 26%|██▋       | 1319/5000 [17:26:06<45:46:30, 44.77s/it]
                                                         
{'loss': '0.2719', 'grad_norm': '6.003', 'learning_rate': '1.502e-05', 'epoch': '0.2638'}

 26%|██▋       | 1319/5000 [17:26:06<45:46:30, 44.77s/it][2026-05-22 09:31:53,071] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 166])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 09:31:53,526] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 496])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 26%|██▋       | 1320/5000 [17:26:53<46:26:16, 45.43s/it]
                                                         
{'loss': '0.1686', 'grad_norm': '5.252', 'learning_rate': '1.502e-05', 'epoch': '0.264'}

 26%|██▋       | 1320/5000 [17:26:53<46:26:16, 45.43s/it][2026-05-22 09:32:43,944] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 129])
[2026-05-22 09:32:44,218] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 286])

 26%|██▋       | 1321/5000 [17:27:44<48:12:07, 47.17s/it]
                                                         
{'loss': '1.005', 'grad_norm': '8.124', 'learning_rate': '1.502e-05', 'epoch': '0.2642'}

 26%|██▋       | 1321/5000 [17:27:44<48:12:07, 47.17s/it][2026-05-22 09:33:35,261] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 128])
[2026-05-22 09:33:35,966] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 126])

 26%|██▋       | 1322/5000 [17:28:31<48:18:52, 47.29s/it]
                                                         
{'loss': '0.7463', 'grad_norm': '11.71', 'learning_rate': '1.501e-05', 'epoch': '0.2644'}

 26%|██▋       | 1322/5000 [17:28:31<48:18:52, 47.29s/it][2026-05-22 09:34:43,411] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 633])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 26%|██▋       | 1323/5000 [17:29:01<42:51:02, 41.95s/it]
                                                         
{'loss': '0.2938', 'grad_norm': '6.969', 'learning_rate': '1.501e-05', 'epoch': '0.2646'}

 26%|██▋       | 1323/5000 [17:29:01<42:51:02, 41.95s/it][2026-05-22 09:34:47,895] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])
[2026-05-22 09:35:12,005] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 975])
[2026-05-22 09:35:14,508] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 26%|██▋       | 1324/5000 [17:29:36<40:47:44, 39.95s/it]
                                                         
{'loss': '0.06946', 'grad_norm': '4.378', 'learning_rate': '1.5e-05', 'epoch': '0.2648'}

 26%|██▋       | 1324/5000 [17:29:36<40:47:44, 39.95s/it][2026-05-22 09:35:23,643] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])
[2026-05-22 09:35:23,741] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 969])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 64])

 26%|██▋       | 1325/5000 [17:30:02<36:22:14, 35.63s/it]
                                                         
{'loss': '0.2277', 'grad_norm': '7.573', 'learning_rate': '1.5e-05', 'epoch': '0.265'}

 26%|██▋       | 1325/5000 [17:30:02<36:22:14, 35.63s/it][2026-05-22 09:35:49,436] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 969])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 53])
[2026-05-22 09:35:50,236] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 69])

 27%|██▋       | 1326/5000 [17:30:31<34:21:07, 33.66s/it]
                                                         
{'loss': '0.2753', 'grad_norm': '7.758', 'learning_rate': '1.5e-05', 'epoch': '0.2652'}

 27%|██▋       | 1326/5000 [17:30:31<34:21:07, 33.66s/it][2026-05-22 09:36:19,423] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 69])
[2026-05-22 09:36:19,761] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 218])

 27%|██▋       | 1327/5000 [17:31:19<38:53:55, 38.13s/it]
                                                         
{'loss': '0.1685', 'grad_norm': '6.85', 'learning_rate': '1.499e-05', 'epoch': '0.2654'}

 27%|██▋       | 1327/5000 [17:31:19<38:53:55, 38.13s/it][2026-05-22 09:37:08,054] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 45])
[2026-05-22 09:37:08,205] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 256])

 27%|██▋       | 1328/5000 [17:32:06<41:36:23, 40.79s/it]
                                                         
{'loss': '0.2686', 'grad_norm': '5.378', 'learning_rate': '1.499e-05', 'epoch': '0.2656'}

 27%|██▋       | 1328/5000 [17:32:06<41:36:23, 40.79s/it][2026-05-22 09:37:58,296] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 136])
[2026-05-22 09:37:59,388] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 97])

 27%|██▋       | 1329/5000 [17:33:15<50:02:02, 49.07s/it]
                                                         
{'loss': '1.28', 'grad_norm': '11.74', 'learning_rate': '1.498e-05', 'epoch': '0.2658'}

 27%|██▋       | 1329/5000 [17:33:15<50:02:02, 49.07s/it][2026-05-22 09:39:06,466] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 97])
[2026-05-22 09:39:06,960] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 128])

 27%|██▋       | 1330/5000 [17:34:27<57:13:41, 56.14s/it]
                                                         
{'loss': '0.2368', 'grad_norm': '9.587', 'learning_rate': '1.498e-05', 'epoch': '0.266'}

 27%|██▋       | 1330/5000 [17:34:27<57:13:41, 56.14s/it][2026-05-22 09:40:21,547] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 09:40:22,954] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 253])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 27%|██▋       | 1331/5000 [17:35:11<53:24:52, 52.41s/it]
                                                         
{'loss': '0.1811', 'grad_norm': '8.185', 'learning_rate': '1.498e-05', 'epoch': '0.2662'}

 27%|██▋       | 1331/5000 [17:35:11<53:24:52, 52.41s/it][2026-05-22 09:41:07,410] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 977])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 09:41:09,484] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 27%|██▋       | 1332/5000 [17:35:57<51:26:32, 50.49s/it]
                                                         
{'loss': '0.2016', 'grad_norm': '9.492', 'learning_rate': '1.497e-05', 'epoch': '0.2664'}

 27%|██▋       | 1332/5000 [17:35:57<51:26:32, 50.49s/it][2026-05-22 09:41:46,129] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 09:41:47,611] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 311])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])

 27%|██▋       | 1333/5000 [17:36:46<50:57:40, 50.03s/it]
                                                         
{'loss': '2.709', 'grad_norm': '25.68', 'learning_rate': '1.497e-05', 'epoch': '0.2666'}

 27%|██▋       | 1333/5000 [17:36:46<50:57:40, 50.03s/it][2026-05-22 09:42:34,706] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 09:42:36,581] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 234])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 27%|██▋       | 1334/5000 [17:37:36<50:55:17, 50.00s/it]
                                                         
{'loss': '3.144', 'grad_norm': '26.27', 'learning_rate': '1.496e-05', 'epoch': '0.2668'}

 27%|██▋       | 1334/5000 [17:37:36<50:55:17, 50.00s/it][2026-05-22 09:43:37,781] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 163])
[2026-05-22 09:43:38,223] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 292])

 27%|██▋       | 1335/5000 [17:38:26<50:46:35, 49.88s/it]
                                                         
{'loss': '1.443', 'grad_norm': '6.515', 'learning_rate': '1.496e-05', 'epoch': '0.267'}

 27%|██▋       | 1335/5000 [17:38:26<50:46:35, 49.88s/it][2026-05-22 09:44:26,282] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 09:44:28,160] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 121])

 27%|██▋       | 1336/5000 [17:39:14<50:25:49, 49.55s/it]
                                                         
{'loss': '1.648', 'grad_norm': '21.28', 'learning_rate': '1.496e-05', 'epoch': '0.2672'}

 27%|██▋       | 1336/5000 [17:39:14<50:25:49, 49.55s/it][2026-05-22 09:45:05,827] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 09:45:06,033] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1017])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 112])

 27%|██▋       | 1337/5000 [17:39:44<44:26:45, 43.68s/it]
                                                         
{'loss': '0.8747', 'grad_norm': '14.89', 'learning_rate': '1.495e-05', 'epoch': '0.2674'}

 27%|██▋       | 1337/5000 [17:39:44<44:26:45, 43.68s/it][2026-05-22 09:45:35,714] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 410])
[2026-05-22 09:45:36,168] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 195])

 27%|██▋       | 1338/5000 [17:40:37<47:14:46, 46.45s/it]
                                                         
{'loss': '1.156', 'grad_norm': '20.36', 'learning_rate': '1.495e-05', 'epoch': '0.2676'}

 27%|██▋       | 1338/5000 [17:40:37<47:14:46, 46.45s/it][2026-05-22 09:46:33,324] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 259])
[2026-05-22 09:46:38,300] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])

 27%|██▋       | 1339/5000 [17:41:27<48:14:45, 47.44s/it]
                                                         
{'loss': '0.7159', 'grad_norm': '9.479', 'learning_rate': '1.494e-05', 'epoch': '0.2678'}

 27%|██▋       | 1339/5000 [17:41:27<48:14:45, 47.44s/it][2026-05-22 09:47:25,753] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 270])
[2026-05-22 09:47:27,109] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1043])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])

 27%|██▋       | 1340/5000 [17:42:16<48:45:59, 47.97s/it]
                                                         
{'loss': '0.3621', 'grad_norm': '9.538', 'learning_rate': '1.494e-05', 'epoch': '0.268'}

 27%|██▋       | 1340/5000 [17:42:16<48:45:59, 47.97s/it][2026-05-22 09:48:15,510] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 983])
[2026-05-22 09:48:16,128] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 27%|██▋       | 1341/5000 [17:43:05<49:07:39, 48.34s/it]
                                                         
{'loss': '0.2215', 'grad_norm': '7.463', 'learning_rate': '1.493e-05', 'epoch': '0.2682'}

 27%|██▋       | 1341/5000 [17:43:05<49:07:39, 48.34s/it][2026-05-22 09:49:04,145] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 09:49:06,358] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 27%|██▋       | 1342/5000 [17:43:55<49:22:47, 48.60s/it]
                                                         
{'loss': '0.2397', 'grad_norm': '8.862', 'learning_rate': '1.493e-05', 'epoch': '0.2684'}

 27%|██▋       | 1342/5000 [17:43:55<49:22:47, 48.60s/it][2026-05-22 09:49:51,242] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 09:49:52,600] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 27%|██▋       | 1343/5000 [17:44:43<49:16:31, 48.51s/it]
                                                         
{'loss': '0.2563', 'grad_norm': '6.162', 'learning_rate': '1.493e-05', 'epoch': '0.2686'}

 27%|██▋       | 1343/5000 [17:44:43<49:16:31, 48.51s/it][2026-05-22 09:50:38,503] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 09:50:38,601] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])

 27%|██▋       | 1344/5000 [17:45:34<50:08:36, 49.38s/it]
                                                         
{'loss': '0.2146', 'grad_norm': '8.405', 'learning_rate': '1.492e-05', 'epoch': '0.2688'}

 27%|██▋       | 1344/5000 [17:45:34<50:08:36, 49.38s/it][2026-05-22 09:51:31,797] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 471])
[2026-05-22 09:51:32,022] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 145])

 27%|██▋       | 1345/5000 [17:46:22<49:27:41, 48.72s/it]
                                                         
{'loss': '0.4171', 'grad_norm': '9.794', 'learning_rate': '1.492e-05', 'epoch': '0.269'}

 27%|██▋       | 1345/5000 [17:46:22<49:27:41, 48.72s/it][2026-05-22 09:52:17,911] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 133])
[2026-05-22 09:52:19,160] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1019])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 147])

 27%|██▋       | 1346/5000 [17:47:12<49:52:37, 49.14s/it]
                                                         
{'loss': '0.1209', 'grad_norm': '5.619', 'learning_rate': '1.491e-05', 'epoch': '0.2692'}

 27%|██▋       | 1346/5000 [17:47:12<49:52:37, 49.14s/it][2026-05-22 09:53:09,739] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 252])
[2026-05-22 09:53:10,467] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 265])

 27%|██▋       | 1347/5000 [17:48:21<56:06:17, 55.29s/it]
                                                         
{'loss': '1.032', 'grad_norm': '7.978', 'learning_rate': '1.491e-05', 'epoch': '0.2694'}

 27%|██▋       | 1347/5000 [17:48:21<56:06:17, 55.29s/it][2026-05-22 09:54:21,060] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 183])
[2026-05-22 09:54:21,804] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 145])

 27%|██▋       | 1348/5000 [17:49:31<60:36:51, 59.75s/it]
                                                         
{'loss': '0.4803', 'grad_norm': '10.25', 'learning_rate': '1.491e-05', 'epoch': '0.2696'}

 27%|██▋       | 1348/5000 [17:49:31<60:36:51, 59.75s/it][2026-05-22 09:55:20,804] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 92])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 09:55:21,257] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 99])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])

 27%|██▋       | 1349/5000 [17:50:41<63:35:26, 62.70s/it]
                                                         
{'loss': '1.224', 'grad_norm': '8.332', 'learning_rate': '1.49e-05', 'epoch': '0.2698'}

 27%|██▋       | 1349/5000 [17:50:41<63:35:26, 62.70s/it][2026-05-22 09:56:28,813] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 187])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 09:56:29,529] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 101])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 27%|██▋       | 1350/5000 [17:51:51<65:47:49, 64.90s/it]
                                                         
{'loss': '0.8418', 'grad_norm': '8.909', 'learning_rate': '1.49e-05', 'epoch': '0.27'}

 27%|██▋       | 1350/5000 [17:51:51<65:47:49, 64.90s/it][2026-05-22 09:57:47,335] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 09:57:47,419] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 27%|██▋       | 1351/5000 [17:52:41<61:15:55, 60.44s/it]
                                                         
{'loss': '0.1156', 'grad_norm': '5.223', 'learning_rate': '1.489e-05', 'epoch': '0.2702'}

 27%|██▋       | 1351/5000 [17:52:41<61:15:55, 60.44s/it][2026-05-22 09:58:35,293] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 09:58:37,961] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 27%|██▋       | 1352/5000 [17:53:28<57:06:45, 56.36s/it]
                                                         
{'loss': '0.1327', 'grad_norm': '6.582', 'learning_rate': '1.489e-05', 'epoch': '0.2704'}

 27%|██▋       | 1352/5000 [17:53:28<57:06:45, 56.36s/it][2026-05-22 09:59:19,430] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1015])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 141])
[2026-05-22 09:59:22,648] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 102])

 27%|██▋       | 1353/5000 [17:54:19<55:28:52, 54.77s/it]
                                                         
{'loss': '0.4277', 'grad_norm': '12.23', 'learning_rate': '1.489e-05', 'epoch': '0.2706'}

 27%|██▋       | 1353/5000 [17:54:19<55:28:52, 54.77s/it][2026-05-22 10:00:10,643] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 109])
[2026-05-22 10:00:13,219] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 27%|██▋       | 1354/5000 [17:55:10<54:11:32, 53.51s/it]
                                                         
{'loss': '0.4311', 'grad_norm': '9.616', 'learning_rate': '1.488e-05', 'epoch': '0.2708'}

 27%|██▋       | 1354/5000 [17:55:10<54:11:32, 53.51s/it][2026-05-22 10:01:04,909] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 286])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 246])
[2026-05-22 10:01:06,635] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 27%|██▋       | 1355/5000 [17:55:56<52:07:19, 51.48s/it]
                                                         
{'loss': '0.9297', 'grad_norm': '13.97', 'learning_rate': '1.488e-05', 'epoch': '0.271'}

 27%|██▋       | 1355/5000 [17:55:56<52:07:19, 51.48s/it][2026-05-22 10:01:53,337] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 10:01:54,868] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 27%|██▋       | 1356/5000 [17:56:43<50:37:37, 50.02s/it]
                                                         
{'loss': '0.9837', 'grad_norm': '12.94', 'learning_rate': '1.487e-05', 'epoch': '0.2712'}

 27%|██▋       | 1356/5000 [17:56:43<50:37:37, 50.02s/it][2026-05-22 10:02:30,329] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 417])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])
[2026-05-22 10:02:30,495] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 242])

 27%|██▋       | 1357/5000 [17:57:20<46:48:53, 46.26s/it]
                                                         
{'loss': '1.784', 'grad_norm': '7.727', 'learning_rate': '1.487e-05', 'epoch': '0.2714'}

 27%|██▋       | 1357/5000 [17:57:20<46:48:53, 46.26s/it][2026-05-22 10:03:08,137] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 254])
[2026-05-22 10:03:09,076] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 504])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 291])

 27%|██▋       | 1358/5000 [17:57:58<44:01:35, 43.52s/it]
                                                         
{'loss': '1.536', 'grad_norm': '3.37', 'learning_rate': '1.487e-05', 'epoch': '0.2716'}

 27%|██▋       | 1358/5000 [17:57:58<44:01:35, 43.52s/it][2026-05-22 10:03:47,756] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 96])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 10:03:49,179] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 106])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 27%|██▋       | 1359/5000 [17:58:48<46:04:56, 45.56s/it]
                                                         
{'loss': '1.773', 'grad_norm': '7.518', 'learning_rate': '1.486e-05', 'epoch': '0.2718'}

 27%|██▋       | 1359/5000 [17:58:48<46:04:56, 45.56s/it][2026-05-22 10:04:37,170] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 97])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 973])
[2026-05-22 10:04:42,456] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 104])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 27%|██▋       | 1360/5000 [17:59:37<47:10:51, 46.66s/it]
                                                         
{'loss': '1.854', 'grad_norm': '14.52', 'learning_rate': '1.486e-05', 'epoch': '0.272'}

 27%|██▋       | 1360/5000 [17:59:37<47:10:51, 46.66s/it][2026-05-22 10:05:31,004] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])
[2026-05-22 10:05:31,174] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 27%|██▋       | 1361/5000 [18:00:15<44:22:19, 43.90s/it]
                                                         
{'loss': '0.4397', 'grad_norm': '9.667', 'learning_rate': '1.485e-05', 'epoch': '0.2722'}

 27%|██▋       | 1361/5000 [18:00:15<44:22:19, 43.90s/it][2026-05-22 10:06:09,303] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 10:06:10,340] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1031])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 27%|██▋       | 1362/5000 [18:00:48<41:14:03, 40.80s/it]
                                                         
{'loss': '0.0271', 'grad_norm': '4.177', 'learning_rate': '1.485e-05', 'epoch': '0.2724'}

 27%|██▋       | 1362/5000 [18:00:48<41:14:03, 40.80s/it][2026-05-22 10:06:57,263] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 123])
[2026-05-22 10:06:57,919] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 27%|██▋       | 1363/5000 [18:01:17<37:30:52, 37.13s/it]
                                                         
{'loss': '0.2286', 'grad_norm': '8.012', 'learning_rate': '1.484e-05', 'epoch': '0.2726'}

 27%|██▋       | 1363/5000 [18:01:17<37:30:52, 37.13s/it][2026-05-22 10:07:26,119] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 81])
[2026-05-22 10:07:27,164] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])

 27%|██▋       | 1364/5000 [18:01:50<36:11:17, 35.83s/it]
                                                         
{'loss': '0.02539', 'grad_norm': '3.98', 'learning_rate': '1.484e-05', 'epoch': '0.2728'}

 27%|██▋       | 1364/5000 [18:01:50<36:11:17, 35.83s/it][2026-05-22 10:07:45,820] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 233])
[2026-05-22 10:07:47,594] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 266])

 27%|██▋       | 1365/5000 [18:02:38<40:02:42, 39.66s/it]
                                                         
{'loss': '1.072', 'grad_norm': '11.26', 'learning_rate': '1.484e-05', 'epoch': '0.273'}

 27%|██▋       | 1365/5000 [18:02:38<40:02:42, 39.66s/it][2026-05-22 10:08:36,210] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 283])
[2026-05-22 10:08:37,315] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 299])

 27%|██▋       | 1366/5000 [18:03:29<43:34:01, 43.16s/it]
                                                         
{'loss': '0.8653', 'grad_norm': '11.47', 'learning_rate': '1.483e-05', 'epoch': '0.2732'}

 27%|██▋       | 1366/5000 [18:03:29<43:34:01, 43.16s/it][2026-05-22 10:09:32,010] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])
[2026-05-22 10:09:32,610] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])

 27%|██▋       | 1367/5000 [18:04:43<52:39:52, 52.19s/it]
                                                         
{'loss': '0.3023', 'grad_norm': '8.79', 'learning_rate': '1.483e-05', 'epoch': '0.2734'}

 27%|██▋       | 1367/5000 [18:04:43<52:39:52, 52.19s/it][2026-05-22 10:10:41,197] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 10:10:44,133] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 27%|██▋       | 1368/5000 [18:06:01<60:25:39, 59.90s/it]
                                                         
{'loss': '0.4212', 'grad_norm': '10.93', 'learning_rate': '1.482e-05', 'epoch': '0.2736'}

 27%|██▋       | 1368/5000 [18:06:01<60:25:39, 59.90s/it][2026-05-22 10:11:50,717] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 10:11:51,741] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 27%|██▋       | 1369/5000 [18:06:49<57:04:05, 56.58s/it]
                                                         
{'loss': '0.147', 'grad_norm': '6.366', 'learning_rate': '1.482e-05', 'epoch': '0.2738'}

 27%|██▋       | 1369/5000 [18:06:49<57:04:05, 56.58s/it][2026-05-22 10:12:39,371] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 10:12:39,932] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 27%|██▋       | 1370/5000 [18:07:38<54:33:25, 54.11s/it]
                                                         
{'loss': '0.155', 'grad_norm': '8.353', 'learning_rate': '1.482e-05', 'epoch': '0.274'}

 27%|██▋       | 1370/5000 [18:07:38<54:33:25, 54.11s/it][2026-05-22 10:13:27,860] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 10:13:28,438] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 27%|██▋       | 1371/5000 [18:08:28<53:14:25, 52.81s/it]
                                                         
{'loss': '0.5249', 'grad_norm': '9.552', 'learning_rate': '1.481e-05', 'epoch': '0.2742'}

 27%|██▋       | 1371/5000 [18:08:28<53:14:25, 52.81s/it][2026-05-22 10:14:18,217] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-22 10:14:18,677] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 177])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 981])

 27%|██▋       | 1372/5000 [18:09:19<52:45:39, 52.35s/it]
                                                         
{'loss': '0.1037', 'grad_norm': '4.663', 'learning_rate': '1.481e-05', 'epoch': '0.2744'}

 27%|██▋       | 1372/5000 [18:09:19<52:45:39, 52.35s/it][2026-05-22 10:15:09,868] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 388])
[2026-05-22 10:15:10,366] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 172])

 27%|██▋       | 1373/5000 [18:10:32<59:05:14, 58.65s/it]
                                                         
{'loss': '0.3352', 'grad_norm': '21.81', 'learning_rate': '1.48e-05', 'epoch': '0.2746'}

 27%|██▋       | 1373/5000 [18:10:32<59:05:14, 58.65s/it][2026-05-22 10:16:23,200] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 164])
[2026-05-22 10:16:24,608] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 203])

 27%|██▋       | 1374/5000 [18:11:44<63:05:17, 62.64s/it]
                                                         
{'loss': '0.4702', 'grad_norm': '10.94', 'learning_rate': '1.48e-05', 'epoch': '0.2748'}

 27%|██▋       | 1374/5000 [18:11:44<63:05:17, 62.64s/it][2026-05-22 10:17:44,147] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 974])
[2026-05-22 10:17:44,841] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 978])

 28%|██▊       | 1375/5000 [18:12:38<60:20:07, 59.92s/it]
                                                         
{'loss': '0.6399', 'grad_norm': '11.44', 'learning_rate': '1.48e-05', 'epoch': '0.275'}

 28%|██▊       | 1375/5000 [18:12:38<60:20:07, 59.92s/it][2026-05-22 10:18:40,719] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 10:18:41,901] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 28%|██▊       | 1376/5000 [18:13:30<57:57:01, 57.57s/it]
                                                         
{'loss': '0.4271', 'grad_norm': '6.336', 'learning_rate': '1.479e-05', 'epoch': '0.2752'}

 28%|██▊       | 1376/5000 [18:13:30<57:57:01, 57.57s/it][2026-05-22 10:19:29,949] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 136])
[2026-05-22 10:19:32,093] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 170])

 28%|██▊       | 1377/5000 [18:14:23<56:40:42, 56.32s/it]
                                                         
{'loss': '1.029', 'grad_norm': '8.601', 'learning_rate': '1.479e-05', 'epoch': '0.2754'}

 28%|██▊       | 1377/5000 [18:14:23<56:40:42, 56.32s/it][2026-05-22 10:20:26,429] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1019])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 200])
[2026-05-22 10:20:27,992] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])

 28%|██▊       | 1378/5000 [18:15:26<58:30:02, 58.15s/it]
                                                         
{'loss': '1.064', 'grad_norm': '8.37', 'learning_rate': '1.478e-05', 'epoch': '0.2756'}

 28%|██▊       | 1378/5000 [18:15:26<58:30:02, 58.15s/it][2026-05-22 10:21:33,816] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 202])
[2026-05-22 10:21:39,551] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 297])

 28%|██▊       | 1379/5000 [18:16:19<56:57:02, 56.62s/it]
                                                         
{'loss': '0.2553', 'grad_norm': '12.32', 'learning_rate': '1.478e-05', 'epoch': '0.2758'}

 28%|██▊       | 1379/5000 [18:16:19<56:57:02, 56.62s/it][2026-05-22 10:22:28,287] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 106])
[2026-05-22 10:22:30,875] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 99])

 28%|██▊       | 1380/5000 [18:17:08<54:48:54, 54.51s/it]
                                                         
{'loss': '0.3338', 'grad_norm': '6.811', 'learning_rate': '1.478e-05', 'epoch': '0.276'}

 28%|██▊       | 1380/5000 [18:17:08<54:48:54, 54.51s/it][2026-05-22 10:22:59,370] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 16])
[2026-05-22 10:22:59,675] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 16])

 28%|██▊       | 1381/5000 [18:18:20<60:07:38, 59.81s/it]
                                                         
{'loss': '1.404', 'grad_norm': '12.12', 'learning_rate': '1.477e-05', 'epoch': '0.2762'}

 28%|██▊       | 1381/5000 [18:18:20<60:07:38, 59.81s/it][2026-05-22 10:24:12,161] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 20])
[2026-05-22 10:24:12,686] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 12])

 28%|██▊       | 1382/5000 [18:19:33<63:59:15, 63.67s/it]
                                                         
{'loss': '1.06', 'grad_norm': '11.65', 'learning_rate': '1.477e-05', 'epoch': '0.2764'}

 28%|██▊       | 1382/5000 [18:19:33<63:59:15, 63.67s/it][2026-05-22 10:25:26,236] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 111])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 986])
[2026-05-22 10:25:28,136] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 108])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 984])

 28%|██▊       | 1383/5000 [18:20:27<60:53:07, 60.60s/it]
                                                         
{'loss': '0.4529', 'grad_norm': '8.331', 'learning_rate': '1.476e-05', 'epoch': '0.2766'}

 28%|██▊       | 1383/5000 [18:20:27<60:53:07, 60.60s/it][2026-05-22 10:26:20,684] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 121])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 992])
[2026-05-22 10:26:22,506] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 104])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])

 28%|██▊       | 1384/5000 [18:21:18<58:11:57, 57.94s/it]
                                                         
{'loss': '0.4449', 'grad_norm': '10.24', 'learning_rate': '1.476e-05', 'epoch': '0.2768'}

 28%|██▊       | 1384/5000 [18:21:18<58:11:57, 57.94s/it][2026-05-22 10:27:12,403] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 99])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 990])
[2026-05-22 10:27:13,922] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 105])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 28%|██▊       | 1385/5000 [18:22:08<55:46:46, 55.55s/it]
                                                         
{'loss': '0.6194', 'grad_norm': '13.14', 'learning_rate': '1.476e-05', 'epoch': '0.277'}

 28%|██▊       | 1385/5000 [18:22:08<55:46:46, 55.55s/it][2026-05-22 10:28:00,614] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 115])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 10:28:07,567] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 89])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 28%|██▊       | 1386/5000 [18:23:01<54:55:34, 54.71s/it]
                                                         
{'loss': '0.6291', 'grad_norm': '14.51', 'learning_rate': '1.475e-05', 'epoch': '0.2772'}

 28%|██▊       | 1386/5000 [18:23:01<54:55:34, 54.71s/it][2026-05-22 10:28:59,420] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 126])
[2026-05-22 10:29:04,285] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 194])

 28%|██▊       | 1387/5000 [18:23:49<52:45:45, 52.57s/it]
                                                         
{'loss': '0.4257', 'grad_norm': '11.08', 'learning_rate': '1.475e-05', 'epoch': '0.2774'}

 28%|██▊       | 1387/5000 [18:23:49<52:45:45, 52.57s/it][2026-05-22 10:29:47,395] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 122])
[2026-05-22 10:29:48,460] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 28%|██▊       | 1388/5000 [18:24:37<51:29:41, 51.32s/it]
                                                         
{'loss': '0.6108', 'grad_norm': '15.98', 'learning_rate': '1.474e-05', 'epoch': '0.2776'}

 28%|██▊       | 1388/5000 [18:24:37<51:29:41, 51.32s/it][2026-05-22 10:30:34,911] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 140])
[2026-05-22 10:30:37,753] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])

 28%|██▊       | 1389/5000 [18:25:26<50:44:18, 50.58s/it]
                                                         
{'loss': '0.6478', 'grad_norm': '9.199', 'learning_rate': '1.474e-05', 'epoch': '0.2778'}

 28%|██▊       | 1389/5000 [18:25:26<50:44:18, 50.58s/it][2026-05-22 10:31:24,848] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 149])
[2026-05-22 10:31:26,260] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 211])

 28%|██▊       | 1390/5000 [18:26:16<50:27:57, 50.33s/it]
                                                         
{'loss': '1.733', 'grad_norm': '17.22', 'learning_rate': '1.473e-05', 'epoch': '0.278'}

 28%|██▊       | 1390/5000 [18:26:16<50:27:57, 50.33s/it][2026-05-22 10:32:03,677] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 312])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 257])
[2026-05-22 10:32:04,835] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 28%|██▊       | 1391/5000 [18:27:04<49:53:54, 49.77s/it]
                                                         
{'loss': '0.3527', 'grad_norm': '12.46', 'learning_rate': '1.473e-05', 'epoch': '0.2782'}

 28%|██▊       | 1391/5000 [18:27:04<49:53:54, 49.77s/it][2026-05-22 10:32:52,418] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 454])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 10:32:52,440] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 471])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 28%|██▊       | 1392/5000 [18:27:52<49:21:01, 49.24s/it]
                                                         
{'loss': '1.467', 'grad_norm': '11.44', 'learning_rate': '1.473e-05', 'epoch': '0.2784'}

 28%|██▊       | 1392/5000 [18:27:52<49:21:01, 49.24s/it][2026-05-22 10:33:38,795] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 236])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 10:33:38,979] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 236])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 28%|██▊       | 1393/5000 [18:28:40<48:53:44, 48.80s/it]
                                                         
{'loss': '2.174', 'grad_norm': '14.51', 'learning_rate': '1.472e-05', 'epoch': '0.2786'}

 28%|██▊       | 1393/5000 [18:28:40<48:53:44, 48.80s/it][2026-05-22 10:34:27,366] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 298])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 10:34:27,462] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 294])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 28%|██▊       | 1394/5000 [18:29:27<48:18:39, 48.23s/it]
                                                         
{'loss': '2.123', 'grad_norm': '12.1', 'learning_rate': '1.472e-05', 'epoch': '0.2788'}

 28%|██▊       | 1394/5000 [18:29:27<48:18:39, 48.23s/it][2026-05-22 10:35:16,563] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 10:35:19,108] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 217])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 28%|██▊       | 1395/5000 [18:30:16<48:45:01, 48.68s/it]
                                                         
{'loss': '0.8978', 'grad_norm': '21.55', 'learning_rate': '1.471e-05', 'epoch': '0.279'}

 28%|██▊       | 1395/5000 [18:30:16<48:45:01, 48.68s/it][2026-05-22 10:36:06,815] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 856])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 10:36:07,276] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 994])

 28%|██▊       | 1396/5000 [18:31:08<49:39:29, 49.60s/it]
                                                         
{'loss': '0.7105', 'grad_norm': '20.39', 'learning_rate': '1.471e-05', 'epoch': '0.2792'}

 28%|██▊       | 1396/5000 [18:31:08<49:39:29, 49.60s/it][2026-05-22 10:36:55,645] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])
[2026-05-22 10:36:56,352] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 189])

 28%|██▊       | 1397/5000 [18:31:42<44:49:00, 44.78s/it]
                                                         
{'loss': '0.364', 'grad_norm': '7.895', 'learning_rate': '1.471e-05', 'epoch': '0.2794'}

 28%|██▊       | 1397/5000 [18:31:42<44:49:00, 44.78s/it][2026-05-22 10:37:30,554] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])
[2026-05-22 10:37:30,615] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])

 28%|██▊       | 1398/5000 [18:32:15<41:20:50, 41.32s/it]
                                                         
{'loss': '0.03485', 'grad_norm': '3.613', 'learning_rate': '1.47e-05', 'epoch': '0.2796'}

 28%|██▊       | 1398/5000 [18:32:15<41:20:50, 41.32s/it][2026-05-22 10:38:02,077] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 10:38:02,080] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])

 28%|██▊       | 1399/5000 [18:32:42<37:10:39, 37.17s/it]
                                                         
{'loss': '0.6699', 'grad_norm': '11.58', 'learning_rate': '1.47e-05', 'epoch': '0.2798'}

 28%|██▊       | 1399/5000 [18:32:42<37:10:39, 37.17s/it][2026-05-22 10:38:30,790] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 18])
[2026-05-22 10:38:30,823] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 65])

 28%|██▊       | 1400/5000 [18:33:12<34:48:08, 34.80s/it]
                                                         
{'loss': '0.2662', 'grad_norm': '8.2', 'learning_rate': '1.469e-05', 'epoch': '0.28'}

 28%|██▊       | 1400/5000 [18:33:12<34:48:08, 34.80s/it][2026-05-22 10:39:02,260] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 398])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 980])
[2026-05-22 10:39:02,747] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 28%|██▊       | 1401/5000 [18:34:02<39:20:25, 39.35s/it]
                                                         
{'loss': '0.7764', 'grad_norm': '12.97', 'learning_rate': '1.469e-05', 'epoch': '0.2802'}

 28%|██▊       | 1401/5000 [18:34:03<39:20:25, 39.35s/it][2026-05-22 10:39:51,848] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 378])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])
[2026-05-22 10:39:53,801] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 28%|██▊       | 1402/5000 [18:35:09<47:36:30, 47.63s/it]
                                                         
{'loss': '1.068', 'grad_norm': '13.52', 'learning_rate': '1.469e-05', 'epoch': '0.2804'}

 28%|██▊       | 1402/5000 [18:35:09<47:36:30, 47.63s/it][2026-05-22 10:40:56,900] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 154])
[2026-05-22 10:40:57,103] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 397])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 28%|██▊       | 1403/5000 [18:35:37<41:52:33, 41.91s/it]
                                                         
{'loss': '1.653', 'grad_norm': '26.08', 'learning_rate': '1.468e-05', 'epoch': '0.2806'}

 28%|██▊       | 1403/5000 [18:35:37<41:52:33, 41.91s/it][2026-05-22 10:41:25,366] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 170])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 112])
[2026-05-22 10:41:26,311] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 70])

 28%|██▊       | 1404/5000 [18:36:17<41:18:31, 41.35s/it]
                                                         
{'loss': '2.235', 'grad_norm': '30.41', 'learning_rate': '1.468e-05', 'epoch': '0.2808'}

 28%|██▊       | 1404/5000 [18:36:17<41:18:31, 41.35s/it][2026-05-22 10:42:06,486] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 125])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 982])
[2026-05-22 10:42:06,759] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 117])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

 28%|██▊       | 1405/5000 [18:36:46<37:36:06, 37.65s/it]
                                                         
{'loss': '1.52', 'grad_norm': '20.33', 'learning_rate': '1.467e-05', 'epoch': '0.281'}

 28%|██▊       | 1405/5000 [18:36:46<37:36:06, 37.65s/it][2026-05-22 10:42:35,343] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 165])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 10:42:37,921] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 149])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 28%|██▊       | 1406/5000 [18:37:28<38:39:57, 38.73s/it]
                                                         
{'loss': '2.015', 'grad_norm': '23.73', 'learning_rate': '1.467e-05', 'epoch': '0.2812'}

 28%|██▊       | 1406/5000 [18:37:28<38:39:57, 38.73s/it][2026-05-22 10:43:18,105] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 222])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 980])
[2026-05-22 10:43:19,337] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 229])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])

 28%|██▊       | 1407/5000 [18:38:19<42:32:58, 42.63s/it]
                                                         
{'loss': '1.011', 'grad_norm': '15.03', 'learning_rate': '1.467e-05', 'epoch': '0.2814'}

 28%|██▊       | 1407/5000 [18:38:19<42:32:58, 42.63s/it][2026-05-22 10:44:09,692] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 170])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 972])
[2026-05-22 10:44:12,522] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 502])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 985])

 28%|██▊       | 1408/5000 [18:39:11<45:22:37, 45.48s/it]
                                                         
{'loss': '1.245', 'grad_norm': '16.62', 'learning_rate': '1.466e-05', 'epoch': '0.2816'}

 28%|██▊       | 1408/5000 [18:39:11<45:22:37, 45.48s/it][2026-05-22 10:45:01,020] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 221])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 10:45:02,092] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 989])

 28%|██▊       | 1409/5000 [18:39:39<40:03:46, 40.16s/it]
                                                         
{'loss': '0.3269', 'grad_norm': '12.96', 'learning_rate': '1.466e-05', 'epoch': '0.2818'}

 28%|██▊       | 1409/5000 [18:39:39<40:03:46, 40.16s/it][2026-05-22 10:45:29,340] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 508])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 960])
[2026-05-22 10:45:31,006] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 201])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 988])

 28%|██▊       | 1410/5000 [18:40:09<36:50:58, 36.95s/it]
                                                         
{'loss': '0.2522', 'grad_norm': '14.03', 'learning_rate': '1.465e-05', 'epoch': '0.282'}

 28%|██▊       | 1410/5000 [18:40:09<36:50:58, 36.95s/it][2026-05-22 10:45:57,591] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 128])
[2026-05-22 10:45:57,683] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 28%|██▊       | 1411/5000 [18:40:55<39:42:56, 39.84s/it]
                                                         
{'loss': '1.346', 'grad_norm': '14.03', 'learning_rate': '1.465e-05', 'epoch': '0.2822'}

 28%|██▊       | 1411/5000 [18:40:55<39:42:56, 39.84s/it][2026-05-22 10:46:44,053] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1016])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 10:46:44,248] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 131])

 28%|██▊       | 1412/5000 [18:41:43<42:02:06, 42.18s/it]
                                                         
{'loss': '0.6089', 'grad_norm': '14.74', 'learning_rate': '1.464e-05', 'epoch': '0.2824'}

 28%|██▊       | 1412/5000 [18:41:43<42:02:06, 42.18s/it][2026-05-22 10:47:53,351] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 115])
[2026-05-22 10:47:53,953] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 85])

 28%|██▊       | 1413/5000 [18:42:31<43:41:15, 43.85s/it]
                                                         
{'loss': '1.223', 'grad_norm': '15.54', 'learning_rate': '1.464e-05', 'epoch': '0.2826'}

 28%|██▊       | 1413/5000 [18:42:31<43:41:15, 43.85s/it][2026-05-22 10:48:44,952] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 114])
[2026-05-22 10:48:45,003] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 94])

 28%|██▊       | 1414/5000 [18:43:23<46:10:14, 46.35s/it]
                                                         
{'loss': '0.4975', 'grad_norm': '7.989', 'learning_rate': '1.464e-05', 'epoch': '0.2828'}

 28%|██▊       | 1414/5000 [18:43:23<46:10:14, 46.35s/it][2026-05-22 10:49:22,844] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 198])
[2026-05-22 10:49:24,396] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 198])

 28%|██▊       | 1415/5000 [18:44:13<47:27:00, 47.65s/it]
                                                         
{'loss': '0.5741', 'grad_norm': '8.74', 'learning_rate': '1.463e-05', 'epoch': '0.283'}

 28%|██▊       | 1415/5000 [18:44:13<47:27:00, 47.65s/it][2026-05-22 10:50:12,946] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 108])
[2026-05-22 10:50:15,295] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 84])

 28%|██▊       | 1416/5000 [18:45:07<49:04:53, 49.30s/it]
                                                         
{'loss': '0.9796', 'grad_norm': '7.09', 'learning_rate': '1.463e-05', 'epoch': '0.2832'}

 28%|██▊       | 1416/5000 [18:45:07<49:04:53, 49.30s/it][2026-05-22 10:50:58,998] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])
[2026-05-22 10:50:59,083] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 80])

 28%|██▊       | 1417/5000 [18:45:54<48:35:27, 48.82s/it]
                                                         
{'loss': '0.3403', 'grad_norm': '6.598', 'learning_rate': '1.462e-05', 'epoch': '0.2834'}

 28%|██▊       | 1417/5000 [18:45:54<48:35:27, 48.82s/it][2026-05-22 10:51:46,719] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 173])
[2026-05-22 10:51:46,903] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 220])

 28%|██▊       | 1418/5000 [18:46:43<48:40:04, 48.91s/it]
                                                         
{'loss': '0.3257', 'grad_norm': '6.307', 'learning_rate': '1.462e-05', 'epoch': '0.2836'}

 28%|██▊       | 1418/5000 [18:46:43<48:40:04, 48.91s/it][2026-05-22 10:52:41,278] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 284])
[2026-05-22 10:52:42,182] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1014])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 28%|██▊       | 1419/5000 [18:47:29<47:35:59, 47.85s/it]
                                                         
{'loss': '0.1581', 'grad_norm': '7.605', 'learning_rate': '1.462e-05', 'epoch': '0.2838'}

 28%|██▊       | 1419/5000 [18:47:29<47:35:59, 47.85s/it][2026-05-22 10:53:26,406] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 10:53:27,662] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 28%|██▊       | 1420/5000 [18:48:15<47:06:52, 47.38s/it]
                                                         
{'loss': '0.1525', 'grad_norm': '6.575', 'learning_rate': '1.461e-05', 'epoch': '0.284'}

 28%|██▊       | 1420/5000 [18:48:15<47:06:52, 47.38s/it][2026-05-22 10:54:10,887] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])
[2026-05-22 10:54:12,322] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 80])

 28%|██▊       | 1421/5000 [18:49:04<47:33:19, 47.83s/it]
                                                         
{'loss': '1.21', 'grad_norm': '10.88', 'learning_rate': '1.461e-05', 'epoch': '0.2842'}

 28%|██▊       | 1421/5000 [18:49:04<47:33:19, 47.83s/it][2026-05-22 10:55:01,003] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 148])
[2026-05-22 10:55:03,227] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 107])

 28%|██▊       | 1422/5000 [18:49:52<47:35:30, 47.88s/it]
                                                         
{'loss': '0.9473', 'grad_norm': '8.591', 'learning_rate': '1.46e-05', 'epoch': '0.2844'}

 28%|██▊       | 1422/5000 [18:49:52<47:35:30, 47.88s/it][2026-05-22 10:55:41,097] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 10:55:41,857] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 28%|██▊       | 1423/5000 [18:50:41<47:52:13, 48.18s/it]
                                                         
{'loss': '1.01', 'grad_norm': '10.04', 'learning_rate': '1.46e-05', 'epoch': '0.2846'}

 28%|██▊       | 1423/5000 [18:50:41<47:52:13, 48.18s/it][2026-05-22 10:56:30,636] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 10:56:30,902] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 28%|██▊       | 1424/5000 [18:51:30<48:07:55, 48.46s/it]
                                                         
{'loss': '0.9563', 'grad_norm': '12.15', 'learning_rate': '1.46e-05', 'epoch': '0.2848'}

 28%|██▊       | 1424/5000 [18:51:30<48:07:55, 48.46s/it][2026-05-22 10:57:29,742] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])
[2026-05-22 10:57:30,624] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 192])

 28%|██▊       | 1425/5000 [18:52:23<49:20:46, 49.69s/it]
                                                         
{'loss': '0.2644', 'grad_norm': '8.017', 'learning_rate': '1.459e-05', 'epoch': '0.285'}

 28%|██▊       | 1425/5000 [18:52:23<49:20:46, 49.69s/it][2026-05-22 10:58:24,672] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 248])
[2026-05-22 10:58:25,522] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 190])

 29%|██▊       | 1426/5000 [18:53:16<50:24:54, 50.78s/it]
                                                         
{'loss': '0.1254', 'grad_norm': '4.158', 'learning_rate': '1.459e-05', 'epoch': '0.2852'}

 29%|██▊       | 1426/5000 [18:53:16<50:24:54, 50.78s/it][2026-05-22 10:59:10,271] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 70])
[2026-05-22 10:59:13,738] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 234])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 48])

 29%|██▊       | 1427/5000 [18:54:01<48:51:23, 49.23s/it]
                                                         
{'loss': '0.1215', 'grad_norm': '6.167', 'learning_rate': '1.458e-05', 'epoch': '0.2854'}

 29%|██▊       | 1427/5000 [18:54:01<48:51:23, 49.23s/it][2026-05-22 10:59:54,134] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 465])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])
[2026-05-22 10:59:56,735] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 203])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 44])

 29%|██▊       | 1428/5000 [18:54:48<48:06:02, 48.48s/it]
                                                         
{'loss': '0.1384', 'grad_norm': '6.6', 'learning_rate': '1.458e-05', 'epoch': '0.2856'}

 29%|██▊       | 1428/5000 [18:54:48<48:06:02, 48.48s/it][2026-05-22 11:00:36,100] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 425])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 101])
[2026-05-22 11:00:36,206] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 357])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])

 29%|██▊       | 1429/5000 [18:55:37<48:16:56, 48.67s/it]
                                                         
{'loss': '0.2886', 'grad_norm': '8.064', 'learning_rate': '1.458e-05', 'epoch': '0.2858'}

 29%|██▊       | 1429/5000 [18:55:37<48:16:56, 48.67s/it][2026-05-22 11:01:25,186] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 199])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 51])
[2026-05-22 11:01:26,944] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 478])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 85])

 29%|██▊       | 1430/5000 [18:56:29<49:09:53, 49.58s/it]
                                                         
{'loss': '0.0777', 'grad_norm': '5.009', 'learning_rate': '1.457e-05', 'epoch': '0.286'}

 29%|██▊       | 1430/5000 [18:56:29<49:09:53, 49.58s/it][2026-05-22 11:02:16,174] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 289])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 11:02:16,854] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 271])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 29%|██▊       | 1431/5000 [18:57:19<49:19:47, 49.76s/it]
                                                         
{'loss': '0.4784', 'grad_norm': '8.333', 'learning_rate': '1.457e-05', 'epoch': '0.2862'}

 29%|██▊       | 1431/5000 [18:57:19<49:19:47, 49.76s/it][2026-05-22 11:03:07,040] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 282])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 11:03:07,786] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 501])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 29%|██▊       | 1432/5000 [18:58:09<49:11:41, 49.64s/it]
                                                         
{'loss': '0.4549', 'grad_norm': '7.384', 'learning_rate': '1.456e-05', 'epoch': '0.2864'}

 29%|██▊       | 1432/5000 [18:58:09<49:11:41, 49.64s/it][2026-05-22 11:03:56,503] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 250])
[2026-05-22 11:03:57,064] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])

 29%|██▊       | 1433/5000 [18:58:32<41:28:37, 41.86s/it]
                                                         
{'loss': '0.3096', 'grad_norm': '8.951', 'learning_rate': '1.456e-05', 'epoch': '0.2866'}

 29%|██▊       | 1433/5000 [18:58:32<41:28:37, 41.86s/it][2026-05-22 11:04:20,808] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])
[2026-05-22 11:04:21,492] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])

 29%|██▊       | 1434/5000 [18:59:01<37:27:52, 37.82s/it]
                                                         
{'loss': '0.3861', 'grad_norm': '9.749', 'learning_rate': '1.456e-05', 'epoch': '0.2868'}

 29%|██▊       | 1434/5000 [18:59:01<37:27:52, 37.82s/it][2026-05-22 11:04:48,011] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 252])
[2026-05-22 11:04:48,845] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 312])

 29%|██▊       | 1435/5000 [18:59:27<33:59:19, 34.32s/it]
                                                         
{'loss': '0.1157', 'grad_norm': '7.287', 'learning_rate': '1.455e-05', 'epoch': '0.287'}

 29%|██▊       | 1435/5000 [18:59:27<33:59:19, 34.32s/it][2026-05-22 11:05:15,014] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 274])
[2026-05-22 11:05:15,094] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 450])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 283])

 29%|██▊       | 1436/5000 [18:59:56<32:22:42, 32.71s/it]
                                                         
{'loss': '0.04295', 'grad_norm': '4.215', 'learning_rate': '1.455e-05', 'epoch': '0.2872'}

 29%|██▊       | 1436/5000 [18:59:56<32:22:42, 32.71s/it][2026-05-22 11:05:48,279] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 234])
[2026-05-22 11:05:48,617] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])

 29%|██▊       | 1437/5000 [19:00:26<31:41:04, 32.01s/it]
                                                         
{'loss': '0.3523', 'grad_norm': '7.73', 'learning_rate': '1.454e-05', 'epoch': '0.2874'}

 29%|██▊       | 1437/5000 [19:00:26<31:41:04, 32.01s/it][2026-05-22 11:06:18,469] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 195])
[2026-05-22 11:06:19,358] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 274])

 29%|██▉       | 1438/5000 [19:01:01<32:24:52, 32.76s/it]
                                                         
{'loss': '0.0256', 'grad_norm': '3.576', 'learning_rate': '1.454e-05', 'epoch': '0.2876'}

 29%|██▉       | 1438/5000 [19:01:01<32:24:52, 32.76s/it]
 29%|██▉       | 1439/5000 [19:01:28<30:41:52, 31.03s/it]
                                                         
{'loss': '0.3358', 'grad_norm': '9.356', 'learning_rate': '1.453e-05', 'epoch': '0.2878'}

 29%|██▉       | 1439/5000 [19:01:28<30:41:52, 31.03s/it][2026-05-22 11:07:16,632] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 165])
[2026-05-22 11:07:18,829] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 134])
[2026-05-22 11:07:44,700] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 150])

 29%|██▉       | 1440/5000 [19:02:01<31:14:18, 31.59s/it]
                                                         
{'loss': '0.0435', 'grad_norm': '4.142', 'learning_rate': '1.453e-05', 'epoch': '0.288'}

 29%|██▉       | 1440/5000 [19:02:01<31:14:18, 31.59s/it][2026-05-22 11:07:45,871] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 230])
[2026-05-22 11:07:52,737] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 214])
[2026-05-22 11:07:52,754] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 112])

 29%|██▉       | 1441/5000 [19:02:36<32:15:05, 32.62s/it]
                                                         
{'loss': '0.4766', 'grad_norm': '9.334', 'learning_rate': '1.453e-05', 'epoch': '0.2882'}

 29%|██▉       | 1441/5000 [19:02:36<32:15:05, 32.62s/it][2026-05-22 11:08:27,115] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 297])
[2026-05-22 11:08:27,331] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 97])

 29%|██▉       | 1442/5000 [19:03:06<31:42:17, 32.08s/it]
                                                         
{'loss': '0.04871', 'grad_norm': '4.313', 'learning_rate': '1.452e-05', 'epoch': '0.2884'}

 29%|██▉       | 1442/5000 [19:03:06<31:42:17, 32.08s/it][2026-05-22 11:09:01,781] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 287])
[2026-05-22 11:09:01,799] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 79])

 29%|██▉       | 1443/5000 [19:03:55<36:40:03, 37.11s/it]
                                                         
{'loss': '0.5164', 'grad_norm': '10.69', 'learning_rate': '1.452e-05', 'epoch': '0.2886'}

 29%|██▉       | 1443/5000 [19:03:55<36:40:03, 37.11s/it][2026-05-22 11:09:49,346] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 147])
[2026-05-22 11:09:52,429] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 86])

 29%|██▉       | 1444/5000 [19:04:45<40:19:34, 40.83s/it]
                                                         
{'loss': '0.2342', 'grad_norm': '7.911', 'learning_rate': '1.451e-05', 'epoch': '0.2888'}

 29%|██▉       | 1444/5000 [19:04:45<40:19:34, 40.83s/it][2026-05-22 11:10:43,029] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 11:10:44,285] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])

 29%|██▉       | 1445/5000 [19:05:32<42:15:12, 42.79s/it]
                                                         
{'loss': '0.2738', 'grad_norm': '9', 'learning_rate': '1.451e-05', 'epoch': '0.289'}

 29%|██▉       | 1445/5000 [19:05:32<42:15:12, 42.79s/it][2026-05-22 11:11:30,933] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-22 11:11:32,059] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 144])

 29%|██▉       | 1446/5000 [19:06:20<43:49:32, 44.39s/it]
                                                         
{'loss': '0.3049', 'grad_norm': '10.32', 'learning_rate': '1.451e-05', 'epoch': '0.2892'}

 29%|██▉       | 1446/5000 [19:06:20<43:49:32, 44.39s/it][2026-05-22 11:12:07,287] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 305])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 11:12:08,344] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 343])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 29%|██▉       | 1447/5000 [19:07:08<44:45:52, 45.36s/it]
                                                         
{'loss': '0.2771', 'grad_norm': '6.971', 'learning_rate': '1.45e-05', 'epoch': '0.2894'}

 29%|██▉       | 1447/5000 [19:07:08<44:45:52, 45.36s/it][2026-05-22 11:12:56,051] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 293])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 11:12:56,148] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 321])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 29%|██▉       | 1448/5000 [19:07:56<45:42:10, 46.32s/it]
                                                         
{'loss': '0.3984', 'grad_norm': '9.48', 'learning_rate': '1.45e-05', 'epoch': '0.2896'}

 29%|██▉       | 1448/5000 [19:07:56<45:42:10, 46.32s/it][2026-05-22 11:13:43,954] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 222])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 223])
[2026-05-22 11:13:44,342] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 293])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 29%|██▉       | 1449/5000 [19:08:43<45:50:58, 46.48s/it]
                                                         
{'loss': '0.3235', 'grad_norm': '5.083', 'learning_rate': '1.449e-05', 'epoch': '0.2898'}

 29%|██▉       | 1449/5000 [19:08:43<45:50:58, 46.48s/it][2026-05-22 11:14:31,065] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 263])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 11:14:31,179] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 299])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 29%|██▉       | 1450/5000 [19:09:29<45:33:54, 46.21s/it]
                                                         
{'loss': '0.3151', 'grad_norm': '5.331', 'learning_rate': '1.449e-05', 'epoch': '0.29'}

 29%|██▉       | 1450/5000 [19:09:29<45:33:54, 46.21s/it][2026-05-22 11:15:17,779] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 169])
[2026-05-22 11:15:18,112] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 102])

 29%|██▉       | 1451/5000 [19:10:16<45:43:57, 46.39s/it]
                                                         
{'loss': '0.2755', 'grad_norm': '8.931', 'learning_rate': '1.449e-05', 'epoch': '0.2902'}

 29%|██▉       | 1451/5000 [19:10:16<45:43:57, 46.39s/it][2026-05-22 11:16:04,828] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])
[2026-05-22 11:16:04,844] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 102])

 29%|██▉       | 1452/5000 [19:11:02<45:40:51, 46.35s/it]
                                                         
{'loss': '0.3346', 'grad_norm': '8.293', 'learning_rate': '1.448e-05', 'epoch': '0.2904'}

 29%|██▉       | 1452/5000 [19:11:02<45:40:51, 46.35s/it][2026-05-22 11:16:57,839] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])
[2026-05-22 11:16:59,805] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])

 29%|██▉       | 1453/5000 [19:11:33<41:14:18, 41.85s/it]
                                                         
{'loss': '0.2217', 'grad_norm': '7.285', 'learning_rate': '1.448e-05', 'epoch': '0.2906'}

 29%|██▉       | 1453/5000 [19:11:33<41:14:18, 41.85s/it][2026-05-22 11:17:30,245] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 248])
[2026-05-22 11:17:33,224] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])

 29%|██▉       | 1454/5000 [19:12:06<38:25:43, 39.01s/it]
                                                         
{'loss': '0.04918', 'grad_norm': '4.009', 'learning_rate': '1.447e-05', 'epoch': '0.2908'}

 29%|██▉       | 1454/5000 [19:12:06<38:25:43, 39.01s/it][2026-05-22 11:17:52,814] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 289])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 11:17:53,281] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 260])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 29%|██▉       | 1455/5000 [19:12:35<35:41:17, 36.24s/it]
                                                         
{'loss': '0.3868', 'grad_norm': '8.299', 'learning_rate': '1.447e-05', 'epoch': '0.291'}

 29%|██▉       | 1455/5000 [19:12:35<35:41:17, 36.24s/it][2026-05-22 11:18:23,430] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 11:18:23,841] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 29%|██▉       | 1456/5000 [19:13:05<33:47:07, 34.32s/it]
                                                         
{'loss': '0.3094', 'grad_norm': '8.2', 'learning_rate': '1.447e-05', 'epoch': '0.2912'}

 29%|██▉       | 1456/5000 [19:13:05<33:47:07, 34.32s/it][2026-05-22 11:18:52,892] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 212])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 241])
[2026-05-22 11:18:53,519] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 569])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 428])

 29%|██▉       | 1457/5000 [19:13:51<37:09:16, 37.75s/it]
                                                         
{'loss': '0.5141', 'grad_norm': '13.41', 'learning_rate': '1.446e-05', 'epoch': '0.2914'}

 29%|██▉       | 1457/5000 [19:13:51<37:09:16, 37.75s/it][2026-05-22 11:19:39,572] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])
[2026-05-22 11:19:39,718] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 300])

 29%|██▉       | 1458/5000 [19:14:36<39:25:30, 40.07s/it]
                                                         
{'loss': '0.6092', 'grad_norm': '13.83', 'learning_rate': '1.446e-05', 'epoch': '0.2916'}

 29%|██▉       | 1458/5000 [19:14:36<39:25:30, 40.07s/it][2026-05-22 11:20:26,132] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 137])
[2026-05-22 11:20:26,907] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 111])

 29%|██▉       | 1459/5000 [19:15:25<41:54:36, 42.61s/it]
                                                         
{'loss': '0.3911', 'grad_norm': '8.902', 'learning_rate': '1.445e-05', 'epoch': '0.2918'}

 29%|██▉       | 1459/5000 [19:15:25<41:54:36, 42.61s/it][2026-05-22 11:21:14,709] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 118])
[2026-05-22 11:21:14,918] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 308])

 29%|██▉       | 1460/5000 [19:16:13<43:32:31, 44.28s/it]
                                                         
{'loss': '0.1129', 'grad_norm': '5.499', 'learning_rate': '1.445e-05', 'epoch': '0.292'}

 29%|██▉       | 1460/5000 [19:16:13<43:32:31, 44.28s/it][2026-05-22 11:22:08,791] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 86])
[2026-05-22 11:22:13,132] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 733])

 29%|██▉       | 1461/5000 [19:16:44<39:38:56, 40.33s/it]
                                                         
{'loss': '0.5564', 'grad_norm': '10.64', 'learning_rate': '1.444e-05', 'epoch': '0.2922'}

 29%|██▉       | 1461/5000 [19:16:44<39:38:56, 40.33s/it][2026-05-22 11:22:40,709] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 70])
[2026-05-22 11:22:44,224] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 105])

 29%|██▉       | 1462/5000 [19:17:27<40:24:29, 41.12s/it]
                                                         
{'loss': '0.3949', 'grad_norm': '11.44', 'learning_rate': '1.444e-05', 'epoch': '0.2924'}

 29%|██▉       | 1462/5000 [19:17:27<40:24:29, 41.12s/it][2026-05-22 11:23:13,932] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 11:23:14,619] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 232])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 29%|██▉       | 1463/5000 [19:18:00<37:59:38, 38.67s/it]
                                                         
{'loss': '0.638', 'grad_norm': '10.09', 'learning_rate': '1.444e-05', 'epoch': '0.2926'}

 29%|██▉       | 1463/5000 [19:18:00<37:59:38, 38.67s/it][2026-05-22 11:23:46,896] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 218])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 11:23:46,923] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 270])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 29%|██▉       | 1464/5000 [19:18:39<38:02:02, 38.72s/it]
                                                         
{'loss': '0.4191', 'grad_norm': '9.38', 'learning_rate': '1.443e-05', 'epoch': '0.2928'}

 29%|██▉       | 1464/5000 [19:18:39<38:02:02, 38.72s/it][2026-05-22 11:24:30,539] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 11:24:31,030] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 29%|██▉       | 1465/5000 [19:19:29<41:17:54, 42.06s/it]
                                                         
{'loss': '0.9614', 'grad_norm': '15.62', 'learning_rate': '1.443e-05', 'epoch': '0.293'}

 29%|██▉       | 1465/5000 [19:19:29<41:17:54, 42.06s/it][2026-05-22 11:25:20,355] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 11:25:20,795] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 29%|██▉       | 1466/5000 [19:20:20<43:49:50, 44.65s/it]
                                                         
{'loss': '0.5441', 'grad_norm': '10.39', 'learning_rate': '1.442e-05', 'epoch': '0.2932'}

 29%|██▉       | 1466/5000 [19:20:20<43:49:50, 44.65s/it][2026-05-22 11:26:16,920] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 70])
[2026-05-22 11:26:18,217] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])

 29%|██▉       | 1467/5000 [19:21:16<47:21:04, 48.25s/it]
                                                         
{'loss': '0.135', 'grad_norm': '5.301', 'learning_rate': '1.442e-05', 'epoch': '0.2934'}

 29%|██▉       | 1467/5000 [19:21:16<47:21:04, 48.25s/it][2026-05-22 11:27:10,630] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 196])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])
[2026-05-22 11:27:12,538] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 501])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 90])

 29%|██▉       | 1468/5000 [19:22:03<46:47:01, 47.68s/it]
                                                         
{'loss': '0.09788', 'grad_norm': '4.185', 'learning_rate': '1.442e-05', 'epoch': '0.2936'}

 29%|██▉       | 1468/5000 [19:22:03<46:47:01, 47.68s/it][2026-05-22 11:27:49,880] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 53])
[2026-05-22 11:27:50,639] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 53])

 29%|██▉       | 1469/5000 [19:22:30<40:47:09, 41.58s/it]
                                                         
{'loss': '0.7145', 'grad_norm': '19.73', 'learning_rate': '1.441e-05', 'epoch': '0.2938'}

 29%|██▉       | 1469/5000 [19:22:30<40:47:09, 41.58s/it][2026-05-22 11:28:17,976] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 428])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 97])
[2026-05-22 11:28:18,434] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 59])

 29%|██▉       | 1470/5000 [19:22:58<36:47:49, 37.53s/it]
                                                         
{'loss': '0.6568', 'grad_norm': '18.74', 'learning_rate': '1.441e-05', 'epoch': '0.294'}

 29%|██▉       | 1470/5000 [19:22:58<36:47:49, 37.53s/it][2026-05-22 11:28:44,913] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 224])
[2026-05-22 11:28:45,201] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 225])

 29%|██▉       | 1471/5000 [19:23:49<40:39:22, 41.47s/it]
                                                         
{'loss': '0.2569', 'grad_norm': '7.115', 'learning_rate': '1.44e-05', 'epoch': '0.2942'}

 29%|██▉       | 1471/5000 [19:23:49<40:39:22, 41.47s/it][2026-05-22 11:29:36,946] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])
[2026-05-22 11:29:37,137] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 236])

 29%|██▉       | 1472/5000 [19:24:40<43:29:18, 44.38s/it]
                                                         
{'loss': '0.3188', 'grad_norm': '8.25', 'learning_rate': '1.44e-05', 'epoch': '0.2944'}

 29%|██▉       | 1472/5000 [19:24:40<43:29:18, 44.38s/it][2026-05-22 11:30:27,928] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 378])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 209])
[2026-05-22 11:30:28,229] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 430])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])

 29%|██▉       | 1473/5000 [19:25:06<38:06:04, 38.89s/it]
                                                         
{'loss': '0.09269', 'grad_norm': '5.912', 'learning_rate': '1.44e-05', 'epoch': '0.2946'}

 29%|██▉       | 1473/5000 [19:25:06<38:06:04, 38.89s/it][2026-05-22 11:30:54,304] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])
[2026-05-22 11:30:54,361] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 329])

 29%|██▉       | 1474/5000 [19:25:34<35:00:14, 35.74s/it]
                                                         
{'loss': '0.07406', 'grad_norm': '5.802', 'learning_rate': '1.439e-05', 'epoch': '0.2948'}

 29%|██▉       | 1474/5000 [19:25:34<35:00:14, 35.74s/it][2026-05-22 11:31:24,318] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 11:31:27,055] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 30%|██▉       | 1475/5000 [19:26:03<32:51:42, 33.56s/it]
                                                         
{'loss': '0.4276', 'grad_norm': '15.43', 'learning_rate': '1.439e-05', 'epoch': '0.295'}

 30%|██▉       | 1475/5000 [19:26:03<32:51:42, 33.56s/it][2026-05-22 11:31:52,365] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 11:31:53,036] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 986])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 30%|██▉       | 1476/5000 [19:26:31<31:10:53, 31.85s/it]
                                                         
{'loss': '0.5486', 'grad_norm': '18.11', 'learning_rate': '1.438e-05', 'epoch': '0.2952'}

 30%|██▉       | 1476/5000 [19:26:31<31:10:53, 31.85s/it][2026-05-22 11:32:20,903] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 976])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 11:32:20,926] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 30%|██▉       | 1477/5000 [19:26:59<30:05:03, 30.74s/it]
                                                         
{'loss': '1.059', 'grad_norm': '16.35', 'learning_rate': '1.438e-05', 'epoch': '0.2954'}

 30%|██▉       | 1477/5000 [19:26:59<30:05:03, 30.74s/it][2026-05-22 11:32:48,730] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 11:32:50,053] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 30%|██▉       | 1478/5000 [19:27:49<35:42:29, 36.50s/it]
                                                         
{'loss': '0.8906', 'grad_norm': '15.65', 'learning_rate': '1.438e-05', 'epoch': '0.2956'}

 30%|██▉       | 1478/5000 [19:27:49<35:42:29, 36.50s/it][2026-05-22 11:33:37,630] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 11:33:37,669] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 177])

 30%|██▉       | 1479/5000 [19:28:19<33:45:14, 34.51s/it]
                                                         
{'loss': '2.011', 'grad_norm': '43.29', 'learning_rate': '1.437e-05', 'epoch': '0.2958'}

 30%|██▉       | 1479/5000 [19:28:19<33:45:14, 34.51s/it][2026-05-22 11:34:07,436] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 11:34:07,444] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 213])

 30%|██▉       | 1480/5000 [19:28:50<32:41:55, 33.44s/it]
                                                         
{'loss': '1.639', 'grad_norm': '26.09', 'learning_rate': '1.437e-05', 'epoch': '0.296'}

 30%|██▉       | 1480/5000 [19:28:50<32:41:55, 33.44s/it][2026-05-22 11:34:45,944] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 137])
[2026-05-22 11:34:47,384] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 176])

 30%|██▉       | 1481/5000 [19:29:41<37:52:35, 38.75s/it]
                                                         
{'loss': '0.4552', 'grad_norm': '11.4', 'learning_rate': '1.436e-05', 'epoch': '0.2962'}

 30%|██▉       | 1481/5000 [19:29:41<37:52:35, 38.75s/it][2026-05-22 11:35:40,623] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 148])
[2026-05-22 11:35:40,628] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 126])

 30%|██▉       | 1482/5000 [19:30:32<41:25:25, 42.39s/it]
                                                         
{'loss': '0.1963', 'grad_norm': '10.02', 'learning_rate': '1.436e-05', 'epoch': '0.2964'}

 30%|██▉       | 1482/5000 [19:30:32<41:25:25, 42.39s/it][2026-05-22 11:36:20,830] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 11:36:20,888] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 58])

 30%|██▉       | 1483/5000 [19:31:23<44:01:21, 45.06s/it]
                                                         
{'loss': '0.4988', 'grad_norm': '14.69', 'learning_rate': '1.436e-05', 'epoch': '0.2966'}

 30%|██▉       | 1483/5000 [19:31:23<44:01:21, 45.06s/it][2026-05-22 11:37:11,711] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 64])
[2026-05-22 11:37:13,239] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 74])

 30%|██▉       | 1484/5000 [19:32:14<45:54:30, 47.01s/it]
                                                         
{'loss': '0.2545', 'grad_norm': '11.98', 'learning_rate': '1.435e-05', 'epoch': '0.2968'}

 30%|██▉       | 1484/5000 [19:32:14<45:54:30, 47.01s/it][2026-05-22 11:38:04,170] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 227])
[2026-05-22 11:38:04,798] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 319])

 30%|██▉       | 1485/5000 [19:33:02<45:55:54, 47.04s/it]
                                                         
{'loss': '0.643', 'grad_norm': '17.71', 'learning_rate': '1.435e-05', 'epoch': '0.297'}

 30%|██▉       | 1485/5000 [19:33:02<45:55:54, 47.04s/it][2026-05-22 11:38:51,182] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 973])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])
[2026-05-22 11:38:51,780] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 281])

 30%|██▉       | 1486/5000 [19:33:49<45:59:43, 47.12s/it]
                                                         
{'loss': '0.5769', 'grad_norm': '17.63', 'learning_rate': '1.434e-05', 'epoch': '0.2972'}

 30%|██▉       | 1486/5000 [19:33:49<45:59:43, 47.12s/it][2026-05-22 11:39:38,078] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 11:39:38,349] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 171])

 30%|██▉       | 1487/5000 [19:34:35<45:44:07, 46.87s/it]
                                                         
{'loss': '0.1112', 'grad_norm': '6.056', 'learning_rate': '1.434e-05', 'epoch': '0.2974'}

 30%|██▉       | 1487/5000 [19:34:35<45:44:07, 46.87s/it][2026-05-22 11:40:24,499] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 138])
[2026-05-22 11:40:25,715] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])

 30%|██▉       | 1488/5000 [19:35:23<46:04:58, 47.24s/it]
                                                         
{'loss': '0.104', 'grad_norm': '5.793', 'learning_rate': '1.433e-05', 'epoch': '0.2976'}

 30%|██▉       | 1488/5000 [19:35:23<46:04:58, 47.24s/it][2026-05-22 11:41:23,443] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1028])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])
[2026-05-22 11:41:24,887] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 293])

 30%|██▉       | 1489/5000 [19:36:10<46:00:58, 47.18s/it]
                                                         
{'loss': '0.1093', 'grad_norm': '6.706', 'learning_rate': '1.433e-05', 'epoch': '0.2978'}

 30%|██▉       | 1489/5000 [19:36:10<46:00:58, 47.18s/it][2026-05-22 11:42:12,797] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 265])
[2026-05-22 11:42:13,766] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1021])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 247])

 30%|██▉       | 1490/5000 [19:36:58<46:06:51, 47.30s/it]
                                                         
{'loss': '0.09014', 'grad_norm': '7.115', 'learning_rate': '1.433e-05', 'epoch': '0.298'}

 30%|██▉       | 1490/5000 [19:36:58<46:06:51, 47.30s/it][2026-05-22 11:43:12,650] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 166])
[2026-05-22 11:43:13,744] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 90])

 30%|██▉       | 1491/5000 [19:37:46<46:21:29, 47.56s/it]
                                                         
{'loss': '0.47', 'grad_norm': '17.78', 'learning_rate': '1.432e-05', 'epoch': '0.2982'}

 30%|██▉       | 1491/5000 [19:37:46<46:21:29, 47.56s/it][2026-05-22 11:43:59,875] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 148])
[2026-05-22 11:44:01,038] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 237])

 30%|██▉       | 1492/5000 [19:38:34<46:28:44, 47.70s/it]
                                                         
{'loss': '0.09331', 'grad_norm': '5.938', 'learning_rate': '1.432e-05', 'epoch': '0.2984'}

 30%|██▉       | 1492/5000 [19:38:34<46:28:44, 47.70s/it][2026-05-22 11:44:25,689] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 186])
[2026-05-22 11:44:25,815] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 85])

 30%|██▉       | 1493/5000 [19:39:23<46:43:34, 47.97s/it]
                                                         
{'loss': '0.1426', 'grad_norm': '6.041', 'learning_rate': '1.431e-05', 'epoch': '0.2986'}

 30%|██▉       | 1493/5000 [19:39:23<46:43:34, 47.97s/it][2026-05-22 11:45:13,885] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 118])
[2026-05-22 11:45:14,079] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 197])

 30%|██▉       | 1494/5000 [19:40:12<47:13:21, 48.49s/it]
                                                         
{'loss': '0.115', 'grad_norm': '5.888', 'learning_rate': '1.431e-05', 'epoch': '0.2988'}

 30%|██▉       | 1494/5000 [19:40:12<47:13:21, 48.49s/it][2026-05-22 11:46:08,066] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 241])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 225])
[2026-05-22 11:46:08,701] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 30%|██▉       | 1495/5000 [19:41:05<48:25:13, 49.73s/it]
                                                         
{'loss': '0.3416', 'grad_norm': '10.21', 'learning_rate': '1.431e-05', 'epoch': '0.299'}

 30%|██▉       | 1495/5000 [19:41:05<48:25:13, 49.73s/it][2026-05-22 11:47:01,151] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 11:47:01,239] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])

 30%|██▉       | 1496/5000 [19:41:55<48:28:59, 49.81s/it]
                                                         
{'loss': '0.293', 'grad_norm': '7.785', 'learning_rate': '1.43e-05', 'epoch': '0.2992'}

 30%|██▉       | 1496/5000 [19:41:55<48:28:59, 49.81s/it][2026-05-22 11:47:42,775] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 293])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 11:47:43,719] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 290])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 30%|██▉       | 1497/5000 [19:42:42<47:38:39, 48.96s/it]
                                                         
{'loss': '1.165', 'grad_norm': '10.41', 'learning_rate': '1.43e-05', 'epoch': '0.2994'}

 30%|██▉       | 1497/5000 [19:42:42<47:38:39, 48.96s/it][2026-05-22 11:48:30,545] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 281])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 11:48:30,748] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 300])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 30%|██▉       | 1498/5000 [19:43:30<47:17:44, 48.62s/it]
                                                         
{'loss': '1.202', 'grad_norm': '11.92', 'learning_rate': '1.429e-05', 'epoch': '0.2996'}

 30%|██▉       | 1498/5000 [19:43:30<47:17:44, 48.62s/it][2026-05-22 11:49:27,102] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])
[2026-05-22 11:49:29,520] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 158])

 30%|██▉       | 1499/5000 [19:44:19<47:21:33, 48.70s/it]
                                                         
{'loss': '1.044', 'grad_norm': '8.157', 'learning_rate': '1.429e-05', 'epoch': '0.2998'}

 30%|██▉       | 1499/5000 [19:44:19<47:21:33, 48.70s/it][2026-05-22 11:50:19,925] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 187])
[2026-05-22 11:50:20,436] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 112])

 30%|███       | 1500/5000 [19:45:10<48:05:03, 49.46s/it]
                                                         
{'loss': '1.174', 'grad_norm': '7.35', 'learning_rate': '1.429e-05', 'epoch': '0.3'}

 30%|███       | 1500/5000 [19:45:10<48:05:03, 49.46s/it][2026-05-22 11:50:55,046] INFO [src.utils.basic_utils:19] Saving model to /workspace/VLM2Vec/outputs/Qwen3.5-08b-continued-mixed-alltasks/checkpoint-1500
[2026-05-22 11:50:55,703] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-22 11:50:55,712] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-22 11:51:08,588] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1021])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])
[2026-05-22 11:51:09,592] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 221])

 30%|███       | 1501/5000 [19:45:42<43:07:58, 44.38s/it]
                                                         
{'loss': '0.6164', 'grad_norm': '8.876', 'learning_rate': '1.428e-05', 'epoch': '0.3002'}

 30%|███       | 1501/5000 [19:45:42<43:07:58, 44.38s/it][2026-05-22 11:51:39,769] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])
[2026-05-22 11:51:42,891] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 241])

 30%|███       | 1502/5000 [19:46:17<40:09:11, 41.32s/it]
                                                         
{'loss': '0.03301', 'grad_norm': '4.372', 'learning_rate': '1.428e-05', 'epoch': '0.3004'}

 30%|███       | 1502/5000 [19:46:17<40:09:11, 41.32s/it][2026-05-22 11:52:03,977] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 220])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 257])
[2026-05-22 11:52:05,352] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 30%|███       | 1503/5000 [19:46:45<36:24:41, 37.48s/it]
                                                         
{'loss': '0.2453', 'grad_norm': '7.662', 'learning_rate': '1.427e-05', 'epoch': '0.3006'}

 30%|███       | 1503/5000 [19:46:45<36:24:41, 37.48s/it][2026-05-22 11:52:33,382] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 11:52:33,568] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 30%|███       | 1504/5000 [19:47:16<34:24:35, 35.43s/it]
                                                         
{'loss': '0.05331', 'grad_norm': '4.689', 'learning_rate': '1.427e-05', 'epoch': '0.3008'}

 30%|███       | 1504/5000 [19:47:16<34:24:35, 35.43s/it][2026-05-22 11:53:03,043] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 260])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 11:53:03,265] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 30%|███       | 1505/5000 [19:48:05<38:31:13, 39.68s/it]
                                                         
{'loss': '0.6022', 'grad_norm': '11.23', 'learning_rate': '1.427e-05', 'epoch': '0.301'}

 30%|███       | 1505/5000 [19:48:05<38:31:13, 39.68s/it][2026-05-22 11:53:54,335] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 11:53:54,368] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 972])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 30%|███       | 1506/5000 [19:48:55<41:17:44, 42.55s/it]
                                                         
{'loss': '0.4922', 'grad_norm': '9.333', 'learning_rate': '1.426e-05', 'epoch': '0.3012'}

 30%|███       | 1506/5000 [19:48:55<41:17:44, 42.55s/it][2026-05-22 11:54:43,960] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 11:54:44,513] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 986])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 30%|███       | 1507/5000 [19:49:43<43:00:38, 44.33s/it]
                                                         
{'loss': '0.5131', 'grad_norm': '8.831', 'learning_rate': '1.426e-05', 'epoch': '0.3014'}

 30%|███       | 1507/5000 [19:49:43<43:00:38, 44.33s/it][2026-05-22 11:55:32,101] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 970])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 11:55:32,665] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 30%|███       | 1508/5000 [19:50:31<44:04:42, 45.44s/it]
                                                         
{'loss': '0.09171', 'grad_norm': '5.562', 'learning_rate': '1.425e-05', 'epoch': '0.3016'}

 30%|███       | 1508/5000 [19:50:31<44:04:42, 45.44s/it][2026-05-22 11:56:20,970] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 11:56:21,000] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 30%|███       | 1509/5000 [19:51:04<40:17:49, 41.56s/it]
                                                         
{'loss': '0.4218', 'grad_norm': '10.72', 'learning_rate': '1.425e-05', 'epoch': '0.3018'}

 30%|███       | 1509/5000 [19:51:04<40:17:49, 41.56s/it][2026-05-22 11:56:53,269] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 11:56:53,846] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 976])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 30%|███       | 1510/5000 [19:51:38<38:14:47, 39.45s/it]
                                                         
{'loss': '0.02788', 'grad_norm': '3.186', 'learning_rate': '1.424e-05', 'epoch': '0.302'}

 30%|███       | 1510/5000 [19:51:38<38:14:47, 39.45s/it][2026-05-22 11:57:25,087] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 432])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 22])
[2026-05-22 11:57:25,126] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 14])

 30%|███       | 1511/5000 [19:52:06<34:51:12, 35.96s/it]
                                                         
{'loss': '0.8952', 'grad_norm': '11.78', 'learning_rate': '1.424e-05', 'epoch': '0.3022'}

 30%|███       | 1511/5000 [19:52:06<34:51:12, 35.96s/it][2026-05-22 11:57:52,893] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 20])
[2026-05-22 11:57:52,895] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 483])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 13])

 30%|███       | 1512/5000 [19:52:55<38:36:14, 39.84s/it]
                                                         
{'loss': '0.9808', 'grad_norm': '18.22', 'learning_rate': '1.424e-05', 'epoch': '0.3024'}

 30%|███       | 1512/5000 [19:52:55<38:36:14, 39.84s/it][2026-05-22 11:58:45,185] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 216])
[2026-05-22 11:58:45,224] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 86])

 30%|███       | 1513/5000 [19:53:46<41:55:54, 43.29s/it]
                                                         
{'loss': '0.15', 'grad_norm': '5.349', 'learning_rate': '1.423e-05', 'epoch': '0.3026'}

 30%|███       | 1513/5000 [19:53:46<41:55:54, 43.29s/it][2026-05-22 11:59:36,472] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])
[2026-05-22 11:59:36,591] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 144])

 30%|███       | 1514/5000 [19:54:36<43:52:29, 45.31s/it]
                                                         
{'loss': '0.1112', 'grad_norm': '9.293', 'learning_rate': '1.423e-05', 'epoch': '0.3028'}

 30%|███       | 1514/5000 [19:54:36<43:52:29, 45.31s/it][2026-05-22 12:00:31,045] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 244])
[2026-05-22 12:00:33,829] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])

 30%|███       | 1515/5000 [19:55:26<45:09:07, 46.64s/it]
                                                         
{'loss': '0.2327', 'grad_norm': '9.035', 'learning_rate': '1.422e-05', 'epoch': '0.303'}

 30%|███       | 1515/5000 [19:55:26<45:09:07, 46.64s/it][2026-05-22 12:01:23,969] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])
[2026-05-22 12:01:24,552] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 235])

 30%|███       | 1516/5000 [19:56:15<45:53:07, 47.41s/it]
                                                         
{'loss': '0.2701', 'grad_norm': '9.764', 'learning_rate': '1.422e-05', 'epoch': '0.3032'}

 30%|███       | 1516/5000 [19:56:15<45:53:07, 47.41s/it][2026-05-22 12:02:03,781] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 656])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 245])
[2026-05-22 12:02:04,039] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 630])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 30%|███       | 1517/5000 [19:56:58<44:38:26, 46.14s/it]
                                                         
{'loss': '1.267', 'grad_norm': '19.92', 'learning_rate': '1.422e-05', 'epoch': '0.3034'}

 30%|███       | 1517/5000 [19:56:58<44:38:26, 46.14s/it][2026-05-22 12:02:47,291] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 642])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 12:02:47,416] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 645])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 30%|███       | 1518/5000 [19:57:26<39:21:47, 40.70s/it]
                                                         
{'loss': '1.325', 'grad_norm': '20.65', 'learning_rate': '1.421e-05', 'epoch': '0.3036'}

 30%|███       | 1518/5000 [19:57:26<39:21:47, 40.70s/it][2026-05-22 12:03:13,286] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 269])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])
[2026-05-22 12:03:13,438] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 257])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 30%|███       | 1519/5000 [19:58:13<41:00:25, 42.41s/it]
                                                         
{'loss': '0.4057', 'grad_norm': '11.21', 'learning_rate': '1.421e-05', 'epoch': '0.3038'}

 30%|███       | 1519/5000 [19:58:13<41:00:25, 42.41s/it][2026-05-22 12:03:59,933] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 271])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 12:03:59,978] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 267])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 30%|███       | 1520/5000 [19:58:59<42:07:08, 43.57s/it]
                                                         
{'loss': '0.5359', 'grad_norm': '13.19', 'learning_rate': '1.42e-05', 'epoch': '0.304'}

 30%|███       | 1520/5000 [19:58:59<42:07:08, 43.57s/it][2026-05-22 12:04:46,922] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 258])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])
[2026-05-22 12:04:47,291] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 261])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 30%|███       | 1521/5000 [19:59:46<42:58:54, 44.48s/it]
                                                         
{'loss': '0.4369', 'grad_norm': '8.734', 'learning_rate': '1.42e-05', 'epoch': '0.3042'}

 30%|███       | 1521/5000 [19:59:46<42:58:54, 44.48s/it][2026-05-22 12:05:33,843] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 295])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 12:05:33,862] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 294])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 30%|███       | 1522/5000 [20:00:33<43:42:40, 45.24s/it]
                                                         
{'loss': '0.2313', 'grad_norm': '6.031', 'learning_rate': '1.42e-05', 'epoch': '0.3044'}

 30%|███       | 1522/5000 [20:00:33<43:42:40, 45.24s/it][2026-05-22 12:06:19,966] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 12:06:21,271] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 30%|███       | 1523/5000 [20:01:11<41:39:01, 43.12s/it]
                                                         
{'loss': '1.746', 'grad_norm': '11.88', 'learning_rate': '1.419e-05', 'epoch': '0.3046'}

 30%|███       | 1523/5000 [20:01:11<41:39:01, 43.12s/it][2026-05-22 12:06:58,978] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 309])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 12:06:59,102] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 301])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 30%|███       | 1524/5000 [20:01:50<40:22:31, 41.82s/it]
                                                         
{'loss': '1.587', 'grad_norm': '9.504', 'learning_rate': '1.419e-05', 'epoch': '0.3048'}

 30%|███       | 1524/5000 [20:01:50<40:22:31, 41.82s/it][2026-05-22 12:07:40,229] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 12:07:40,300] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 30%|███       | 1525/5000 [20:02:19<36:51:04, 38.18s/it]
                                                         
{'loss': '0.2931', 'grad_norm': '8.724', 'learning_rate': '1.418e-05', 'epoch': '0.305'}

 30%|███       | 1525/5000 [20:02:19<36:51:04, 38.18s/it][2026-05-22 12:08:10,312] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 12:08:11,050] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 31%|███       | 1526/5000 [20:02:50<34:48:37, 36.07s/it]
                                                         
{'loss': '0.2283', 'grad_norm': '7.351', 'learning_rate': '1.418e-05', 'epoch': '0.3052'}

 31%|███       | 1526/5000 [20:02:50<34:48:37, 36.07s/it][2026-05-22 12:08:51,214] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])
[2026-05-22 12:08:52,185] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])

 31%|███       | 1527/5000 [20:03:22<33:23:08, 34.61s/it]
                                                         
{'loss': '0.3274', 'grad_norm': '9.909', 'learning_rate': '1.418e-05', 'epoch': '0.3054'}

 31%|███       | 1527/5000 [20:03:22<33:23:08, 34.61s/it][2026-05-22 12:09:21,842] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 206])
[2026-05-22 12:09:23,183] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])

 31%|███       | 1528/5000 [20:03:54<32:38:53, 33.85s/it]
                                                         
{'loss': '0.1343', 'grad_norm': '7.072', 'learning_rate': '1.417e-05', 'epoch': '0.3056'}

 31%|███       | 1528/5000 [20:03:54<32:38:53, 33.85s/it][2026-05-22 12:09:41,241] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 219])
[2026-05-22 12:09:41,425] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 393])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])

 31%|███       | 1529/5000 [20:04:23<31:22:23, 32.54s/it]
                                                         
{'loss': '0.5211', 'grad_norm': '9.469', 'learning_rate': '1.417e-05', 'epoch': '0.3058'}

 31%|███       | 1529/5000 [20:04:23<31:22:23, 32.54s/it][2026-05-22 12:10:11,419] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 452])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])
[2026-05-22 12:10:11,425] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 569])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 236])

 31%|███       | 1530/5000 [20:04:53<30:36:23, 31.75s/it]
                                                         
{'loss': '0.02343', 'grad_norm': '3.269', 'learning_rate': '1.416e-05', 'epoch': '0.306'}

 31%|███       | 1530/5000 [20:04:53<30:36:23, 31.75s/it][2026-05-22 12:10:40,255] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 250])
[2026-05-22 12:10:41,465] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 272])

 31%|███       | 1531/5000 [20:05:44<36:06:24, 37.47s/it]
                                                         
{'loss': '0.2944', 'grad_norm': '6.663', 'learning_rate': '1.416e-05', 'epoch': '0.3062'}

 31%|███       | 1531/5000 [20:05:44<36:06:24, 37.47s/it][2026-05-22 12:11:31,738] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])
[2026-05-22 12:11:31,811] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])

 31%|███       | 1532/5000 [20:06:34<39:39:44, 41.17s/it]
                                                         
{'loss': '0.1095', 'grad_norm': '7.565', 'learning_rate': '1.416e-05', 'epoch': '0.3064'}

 31%|███       | 1532/5000 [20:06:34<39:39:44, 41.17s/it][2026-05-22 12:12:23,968] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 237])
[2026-05-22 12:12:24,322] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])

 31%|███       | 1533/5000 [20:07:24<42:18:06, 43.92s/it]
                                                         
{'loss': '0.3414', 'grad_norm': '6.531', 'learning_rate': '1.415e-05', 'epoch': '0.3066'}

 31%|███       | 1533/5000 [20:07:24<42:18:06, 43.92s/it][2026-05-22 12:13:15,063] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])
[2026-05-22 12:13:15,069] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1014])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 413])

 31%|███       | 1534/5000 [20:08:15<44:19:45, 46.04s/it]
                                                         
{'loss': '0.3484', 'grad_norm': '9.288', 'learning_rate': '1.415e-05', 'epoch': '0.3068'}

 31%|███       | 1534/5000 [20:08:15<44:19:45, 46.04s/it][2026-05-22 12:14:27,451] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 244])

 31%|███       | 1535/5000 [20:08:43<39:01:37, 40.55s/it]
                                                         
{'loss': '0.3619', 'grad_norm': '8.131', 'learning_rate': '1.414e-05', 'epoch': '0.307'}

 31%|███       | 1535/5000 [20:08:43<39:01:37, 40.55s/it][2026-05-22 12:14:30,478] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 12:14:56,031] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-22 12:14:58,330] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 219])

 31%|███       | 1536/5000 [20:09:20<37:58:22, 39.46s/it]
                                                         
{'loss': '0.1442', 'grad_norm': '4.168', 'learning_rate': '1.414e-05', 'epoch': '0.3072'}

 31%|███       | 1536/5000 [20:09:20<37:58:22, 39.46s/it][2026-05-22 12:15:06,162] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 141])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 12:15:06,867] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 157])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 279])

 31%|███       | 1537/5000 [20:09:49<34:55:45, 36.31s/it]
                                                         
{'loss': '0.3015', 'grad_norm': '8.718', 'learning_rate': '1.413e-05', 'epoch': '0.3074'}

 31%|███       | 1537/5000 [20:09:49<34:55:45, 36.31s/it][2026-05-22 12:15:35,215] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 165])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])
[2026-05-22 12:15:35,835] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 125])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 31%|███       | 1538/5000 [20:10:17<32:42:01, 34.00s/it]
                                                         
{'loss': '0.1004', 'grad_norm': '5.436', 'learning_rate': '1.413e-05', 'epoch': '0.3076'}

 31%|███       | 1538/5000 [20:10:17<32:42:01, 34.00s/it][2026-05-22 12:16:04,529] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 875])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 296])
[2026-05-22 12:16:05,763] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])

 31%|███       | 1539/5000 [20:11:05<36:29:19, 37.95s/it]
                                                         
{'loss': '0.2931', 'grad_norm': '6.746', 'learning_rate': '1.413e-05', 'epoch': '0.3078'}

 31%|███       | 1539/5000 [20:11:05<36:29:19, 37.95s/it][2026-05-22 12:16:52,300] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 958])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 256])
[2026-05-22 12:16:52,853] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 885])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 282])

 31%|███       | 1540/5000 [20:11:54<39:48:08, 41.41s/it]
                                                         
{'loss': '0.1655', 'grad_norm': '7.832', 'learning_rate': '1.412e-05', 'epoch': '0.308'}

 31%|███       | 1540/5000 [20:11:54<39:48:08, 41.41s/it][2026-05-22 12:17:41,630] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 802])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 11])
[2026-05-22 12:17:42,364] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 23])

 31%|███       | 1541/5000 [20:12:43<41:53:53, 43.61s/it]
                                                         
{'loss': '0.4562', 'grad_norm': '8.469', 'learning_rate': '1.412e-05', 'epoch': '0.3082'}

 31%|███       | 1541/5000 [20:12:43<41:53:53, 43.61s/it][2026-05-22 12:18:30,456] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 8])
[2026-05-22 12:18:31,028] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 967])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 7])

 31%|███       | 1542/5000 [20:13:32<43:23:05, 45.17s/it]
                                                         
{'loss': '0.3785', 'grad_norm': '6.782', 'learning_rate': '1.411e-05', 'epoch': '0.3084'}

 31%|███       | 1542/5000 [20:13:32<43:23:05, 45.17s/it][2026-05-22 12:19:18,119] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-22 12:19:18,707] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 31%|███       | 1543/5000 [20:13:55<37:09:19, 38.69s/it]
                                                         
{'loss': '0.5265', 'grad_norm': '16.19', 'learning_rate': '1.411e-05', 'epoch': '0.3086'}

 31%|███       | 1543/5000 [20:13:55<37:09:19, 38.69s/it][2026-05-22 12:19:41,736] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 404])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-22 12:19:41,799] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 31%|███       | 1544/5000 [20:14:18<32:29:16, 33.84s/it]
                                                         
{'loss': '0.5214', 'grad_norm': '15.23', 'learning_rate': '1.411e-05', 'epoch': '0.3088'}

 31%|███       | 1544/5000 [20:14:18<32:29:16, 33.84s/it][2026-05-22 12:20:04,766] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 352])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 12:20:05,091] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 267])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])

 31%|███       | 1545/5000 [20:15:03<35:46:31, 37.28s/it]
                                                         
{'loss': '0.5668', 'grad_norm': '14.43', 'learning_rate': '1.41e-05', 'epoch': '0.309'}

 31%|███       | 1545/5000 [20:15:03<35:46:31, 37.28s/it][2026-05-22 12:20:50,665] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 252])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 12:20:50,755] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 236])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 31%|███       | 1546/5000 [20:15:46<37:27:50, 39.05s/it]
                                                         
{'loss': '0.2119', 'grad_norm': '10.73', 'learning_rate': '1.41e-05', 'epoch': '0.3092'}

 31%|███       | 1546/5000 [20:15:46<37:27:50, 39.05s/it][2026-05-22 12:21:46,278] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 12:21:46,504] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 225])

 31%|███       | 1547/5000 [20:16:31<39:15:22, 40.93s/it]
                                                         
{'loss': '1.091', 'grad_norm': '21.09', 'learning_rate': '1.409e-05', 'epoch': '0.3094'}

 31%|███       | 1547/5000 [20:16:31<39:15:22, 40.93s/it][2026-05-22 12:22:29,613] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1021])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 12:22:31,773] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 31%|███       | 1548/5000 [20:17:16<40:20:57, 42.08s/it]
                                                         
{'loss': '0.7115', 'grad_norm': '16.9', 'learning_rate': '1.409e-05', 'epoch': '0.3096'}

 31%|███       | 1548/5000 [20:17:16<40:20:57, 42.08s/it][2026-05-22 12:23:03,230] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 243])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 58])
[2026-05-22 12:23:03,849] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 259])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 59])

 31%|███       | 1549/5000 [20:17:42<35:45:51, 37.31s/it]
                                                         
{'loss': '0.8616', 'grad_norm': '22.1', 'learning_rate': '1.409e-05', 'epoch': '0.3098'}

 31%|███       | 1549/5000 [20:17:42<35:45:51, 37.31s/it][2026-05-22 12:23:30,334] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 795])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 62])
[2026-05-22 12:23:31,771] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 79])

 31%|███       | 1550/5000 [20:18:10<33:01:57, 34.47s/it]
                                                         
{'loss': '0.6943', 'grad_norm': '20.37', 'learning_rate': '1.408e-05', 'epoch': '0.31'}

 31%|███       | 1550/5000 [20:18:10<33:01:57, 34.47s/it][2026-05-22 12:24:11,624] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 182])
[2026-05-22 12:24:12,343] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 114])

 31%|███       | 1551/5000 [20:18:41<31:52:22, 33.27s/it]
                                                         
{'loss': '0.3998', 'grad_norm': '13.56', 'learning_rate': '1.408e-05', 'epoch': '0.3102'}

 31%|███       | 1551/5000 [20:18:41<31:52:22, 33.27s/it][2026-05-22 12:24:40,377] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 144])
[2026-05-22 12:24:42,390] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 222])

 31%|███       | 1552/5000 [20:19:12<31:17:17, 32.67s/it]
                                                         
{'loss': '0.3197', 'grad_norm': '14.76', 'learning_rate': '1.407e-05', 'epoch': '0.3104'}

 31%|███       | 1552/5000 [20:19:12<31:17:17, 32.67s/it][2026-05-22 12:25:12,505] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 12:25:12,830] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 31%|███       | 1553/5000 [20:20:04<36:53:49, 38.53s/it]
                                                         
{'loss': '0.3405', 'grad_norm': '8.045', 'learning_rate': '1.407e-05', 'epoch': '0.3106'}

 31%|███       | 1553/5000 [20:20:04<36:53:49, 38.53s/it][2026-05-22 12:26:02,774] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 12:26:03,200] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 31%|███       | 1554/5000 [20:20:59<41:39:10, 43.51s/it]
                                                         
{'loss': '0.1928', 'grad_norm': '7.164', 'learning_rate': '1.407e-05', 'epoch': '0.3108'}

 31%|███       | 1554/5000 [20:20:59<41:39:10, 43.51s/it][2026-05-22 12:26:46,756] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 12:26:47,456] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 317])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 31%|███       | 1555/5000 [20:21:23<35:58:27, 37.59s/it]
                                                         
{'loss': '0.5307', 'grad_norm': '10.57', 'learning_rate': '1.406e-05', 'epoch': '0.311'}

 31%|███       | 1555/5000 [20:21:23<35:58:27, 37.59s/it][2026-05-22 12:27:11,028] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 12:27:11,642] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 481])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 31%|███       | 1556/5000 [20:22:09<38:13:33, 39.96s/it]
                                                         
{'loss': '0.6832', 'grad_norm': '17.36', 'learning_rate': '1.406e-05', 'epoch': '0.3112'}

 31%|███       | 1556/5000 [20:22:09<38:13:33, 39.96s/it][2026-05-22 12:27:55,912] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 277])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 12:27:56,029] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 240])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 31%|███       | 1557/5000 [20:22:56<40:23:10, 42.23s/it]
                                                         
{'loss': '0.5058', 'grad_norm': '8.355', 'learning_rate': '1.405e-05', 'epoch': '0.3114'}

 31%|███       | 1557/5000 [20:22:56<40:23:10, 42.23s/it][2026-05-22 12:28:43,421] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 284])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 12:28:43,834] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 293])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 31%|███       | 1558/5000 [20:23:44<42:04:59, 44.01s/it]
                                                         
{'loss': '0.5609', 'grad_norm': '12.6', 'learning_rate': '1.405e-05', 'epoch': '0.3116'}

 31%|███       | 1558/5000 [20:23:44<42:04:59, 44.01s/it][2026-05-22 12:29:32,045] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-22 12:29:32,112] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 31%|███       | 1559/5000 [20:24:34<43:37:59, 45.65s/it]
                                                         
{'loss': '0.3748', 'grad_norm': '7.766', 'learning_rate': '1.404e-05', 'epoch': '0.3118'}

 31%|███       | 1559/5000 [20:24:34<43:37:59, 45.65s/it][2026-05-22 12:30:21,015] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 203])
[2026-05-22 12:30:21,810] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 223])

 31%|███       | 1560/5000 [20:25:25<45:20:56, 47.46s/it]
                                                         
{'loss': '0.2906', 'grad_norm': '8.845', 'learning_rate': '1.404e-05', 'epoch': '0.312'}

 31%|███       | 1560/5000 [20:25:25<45:20:56, 47.46s/it][2026-05-22 12:31:16,494] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 280])
[2026-05-22 12:31:16,581] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 256])

 31%|███       | 1561/5000 [20:25:58<41:01:22, 42.94s/it]
                                                         
{'loss': '0.2326', 'grad_norm': '7.779', 'learning_rate': '1.404e-05', 'epoch': '0.3122'}

 31%|███       | 1561/5000 [20:25:58<41:01:22, 42.94s/it][2026-05-22 12:31:48,994] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 320])
[2026-05-22 12:31:49,284] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 249])

 31%|███       | 1562/5000 [20:26:31<38:16:58, 40.09s/it]
                                                         
{'loss': '0.02733', 'grad_norm': '4.248', 'learning_rate': '1.403e-05', 'epoch': '0.3124'}

 31%|███       | 1562/5000 [20:26:31<38:16:58, 40.09s/it][2026-05-22 12:32:19,415] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 618])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 247])
[2026-05-22 12:32:19,925] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 632])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 31%|███▏      | 1563/5000 [20:26:59<34:46:47, 36.43s/it]
                                                         
{'loss': '0.3992', 'grad_norm': '14.03', 'learning_rate': '1.403e-05', 'epoch': '0.3126'}

 31%|███▏      | 1563/5000 [20:26:59<34:46:47, 36.43s/it][2026-05-22 12:32:47,517] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 628])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 12:32:47,916] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 616])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 31%|███▏      | 1564/5000 [20:27:29<32:56:00, 34.51s/it]
                                                         
{'loss': '0.2249', 'grad_norm': '10.98', 'learning_rate': '1.402e-05', 'epoch': '0.3128'}

 31%|███▏      | 1564/5000 [20:27:29<32:56:00, 34.51s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 12:33:16,692] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 647])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])
[2026-05-22 12:33:16,741] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 651])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 223])

 31%|███▏      | 1565/5000 [20:28:14<35:45:28, 37.48s/it]
                                                         
{'loss': '0.4602', 'grad_norm': '14.11', 'learning_rate': '1.402e-05', 'epoch': '0.313'}

 31%|███▏      | 1565/5000 [20:28:14<35:45:28, 37.48s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 12:34:01,018] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 644])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 228])
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 12:34:01,636] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 681])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 207])

 31%|███▏      | 1566/5000 [20:28:59<37:55:05, 39.75s/it]
                                                         
{'loss': '0.4644', 'grad_norm': '13.64', 'learning_rate': '1.402e-05', 'epoch': '0.3132'}

 31%|███▏      | 1566/5000 [20:28:59<37:55:05, 39.75s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 12:34:46,396] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 309])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 240])
[2026-05-22 12:34:46,912] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 31%|███▏      | 1567/5000 [20:29:46<40:05:35, 42.04s/it]
                                                         
{'loss': '0.3749', 'grad_norm': '7.998', 'learning_rate': '1.401e-05', 'epoch': '0.3134'}

 31%|███▏      | 1567/5000 [20:29:46<40:05:35, 42.04s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 12:35:33,860] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 12:35:34,629] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 31%|███▏      | 1568/5000 [20:30:34<41:39:59, 43.71s/it]
                                                         
{'loss': '0.1616', 'grad_norm': '6.685', 'learning_rate': '1.401e-05', 'epoch': '0.3136'}

 31%|███▏      | 1568/5000 [20:30:34<41:39:59, 43.71s/it][2026-05-22 12:36:21,116] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 398])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 12:36:21,447] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 31%|███▏      | 1569/5000 [20:31:10<39:35:55, 41.55s/it]
                                                         
{'loss': '0.6223', 'grad_norm': '9.073', 'learning_rate': '1.4e-05', 'epoch': '0.3138'}

 31%|███▏      | 1569/5000 [20:31:10<39:35:55, 41.55s/it][2026-05-22 12:36:58,076] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 12:36:58,561] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 31%|███▏      | 1570/5000 [20:31:47<38:21:07, 40.25s/it]
                                                         
{'loss': '0.1773', 'grad_norm': '4.632', 'learning_rate': '1.4e-05', 'epoch': '0.314'}

 31%|███▏      | 1570/5000 [20:31:47<38:21:07, 40.25s/it][2026-05-22 12:37:36,105] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 21])
[2026-05-22 12:37:36,276] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 23])

 31%|███▏      | 1571/5000 [20:32:23<36:54:20, 38.75s/it]
                                                         
{'loss': '1.619', 'grad_norm': '11.91', 'learning_rate': '1.4e-05', 'epoch': '0.3142'}

 31%|███▏      | 1571/5000 [20:32:23<36:54:20, 38.75s/it][2026-05-22 12:38:10,497] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])
[2026-05-22 12:38:11,523] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 31%|███▏      | 1572/5000 [20:32:58<35:51:39, 37.66s/it]
                                                         
{'loss': '1.697', 'grad_norm': '12.64', 'learning_rate': '1.399e-05', 'epoch': '0.3144'}

 31%|███▏      | 1572/5000 [20:32:58<35:51:39, 37.66s/it][2026-05-22 12:38:44,329] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 540])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 12:38:44,387] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 483])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])

 31%|███▏      | 1573/5000 [20:33:27<33:29:44, 35.19s/it]
                                                         
{'loss': '0.6602', 'grad_norm': '8.218', 'learning_rate': '1.399e-05', 'epoch': '0.3146'}

 31%|███▏      | 1573/5000 [20:33:27<33:29:44, 35.19s/it][2026-05-22 12:39:13,830] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 12:39:14,357] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 568])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 215])

 31%|███▏      | 1574/5000 [20:33:57<31:53:44, 33.52s/it]
                                                         
{'loss': '0.02691', 'grad_norm': '3.124', 'learning_rate': '1.398e-05', 'epoch': '0.3148'}

 31%|███▏      | 1574/5000 [20:33:57<31:53:44, 33.52s/it][2026-05-22 12:39:44,490] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 251])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 12:39:45,064] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 438])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 565])

 32%|███▏      | 1575/5000 [20:34:29<31:31:12, 33.13s/it]
                                                         
{'loss': '1.076', 'grad_norm': '18.52', 'learning_rate': '1.398e-05', 'epoch': '0.315'}

 32%|███▏      | 1575/5000 [20:34:29<31:31:12, 33.13s/it][2026-05-22 12:40:16,618] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 310])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 12:40:17,342] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 319])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 32%|███▏      | 1576/5000 [20:35:01<31:14:54, 32.85s/it]
                                                         
{'loss': '0.7374', 'grad_norm': '12.4', 'learning_rate': '1.398e-05', 'epoch': '0.3152'}

 32%|███▏      | 1576/5000 [20:35:01<31:14:54, 32.85s/it][2026-05-22 12:40:48,244] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 234])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 12:40:48,409] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 243])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 32%|███▏      | 1577/5000 [20:35:45<34:25:36, 36.21s/it]
                                                         
{'loss': '3.108', 'grad_norm': '18.02', 'learning_rate': '1.397e-05', 'epoch': '0.3154'}

 32%|███▏      | 1577/5000 [20:35:45<34:25:36, 36.21s/it][2026-05-22 12:41:32,595] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 290])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 12:41:33,169] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 253])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 32%|███▏      | 1578/5000 [20:36:32<37:19:50, 39.27s/it]
                                                         
{'loss': '3.048', 'grad_norm': '18.02', 'learning_rate': '1.397e-05', 'epoch': '0.3156'}

 32%|███▏      | 1578/5000 [20:36:32<37:19:50, 39.27s/it][2026-05-22 12:42:19,563] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 960])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 12:42:19,642] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 185])

 32%|███▏      | 1579/5000 [20:37:00<34:21:23, 36.15s/it]
                                                         
{'loss': '0.5132', 'grad_norm': '13.25', 'learning_rate': '1.396e-05', 'epoch': '0.3158'}

 32%|███▏      | 1579/5000 [20:37:00<34:21:23, 36.15s/it][2026-05-22 12:42:48,702] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 12:42:49,155] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 32%|███▏      | 1580/5000 [20:37:44<36:34:31, 38.50s/it]
                                                         
{'loss': '0.3986', 'grad_norm': '10.8', 'learning_rate': '1.396e-05', 'epoch': '0.316'}

 32%|███▏      | 1580/5000 [20:37:44<36:34:31, 38.50s/it][2026-05-22 12:43:33,073] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-22 12:43:33,537] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 32%|███▏      | 1581/5000 [20:38:18<35:15:08, 37.12s/it]
                                                         
{'loss': '0.2755', 'grad_norm': '7.87', 'learning_rate': '1.396e-05', 'epoch': '0.3162'}

 32%|███▏      | 1581/5000 [20:38:18<35:15:08, 37.12s/it][2026-05-22 12:44:06,649] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 12:44:06,974] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 32%|███▏      | 1582/5000 [20:38:47<32:51:11, 34.60s/it]
                                                         
{'loss': '0.05438', 'grad_norm': '4.617', 'learning_rate': '1.395e-05', 'epoch': '0.3164'}

 32%|███▏      | 1582/5000 [20:38:47<32:51:11, 34.60s/it][2026-05-22 12:44:34,082] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 133])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 12:44:34,108] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 157])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])

 32%|███▏      | 1583/5000 [20:39:14<30:36:23, 32.25s/it]
                                                         
{'loss': '0.4317', 'grad_norm': '10.91', 'learning_rate': '1.395e-05', 'epoch': '0.3166'}

 32%|███▏      | 1583/5000 [20:39:14<30:36:23, 32.25s/it][2026-05-22 12:45:00,332] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 149])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-22 12:45:00,877] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 141])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 32%|███▏      | 1584/5000 [20:39:43<29:40:45, 31.28s/it]
                                                         
{'loss': '0.2154', 'grad_norm': '7.574', 'learning_rate': '1.394e-05', 'epoch': '0.3168'}

 32%|███▏      | 1584/5000 [20:39:43<29:40:45, 31.28s/it][2026-05-22 12:45:32,542] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 12:45:32,544] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 32%|███▏      | 1585/5000 [20:40:31<34:31:54, 36.40s/it]
                                                         
{'loss': '0.5451', 'grad_norm': '7.422', 'learning_rate': '1.394e-05', 'epoch': '0.317'}

 32%|███▏      | 1585/5000 [20:40:31<34:31:54, 36.40s/it][2026-05-22 12:46:20,124] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])
[2026-05-22 12:46:22,057] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 183])

 32%|███▏      | 1586/5000 [20:41:21<38:17:22, 40.38s/it]
                                                         
{'loss': '0.4027', 'grad_norm': '7.519', 'learning_rate': '1.393e-05', 'epoch': '0.3172'}

 32%|███▏      | 1586/5000 [20:41:21<38:17:22, 40.38s/it][2026-05-22 12:47:09,438] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 12:47:09,513] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 129])

 32%|███▏      | 1587/5000 [20:42:06<39:43:34, 41.90s/it]
                                                         
{'loss': '2.189', 'grad_norm': '9.598', 'learning_rate': '1.393e-05', 'epoch': '0.3174'}

 32%|███▏      | 1587/5000 [20:42:06<39:43:34, 41.90s/it][2026-05-22 12:47:54,326] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 978])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 12:47:54,606] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 32%|███▏      | 1588/5000 [20:42:51<40:38:05, 42.87s/it]
                                                         
{'loss': '2.324', 'grad_norm': '20.61', 'learning_rate': '1.393e-05', 'epoch': '0.3176'}

 32%|███▏      | 1588/5000 [20:42:51<40:38:05, 42.87s/it][2026-05-22 12:48:38,096] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 12:48:38,138] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

 32%|███▏      | 1589/5000 [20:43:14<34:54:42, 36.85s/it]
                                                         
{'loss': '1.05', 'grad_norm': '25.24', 'learning_rate': '1.392e-05', 'epoch': '0.3178'}

 32%|███▏      | 1589/5000 [20:43:14<34:54:42, 36.85s/it][2026-05-22 12:49:01,610] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 865])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 12:49:02,009] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 981])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 32%|███▏      | 1590/5000 [20:43:38<31:03:25, 32.79s/it]
                                                         
{'loss': '0.9018', 'grad_norm': '23.07', 'learning_rate': '1.392e-05', 'epoch': '0.318'}

 32%|███▏      | 1590/5000 [20:43:38<31:03:25, 32.79s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 12:49:25,270] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 271])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 12:49:25,713] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 329])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 32%|███▏      | 1591/5000 [20:44:24<34:47:34, 36.74s/it]
                                                         
{'loss': '0.3361', 'grad_norm': '17.18', 'learning_rate': '1.391e-05', 'epoch': '0.3182'}

 32%|███▏      | 1591/5000 [20:44:24<34:47:34, 36.74s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 12:50:11,624] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 12:50:12,147] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 788])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 32%|███▏      | 1592/5000 [20:45:10<37:31:47, 39.64s/it]
                                                         
{'loss': '0.3129', 'grad_norm': '10.28', 'learning_rate': '1.391e-05', 'epoch': '0.3184'}

 32%|███▏      | 1592/5000 [20:45:10<37:31:47, 39.64s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 12:50:56,401] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 149])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 228])
[2026-05-22 12:50:56,417] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 245])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 179])

 32%|███▏      | 1593/5000 [20:45:54<38:53:19, 41.09s/it]
                                                         
{'loss': '1.148', 'grad_norm': '24.57', 'learning_rate': '1.391e-05', 'epoch': '0.3186'}

 32%|███▏      | 1593/5000 [20:45:54<38:53:19, 41.09s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 12:51:40,825] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 229])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 177])
[2026-05-22 12:51:41,504] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 141])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 189])

 32%|███▏      | 1594/5000 [20:46:39<39:48:34, 42.08s/it]
                                                         
{'loss': '0.9372', 'grad_norm': '21.7', 'learning_rate': '1.39e-05', 'epoch': '0.3188'}

 32%|███▏      | 1594/5000 [20:46:39<39:48:34, 42.08s/it][2026-05-22 12:52:26,575] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 235])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 12:52:27,069] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 272])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 32%|███▏      | 1595/5000 [20:47:03<34:47:41, 36.79s/it]
                                                         
{'loss': '1.105', 'grad_norm': '22.34', 'learning_rate': '1.39e-05', 'epoch': '0.319'}

 32%|███▏      | 1595/5000 [20:47:03<34:47:41, 36.79s/it][2026-05-22 12:52:51,080] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 428])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 12:52:51,442] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 331])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 32%|███▏      | 1596/5000 [20:47:48<36:59:04, 39.11s/it]
                                                         
{'loss': '1.111', 'grad_norm': '22.75', 'learning_rate': '1.389e-05', 'epoch': '0.3192'}

 32%|███▏      | 1596/5000 [20:47:48<36:59:04, 39.11s/it][2026-05-22 12:53:37,875] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 69])
[2026-05-22 12:53:38,215] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 53])

 32%|███▏      | 1597/5000 [20:48:18<34:21:41, 36.35s/it]
                                                         
{'loss': '0.4692', 'grad_norm': '8.825', 'learning_rate': '1.389e-05', 'epoch': '0.3194'}

 32%|███▏      | 1597/5000 [20:48:18<34:21:41, 36.35s/it][2026-05-22 12:54:07,807] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 45])
[2026-05-22 12:54:09,502] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 54])

 32%|███▏      | 1598/5000 [20:49:02<36:33:20, 38.68s/it]
                                                         
{'loss': '0.2331', 'grad_norm': '8.178', 'learning_rate': '1.389e-05', 'epoch': '0.3196'}

 32%|███▏      | 1598/5000 [20:49:02<36:33:20, 38.68s/it][2026-05-22 12:54:54,091] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 19])
[2026-05-22 12:54:56,971] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 15])

 32%|███▏      | 1599/5000 [20:49:30<33:42:06, 35.67s/it]
                                                         
{'loss': '0.5618', 'grad_norm': '15.69', 'learning_rate': '1.388e-05', 'epoch': '0.3198'}

 32%|███▏      | 1599/5000 [20:49:30<33:42:06, 35.67s/it][2026-05-22 12:55:22,762] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 18])
[2026-05-22 12:55:25,713] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 17])

 32%|███▏      | 1600/5000 [20:49:59<31:39:25, 33.52s/it]
                                                         
{'loss': '0.4223', 'grad_norm': '15.51', 'learning_rate': '1.388e-05', 'epoch': '0.32'}

 32%|███▏      | 1600/5000 [20:49:59<31:39:25, 33.52s/it][2026-05-22 12:55:49,728] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1012])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 17])
[2026-05-22 12:55:51,350] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 14])

 32%|███▏      | 1601/5000 [20:50:30<30:58:52, 32.81s/it]
                                                         
{'loss': '0.4555', 'grad_norm': '11.19', 'learning_rate': '1.387e-05', 'epoch': '0.3202'}

 32%|███▏      | 1601/5000 [20:50:30<30:58:52, 32.81s/it][2026-05-22 12:56:21,374] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 18])
[2026-05-22 12:56:21,761] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 11])

 32%|███▏      | 1602/5000 [20:51:05<31:29:48, 33.37s/it]
                                                         
{'loss': '0.1116', 'grad_norm': '8.77', 'learning_rate': '1.387e-05', 'epoch': '0.3204'}

 32%|███▏      | 1602/5000 [20:51:05<31:29:48, 33.37s/it][2026-05-22 12:56:52,633] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 807])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 210])
[2026-05-22 12:56:53,447] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 804])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 32%|███▏      | 1603/5000 [20:51:51<34:59:01, 37.07s/it]
                                                         
{'loss': '0.1128', 'grad_norm': '6.484', 'learning_rate': '1.387e-05', 'epoch': '0.3206'}

 32%|███▏      | 1603/5000 [20:51:51<34:59:01, 37.07s/it][2026-05-22 12:57:39,208] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 930])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 12:57:39,384] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 927])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 32%|███▏      | 1604/5000 [20:52:36<37:26:27, 39.69s/it]
                                                         
{'loss': '0.1033', 'grad_norm': '9.66', 'learning_rate': '1.386e-05', 'epoch': '0.3208'}

 32%|███▏      | 1604/5000 [20:52:36<37:26:27, 39.69s/it][2026-05-22 12:58:23,713] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 267])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 228])
[2026-05-22 12:58:24,345] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 32%|███▏      | 1605/5000 [20:53:24<39:41:12, 42.08s/it]
                                                         
{'loss': '0.8944', 'grad_norm': '23.19', 'learning_rate': '1.386e-05', 'epoch': '0.321'}

 32%|███▏      | 1605/5000 [20:53:24<39:41:12, 42.08s/it][2026-05-22 12:59:12,028] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 12:59:12,205] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 32%|███▏      | 1606/5000 [20:54:12<41:14:43, 43.75s/it]
                                                         
{'loss': '0.9534', 'grad_norm': '24.46', 'learning_rate': '1.385e-05', 'epoch': '0.3212'}

 32%|███▏      | 1606/5000 [20:54:12<41:14:43, 43.75s/it][2026-05-22 12:59:59,482] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 564])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 12:59:59,767] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 32%|███▏      | 1607/5000 [20:54:58<41:55:06, 44.48s/it]
                                                         
{'loss': '1.306', 'grad_norm': '19.7', 'learning_rate': '1.385e-05', 'epoch': '0.3214'}

 32%|███▏      | 1607/5000 [20:54:58<41:55:06, 44.48s/it][2026-05-22 13:00:45,779] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:00:46,971] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 543])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 32%|███▏      | 1608/5000 [20:55:44<42:29:12, 45.09s/it]
                                                         
{'loss': '0.9739', 'grad_norm': '20.25', 'learning_rate': '1.384e-05', 'epoch': '0.3216'}

 32%|███▏      | 1608/5000 [20:55:44<42:29:12, 45.09s/it][2026-05-22 13:01:31,646] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 64])
[2026-05-22 13:01:31,697] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 62])

 32%|███▏      | 1609/5000 [20:56:25<41:15:08, 43.79s/it]
                                                         
{'loss': '0.5933', 'grad_norm': '10.18', 'learning_rate': '1.384e-05', 'epoch': '0.3218'}

 32%|███▏      | 1609/5000 [20:56:25<41:15:08, 43.79s/it][2026-05-22 13:02:12,413] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 55])
[2026-05-22 13:02:13,166] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])

 32%|███▏      | 1610/5000 [20:57:12<42:13:41, 44.84s/it]
                                                         
{'loss': '0.08708', 'grad_norm': '5.318', 'learning_rate': '1.384e-05', 'epoch': '0.322'}

 32%|███▏      | 1610/5000 [20:57:12<42:13:41, 44.84s/it][2026-05-22 13:03:00,289] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 222])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 230])
[2026-05-22 13:03:00,644] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])

 32%|███▏      | 1611/5000 [20:57:45<38:49:54, 41.25s/it]
                                                         
{'loss': '0.4838', 'grad_norm': '8.849', 'learning_rate': '1.383e-05', 'epoch': '0.3222'}

 32%|███▏      | 1611/5000 [20:57:45<38:49:54, 41.25s/it][2026-05-22 13:03:33,766] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 431])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])
[2026-05-22 13:03:34,032] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 471])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 260])

 32%|███▏      | 1612/5000 [20:58:17<36:10:17, 38.43s/it]
                                                         
{'loss': '0.0713', 'grad_norm': '5.379', 'learning_rate': '1.383e-05', 'epoch': '0.3224'}

 32%|███▏      | 1612/5000 [20:58:17<36:10:17, 38.43s/it][2026-05-22 13:04:04,139] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])
[2026-05-22 13:04:04,737] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 32%|███▏      | 1613/5000 [20:58:54<35:46:30, 38.02s/it]
                                                         
{'loss': '0.568', 'grad_norm': '6.281', 'learning_rate': '1.382e-05', 'epoch': '0.3226'}

 32%|███▏      | 1613/5000 [20:58:54<35:46:30, 38.02s/it][2026-05-22 13:04:41,907] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 264])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 13:04:42,088] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 479])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

 32%|███▏      | 1614/5000 [20:59:30<35:06:24, 37.33s/it]
                                                         
{'loss': '0.4893', 'grad_norm': '5.523', 'learning_rate': '1.382e-05', 'epoch': '0.3228'}

 32%|███▏      | 1614/5000 [20:59:30<35:06:24, 37.33s/it][2026-05-22 13:05:17,142] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 231])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 61])
[2026-05-22 13:05:17,662] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 317])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 56])

 32%|███▏      | 1615/5000 [21:00:02<33:30:32, 35.64s/it]
                                                         
{'loss': '2.86', 'grad_norm': '6.822', 'learning_rate': '1.382e-05', 'epoch': '0.323'}

 32%|███▏      | 1615/5000 [21:00:02<33:30:32, 35.64s/it][2026-05-22 13:05:48,884] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 696])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 85])
[2026-05-22 13:05:50,861] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 695])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 105])

 32%|███▏      | 1616/5000 [21:00:35<32:58:26, 35.08s/it]
                                                         
{'loss': '2.91', 'grad_norm': '6.603', 'learning_rate': '1.381e-05', 'epoch': '0.3232'}

 32%|███▏      | 1616/5000 [21:00:35<32:58:26, 35.08s/it][2026-05-22 13:06:22,010] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 105])
[2026-05-22 13:06:22,664] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 278])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 111])

 32%|███▏      | 1617/5000 [21:01:04<31:08:35, 33.14s/it]
                                                         
{'loss': '0.2612', 'grad_norm': '7.361', 'learning_rate': '1.381e-05', 'epoch': '0.3234'}

 32%|███▏      | 1617/5000 [21:01:04<31:08:35, 33.14s/it][2026-05-22 13:06:51,705] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 51])
[2026-05-22 13:06:52,208] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 854])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 169])

 32%|███▏      | 1618/5000 [21:01:34<30:15:03, 32.20s/it]
                                                         
{'loss': '0.05258', 'grad_norm': '3.837', 'learning_rate': '1.38e-05', 'epoch': '0.3236'}

 32%|███▏      | 1618/5000 [21:01:34<30:15:03, 32.20s/it][2026-05-22 13:07:25,795] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:07:26,054] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 32%|███▏      | 1619/5000 [21:02:01<28:45:19, 30.62s/it]
                                                         
{'loss': '0.2291', 'grad_norm': '8.409', 'learning_rate': '1.38e-05', 'epoch': '0.3238'}

 32%|███▏      | 1619/5000 [21:02:01<28:45:19, 30.62s/it][2026-05-22 13:07:52,357] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1007])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:07:53,166] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 32%|███▏      | 1620/5000 [21:02:31<28:41:18, 30.56s/it]
                                                         
{'loss': '0.2039', 'grad_norm': '7.356', 'learning_rate': '1.38e-05', 'epoch': '0.324'}

 32%|███▏      | 1620/5000 [21:02:31<28:41:18, 30.56s/it][2026-05-22 13:08:19,194] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 244])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 13:08:19,681] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 32%|███▏      | 1621/5000 [21:02:55<26:41:35, 28.44s/it]
                                                         
{'loss': '0.1495', 'grad_norm': '6.749', 'learning_rate': '1.379e-05', 'epoch': '0.3242'}

 32%|███▏      | 1621/5000 [21:02:55<26:41:35, 28.44s/it][2026-05-22 13:08:43,071] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 245])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:08:43,198] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 497])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 32%|███▏      | 1622/5000 [21:03:29<28:14:48, 30.10s/it]
                                                         
{'loss': '0.1936', 'grad_norm': '8.364', 'learning_rate': '1.379e-05', 'epoch': '0.3244'}

 32%|███▏      | 1622/5000 [21:03:29<28:14:48, 30.10s/it][2026-05-22 13:09:16,914] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])
[2026-05-22 13:09:16,990] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 976])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 49])

 32%|███▏      | 1623/5000 [21:03:53<26:42:15, 28.47s/it]
                                                         
{'loss': '1.306', 'grad_norm': '21.29', 'learning_rate': '1.378e-05', 'epoch': '0.3246'}

 32%|███▏      | 1623/5000 [21:03:53<26:42:15, 28.47s/it][2026-05-22 13:09:41,443] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 91])
[2026-05-22 13:09:43,395] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 977])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 74])

 32%|███▏      | 1624/5000 [21:04:39<31:38:09, 33.74s/it]
                                                         
{'loss': '1.22', 'grad_norm': '20.32', 'learning_rate': '1.378e-05', 'epoch': '0.3248'}

 32%|███▏      | 1624/5000 [21:04:39<31:38:09, 33.74s/it][2026-05-22 13:10:30,571] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 13:10:30,674] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 981])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])

 32%|███▎      | 1625/5000 [21:05:32<36:53:35, 39.35s/it]
                                                         
{'loss': '0.3371', 'grad_norm': '10.63', 'learning_rate': '1.378e-05', 'epoch': '0.325'}

 32%|███▎      | 1625/5000 [21:05:32<36:53:35, 39.35s/it][2026-05-22 13:11:22,708] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-22 13:11:23,102] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])

 33%|███▎      | 1626/5000 [21:06:26<40:53:47, 43.64s/it]
                                                         
{'loss': '0.3437', 'grad_norm': '10.59', 'learning_rate': '1.377e-05', 'epoch': '0.3252'}

 33%|███▎      | 1626/5000 [21:06:26<40:53:47, 43.64s/it][2026-05-22 13:12:12,769] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 240])
[2026-05-22 13:12:13,181] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 467])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])

 33%|███▎      | 1627/5000 [21:06:57<37:27:23, 39.98s/it]
                                                         
{'loss': '0.1788', 'grad_norm': '6.516', 'learning_rate': '1.377e-05', 'epoch': '0.3254'}

 33%|███▎      | 1627/5000 [21:06:57<37:27:23, 39.98s/it][2026-05-22 13:12:44,727] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])
[2026-05-22 13:12:45,346] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 370])

 33%|███▎      | 1628/5000 [21:07:29<35:09:33, 37.54s/it]
                                                         
{'loss': '0.01873', 'grad_norm': '2.661', 'learning_rate': '1.376e-05', 'epoch': '0.3256'}

 33%|███▎      | 1628/5000 [21:07:29<35:09:33, 37.54s/it][2026-05-22 13:13:16,272] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 209])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 293])
[2026-05-22 13:13:16,635] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 361])

 33%|███▎      | 1629/5000 [21:08:13<36:59:39, 39.51s/it]
                                                         
{'loss': '0.09696', 'grad_norm': '5.859', 'learning_rate': '1.376e-05', 'epoch': '0.3258'}

 33%|███▎      | 1629/5000 [21:08:13<36:59:39, 39.51s/it][2026-05-22 13:14:00,115] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 216])
[2026-05-22 13:14:00,817] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 241])

 33%|███▎      | 1630/5000 [21:08:57<38:21:26, 40.98s/it]
                                                         
{'loss': '0.05316', 'grad_norm': '5.009', 'learning_rate': '1.376e-05', 'epoch': '0.326'}

 33%|███▎      | 1630/5000 [21:08:57<38:21:26, 40.98s/it][2026-05-22 13:14:44,164] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 241])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 290])
[2026-05-22 13:14:44,321] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 33%|███▎      | 1631/5000 [21:09:44<39:55:01, 42.65s/it]
                                                         
{'loss': '0.2024', 'grad_norm': '8.314', 'learning_rate': '1.375e-05', 'epoch': '0.3262'}

 33%|███▎      | 1631/5000 [21:09:44<39:55:01, 42.65s/it][2026-05-22 13:15:31,004] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 473])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 13:15:31,653] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 33%|███▎      | 1632/5000 [21:10:31<41:02:40, 43.87s/it]
                                                         
{'loss': '0.1836', 'grad_norm': '5.979', 'learning_rate': '1.375e-05', 'epoch': '0.3264'}

 33%|███▎      | 1632/5000 [21:10:31<41:02:40, 43.87s/it][2026-05-22 13:16:18,524] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 385])
[2026-05-22 13:16:19,149] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 33%|███▎      | 1633/5000 [21:11:00<36:53:01, 39.44s/it]
                                                         
{'loss': '0.3279', 'grad_norm': '8.206', 'learning_rate': '1.374e-05', 'epoch': '0.3266'}

 33%|███▎      | 1633/5000 [21:11:00<36:53:01, 39.44s/it][2026-05-22 13:16:48,181] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 246])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 565])
[2026-05-22 13:16:48,449] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 33%|███▎      | 1634/5000 [21:11:32<34:54:50, 37.34s/it]
                                                         
{'loss': '0.06161', 'grad_norm': '4.404', 'learning_rate': '1.374e-05', 'epoch': '0.3268'}

 33%|███▎      | 1634/5000 [21:11:32<34:54:50, 37.34s/it][2026-05-22 13:17:19,512] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 56])
[2026-05-22 13:17:19,583] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 93])

 33%|███▎      | 1635/5000 [21:12:01<32:25:08, 34.68s/it]
                                                         
{'loss': '0.3148', 'grad_norm': '8.669', 'learning_rate': '1.373e-05', 'epoch': '0.327'}

 33%|███▎      | 1635/5000 [21:12:01<32:25:08, 34.68s/it][2026-05-22 13:17:47,973] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 200])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 76])
[2026-05-22 13:17:50,224] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 506])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 72])

 33%|███▎      | 1636/5000 [21:12:31<31:13:59, 33.42s/it]
                                                         
{'loss': '0.2022', 'grad_norm': '6.508', 'learning_rate': '1.373e-05', 'epoch': '0.3272'}

 33%|███▎      | 1636/5000 [21:12:31<31:13:59, 33.42s/it][2026-05-22 13:18:18,852] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 63])
[2026-05-22 13:18:18,915] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 67])

 33%|███▎      | 1637/5000 [21:13:01<30:08:58, 32.27s/it]
                                                         
{'loss': '0.3', 'grad_norm': '6.647', 'learning_rate': '1.373e-05', 'epoch': '0.3274'}

 33%|███▎      | 1637/5000 [21:13:01<30:08:58, 32.27s/it][2026-05-22 13:18:48,451] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 483])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 13:18:48,759] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 73])

 33%|███▎      | 1638/5000 [21:13:34<30:21:04, 32.50s/it]
                                                         
{'loss': '0.1852', 'grad_norm': '4.536', 'learning_rate': '1.372e-05', 'epoch': '0.3276'}

 33%|███▎      | 1638/5000 [21:13:34<30:21:04, 32.50s/it][2026-05-22 13:19:21,226] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 783])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 16])
[2026-05-22 13:19:21,250] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 18])

 33%|███▎      | 1639/5000 [21:14:04<29:47:57, 31.92s/it]
                                                         
{'loss': '0.2281', 'grad_norm': '7.283', 'learning_rate': '1.372e-05', 'epoch': '0.3278'}

 33%|███▎      | 1639/5000 [21:14:04<29:47:57, 31.92s/it][2026-05-22 13:19:51,820] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 30])
[2026-05-22 13:19:52,312] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 668])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 22])

 33%|███▎      | 1640/5000 [21:14:39<30:38:44, 32.83s/it]
                                                         
{'loss': '0.03294', 'grad_norm': '3.912', 'learning_rate': '1.371e-05', 'epoch': '0.328'}

 33%|███▎      | 1640/5000 [21:14:39<30:38:44, 32.83s/it][2026-05-22 13:20:27,860] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 636])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 267])
[2026-05-22 13:20:28,265] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 655])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 252])

 33%|███▎      | 1641/5000 [21:15:06<28:58:40, 31.06s/it]
                                                         
{'loss': '0.1201', 'grad_norm': '6.963', 'learning_rate': '1.371e-05', 'epoch': '0.3282'}

 33%|███▎      | 1641/5000 [21:15:06<28:58:40, 31.06s/it][2026-05-22 13:20:54,645] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 635])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 278])
[2026-05-22 13:20:55,298] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 649])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 280])

 33%|███▎      | 1642/5000 [21:15:36<28:31:48, 30.59s/it]
                                                         
{'loss': '0.05786', 'grad_norm': '5.165', 'learning_rate': '1.371e-05', 'epoch': '0.3284'}

 33%|███▎      | 1642/5000 [21:15:36<28:31:48, 30.59s/it][2026-05-22 13:21:22,955] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 483])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 260])
[2026-05-22 13:21:23,699] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 33%|███▎      | 1643/5000 [21:16:19<32:08:54, 34.48s/it]
                                                         
{'loss': '0.2445', 'grad_norm': '9.814', 'learning_rate': '1.37e-05', 'epoch': '0.3286'}

 33%|███▎      | 1643/5000 [21:16:19<32:08:54, 34.48s/it][2026-05-22 13:22:07,408] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 13:22:07,932] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 33%|███▎      | 1644/5000 [21:16:56<32:43:28, 35.10s/it]
                                                         
{'loss': '0.2608', 'grad_norm': '10.62', 'learning_rate': '1.37e-05', 'epoch': '0.3288'}

 33%|███▎      | 1644/5000 [21:16:56<32:43:28, 35.10s/it][2026-05-22 13:22:43,476] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 433])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 273])
[2026-05-22 13:22:43,824] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 266])

 33%|███▎      | 1645/5000 [21:17:32<33:00:52, 35.43s/it]
                                                         
{'loss': '3.916', 'grad_norm': '12.02', 'learning_rate': '1.369e-05', 'epoch': '0.329'}

 33%|███▎      | 1645/5000 [21:17:32<33:00:52, 35.43s/it][2026-05-22 13:23:19,430] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 252])
[2026-05-22 13:23:19,935] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 540])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 212])

 33%|███▎      | 1646/5000 [21:18:15<35:11:17, 37.77s/it]
                                                         
{'loss': '3.935', 'grad_norm': '9.993', 'learning_rate': '1.369e-05', 'epoch': '0.3292'}

 33%|███▎      | 1646/5000 [21:18:15<35:11:17, 37.77s/it][2026-05-22 13:24:02,550] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 286])
[2026-05-22 13:24:03,075] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 33%|███▎      | 1647/5000 [21:18:50<34:16:29, 36.80s/it]
                                                         
{'loss': '1.761', 'grad_norm': '12.4', 'learning_rate': '1.369e-05', 'epoch': '0.3294'}

 33%|███▎      | 1647/5000 [21:18:50<34:16:29, 36.80s/it][2026-05-22 13:24:37,786] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:24:38,318] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 33%|███▎      | 1648/5000 [21:19:25<33:48:16, 36.31s/it]
                                                         
{'loss': '1.707', 'grad_norm': '11.39', 'learning_rate': '1.368e-05', 'epoch': '0.3296'}

 33%|███▎      | 1648/5000 [21:19:25<33:48:16, 36.31s/it][2026-05-22 13:25:12,917] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 13:25:13,082] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])

 33%|███▎      | 1649/5000 [21:19:58<32:54:35, 35.36s/it]
                                                         
{'loss': '0.2597', 'grad_norm': '7.732', 'learning_rate': '1.368e-05', 'epoch': '0.3298'}

 33%|███▎      | 1649/5000 [21:19:58<32:54:35, 35.36s/it][2026-05-22 13:25:45,459] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 319])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 13:25:46,203] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 33%|███▎      | 1650/5000 [21:20:40<34:40:30, 37.26s/it]
                                                         
{'loss': '0.06785', 'grad_norm': '5.592', 'learning_rate': '1.367e-05', 'epoch': '0.33'}

 33%|███▎      | 1650/5000 [21:20:40<34:40:30, 37.26s/it][2026-05-22 13:26:27,077] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 221])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])
[2026-05-22 13:26:27,596] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 33%|███▎      | 1651/5000 [21:21:25<36:46:46, 39.54s/it]
                                                         
{'loss': '0.5418', 'grad_norm': '12.22', 'learning_rate': '1.367e-05', 'epoch': '0.3302'}

 33%|███▎      | 1651/5000 [21:21:25<36:46:46, 39.54s/it][2026-05-22 13:27:12,407] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 433])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 346])
[2026-05-22 13:27:13,348] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 467])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 33%|███▎      | 1652/5000 [21:22:09<38:12:51, 41.09s/it]
                                                         
{'loss': '0.3575', 'grad_norm': '11.88', 'learning_rate': '1.367e-05', 'epoch': '0.3304'}

 33%|███▎      | 1652/5000 [21:22:09<38:12:51, 41.09s/it][2026-05-22 13:27:57,113] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 231])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 76])
[2026-05-22 13:27:57,187] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 110])

 33%|███▎      | 1653/5000 [21:22:39<34:59:07, 37.63s/it]
                                                         
{'loss': '0.4229', 'grad_norm': '10.24', 'learning_rate': '1.366e-05', 'epoch': '0.3306'}

 33%|███▎      | 1653/5000 [21:22:39<34:59:07, 37.63s/it][2026-05-22 13:28:26,846] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 196])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 13:28:27,623] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 442])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 76])

 33%|███▎      | 1654/5000 [21:23:11<33:30:18, 36.05s/it]
                                                         
{'loss': '0.2564', 'grad_norm': '10.04', 'learning_rate': '1.366e-05', 'epoch': '0.3308'}

 33%|███▎      | 1654/5000 [21:23:11<33:30:18, 36.05s/it][2026-05-22 13:28:58,464] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 306])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])
[2026-05-22 13:28:59,097] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 33%|███▎      | 1655/5000 [21:23:42<32:03:04, 34.49s/it]
                                                         
{'loss': '0.598', 'grad_norm': '14.5', 'learning_rate': '1.365e-05', 'epoch': '0.331'}

 33%|███▎      | 1655/5000 [21:23:42<32:03:04, 34.49s/it][2026-05-22 13:29:29,806] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 292])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:29:30,842] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 444])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 33%|███▎      | 1656/5000 [21:24:13<31:05:16, 33.47s/it]
                                                         
{'loss': '0.8737', 'grad_norm': '19.4', 'learning_rate': '1.365e-05', 'epoch': '0.3312'}

 33%|███▎      | 1656/5000 [21:24:13<31:05:16, 33.47s/it][2026-05-22 13:30:00,623] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 266])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 13:30:00,952] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 290])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 33%|███▎      | 1657/5000 [21:24:44<30:27:04, 32.79s/it]
                                                         
{'loss': '0.2623', 'grad_norm': '8.456', 'learning_rate': '1.364e-05', 'epoch': '0.3314'}

 33%|███▎      | 1657/5000 [21:24:44<30:27:04, 32.79s/it][2026-05-22 13:30:31,579] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 259])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 13:30:32,266] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 330])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 33%|███▎      | 1658/5000 [21:25:16<30:03:16, 32.37s/it]
                                                         
{'loss': '0.04023', 'grad_norm': '3.912', 'learning_rate': '1.364e-05', 'epoch': '0.3316'}

 33%|███▎      | 1658/5000 [21:25:16<30:03:16, 32.37s/it][2026-05-22 13:31:07,640] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])
[2026-05-22 13:31:08,098] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 987])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 33%|███▎      | 1659/5000 [21:25:45<29:05:06, 31.34s/it]
                                                         
{'loss': '0.1162', 'grad_norm': '7.246', 'learning_rate': '1.364e-05', 'epoch': '0.3318'}

 33%|███▎      | 1659/5000 [21:25:45<29:05:06, 31.34s/it][2026-05-22 13:31:36,682] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 13:31:37,465] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 33%|███▎      | 1660/5000 [21:26:15<28:54:38, 31.16s/it]
                                                         
{'loss': '0.09313', 'grad_norm': '6.537', 'learning_rate': '1.363e-05', 'epoch': '0.332'}

 33%|███▎      | 1660/5000 [21:26:15<28:54:38, 31.16s/it][2026-05-22 13:32:03,294] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 278])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:32:03,553] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 250])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 33%|███▎      | 1661/5000 [21:26:45<28:30:28, 30.74s/it]
                                                         
{'loss': '0.3109', 'grad_norm': '8.583', 'learning_rate': '1.363e-05', 'epoch': '0.3322'}

 33%|███▎      | 1661/5000 [21:26:45<28:30:28, 30.74s/it][2026-05-22 13:32:33,697] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 257])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:32:33,732] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 523])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 33%|███▎      | 1662/5000 [21:27:18<28:56:11, 31.21s/it]
                                                         
{'loss': '0.06067', 'grad_norm': '5.77', 'learning_rate': '1.362e-05', 'epoch': '0.3324'}

 33%|███▎      | 1662/5000 [21:27:18<28:56:11, 31.21s/it][2026-05-22 13:33:04,753] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 745])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 254])
[2026-05-22 13:33:06,205] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])

 33%|███▎      | 1663/5000 [21:27:47<28:30:20, 30.75s/it]
                                                         
{'loss': '0.7812', 'grad_norm': '18.35', 'learning_rate': '1.362e-05', 'epoch': '0.3326'}

 33%|███▎      | 1663/5000 [21:27:47<28:30:20, 30.75s/it][2026-05-22 13:33:35,240] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1014])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])
[2026-05-22 13:33:36,126] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 970])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 241])

 33%|███▎      | 1664/5000 [21:28:18<28:33:42, 30.82s/it]
                                                         
{'loss': '0.5504', 'grad_norm': '17.53', 'learning_rate': '1.362e-05', 'epoch': '0.3328'}

 33%|███▎      | 1664/5000 [21:28:18<28:33:42, 30.82s/it][2026-05-22 13:34:06,114] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 238])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 244])
[2026-05-22 13:34:06,293] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 33%|███▎      | 1665/5000 [21:29:09<33:58:18, 36.67s/it]
                                                         
{'loss': '0.2298', 'grad_norm': '8.165', 'learning_rate': '1.361e-05', 'epoch': '0.333'}

 33%|███▎      | 1665/5000 [21:29:09<33:58:18, 36.67s/it][2026-05-22 13:34:57,057] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 13:34:57,304] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 33%|███▎      | 1666/5000 [21:30:01<38:13:08, 41.27s/it]
                                                         
{'loss': '0.1789', 'grad_norm': '20.59', 'learning_rate': '1.361e-05', 'epoch': '0.3332'}

 33%|███▎      | 1666/5000 [21:30:01<38:13:08, 41.27s/it][2026-05-22 13:35:48,071] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 294])
[2026-05-22 13:35:48,685] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 33%|███▎      | 1667/5000 [21:30:28<34:28:51, 37.24s/it]
                                                         
{'loss': '0.5192', 'grad_norm': '9.092', 'learning_rate': '1.36e-05', 'epoch': '0.3334'}

 33%|███▎      | 1667/5000 [21:30:28<34:28:51, 37.24s/it][2026-05-22 13:36:15,932] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 13:36:16,032] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 338])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 33%|███▎      | 1668/5000 [21:31:02<33:20:05, 36.02s/it]
                                                         
{'loss': '0.01645', 'grad_norm': '2.728', 'learning_rate': '1.36e-05', 'epoch': '0.3336'}

 33%|███▎      | 1668/5000 [21:31:02<33:20:05, 36.02s/it][2026-05-22 13:36:49,531] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 13:36:49,813] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 657])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 33%|███▎      | 1669/5000 [21:31:47<36:04:41, 38.99s/it]
                                                         
{'loss': '0.9253', 'grad_norm': '16.19', 'learning_rate': '1.36e-05', 'epoch': '0.3338'}

 33%|███▎      | 1669/5000 [21:31:47<36:04:41, 38.99s/it][2026-05-22 13:37:36,159] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 670])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 13:37:36,210] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 634])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 33%|███▎      | 1670/5000 [21:32:34<38:17:47, 41.40s/it]
                                                         
{'loss': '0.4395', 'grad_norm': '11.47', 'learning_rate': '1.359e-05', 'epoch': '0.334'}

 33%|███▎      | 1670/5000 [21:32:34<38:17:47, 41.40s/it][2026-05-22 13:38:22,062] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 641])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 13:38:22,693] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])

 33%|███▎      | 1671/5000 [21:33:06<35:33:08, 38.45s/it]
                                                         
{'loss': '0.457', 'grad_norm': '8.71', 'learning_rate': '1.359e-05', 'epoch': '0.3342'}

 33%|███▎      | 1671/5000 [21:33:06<35:33:08, 38.45s/it][2026-05-22 13:38:53,583] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 637])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 13:38:53,639] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 660])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 33%|███▎      | 1672/5000 [21:33:38<33:44:42, 36.50s/it]
                                                         
{'loss': '0.03256', 'grad_norm': '4.078', 'learning_rate': '1.358e-05', 'epoch': '0.3344'}

 33%|███▎      | 1672/5000 [21:33:38<33:44:42, 36.50s/it][2026-05-22 13:39:25,019] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 590])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 13:39:25,620] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 33%|███▎      | 1673/5000 [21:34:09<32:12:43, 34.86s/it]
                                                         
{'loss': '1.035', 'grad_norm': '21.42', 'learning_rate': '1.358e-05', 'epoch': '0.3346'}

 33%|███▎      | 1673/5000 [21:34:09<32:12:43, 34.86s/it][2026-05-22 13:39:56,628] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-22 13:39:56,678] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 668])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 33%|███▎      | 1674/5000 [21:34:40<31:03:07, 33.61s/it]
                                                         
{'loss': '1.83', 'grad_norm': '25.57', 'learning_rate': '1.358e-05', 'epoch': '0.3348'}

 33%|███▎      | 1674/5000 [21:34:40<31:03:07, 33.61s/it][2026-05-22 13:40:26,370] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 13:40:26,909] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 34%|███▎      | 1675/5000 [21:35:18<32:14:40, 34.91s/it]
                                                         
{'loss': '1.478', 'grad_norm': '11.99', 'learning_rate': '1.357e-05', 'epoch': '0.335'}

 34%|███▎      | 1675/5000 [21:35:18<32:14:40, 34.91s/it][2026-05-22 13:41:05,453] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 13:41:05,551] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 955])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])

 34%|███▎      | 1676/5000 [21:35:56<33:16:22, 36.04s/it]
                                                         
{'loss': '1.251', 'grad_norm': '9.501', 'learning_rate': '1.357e-05', 'epoch': '0.3352'}

 34%|███▎      | 1676/5000 [21:35:56<33:16:22, 36.04s/it][2026-05-22 13:41:45,722] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])
[2026-05-22 13:41:46,488] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 34%|███▎      | 1677/5000 [21:36:34<33:47:06, 36.60s/it]
                                                         
{'loss': '1.101', 'grad_norm': '11.17', 'learning_rate': '1.356e-05', 'epoch': '0.3354'}

 34%|███▎      | 1677/5000 [21:36:34<33:47:06, 36.60s/it][2026-05-22 13:42:24,409] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 223])
[2026-05-22 13:42:24,632] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 34%|███▎      | 1678/5000 [21:37:12<34:10:01, 37.03s/it]
                                                         
{'loss': '1.165', 'grad_norm': '12.05', 'learning_rate': '1.356e-05', 'epoch': '0.3356'}

 34%|███▎      | 1678/5000 [21:37:12<34:10:01, 37.03s/it][2026-05-22 13:43:01,027] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 13:43:02,569] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 984])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▎      | 1679/5000 [21:37:46<33:16:58, 36.08s/it]
                                                         
{'loss': '2.956', 'grad_norm': '6.496', 'learning_rate': '1.356e-05', 'epoch': '0.3358'}

 34%|███▎      | 1679/5000 [21:37:46<33:16:58, 36.08s/it][2026-05-22 13:43:34,803] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 973])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 13:43:36,115] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 978])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▎      | 1680/5000 [21:38:21<32:56:07, 35.71s/it]
                                                         
{'loss': '2.994', 'grad_norm': '6.643', 'learning_rate': '1.355e-05', 'epoch': '0.336'}

 34%|███▎      | 1680/5000 [21:38:21<32:56:07, 35.71s/it][2026-05-22 13:44:08,906] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 233])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 13:44:09,316] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 481])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 34%|███▎      | 1681/5000 [21:38:46<29:58:15, 32.51s/it]
                                                         
{'loss': '0.9329', 'grad_norm': '16.15', 'learning_rate': '1.355e-05', 'epoch': '0.3362'}

 34%|███▎      | 1681/5000 [21:38:46<29:58:15, 32.51s/it][2026-05-22 13:44:34,201] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 233])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 13:44:34,875] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▎      | 1682/5000 [21:39:29<32:54:44, 35.71s/it]
                                                         
{'loss': '1.23', 'grad_norm': '18.48', 'learning_rate': '1.354e-05', 'epoch': '0.3364'}

 34%|███▎      | 1682/5000 [21:39:29<32:54:44, 35.71s/it][2026-05-22 13:45:16,384] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 398])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 13:45:17,393] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 432])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▎      | 1683/5000 [21:40:16<35:52:02, 38.93s/it]
                                                         
{'loss': '0.546', 'grad_norm': '17.38', 'learning_rate': '1.354e-05', 'epoch': '0.3366'}

 34%|███▎      | 1683/5000 [21:40:16<35:52:02, 38.93s/it][2026-05-22 13:46:02,955] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:46:03,205] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▎      | 1684/5000 [21:41:02<37:48:42, 41.05s/it]
                                                         
{'loss': '0.5326', 'grad_norm': '17.95', 'learning_rate': '1.353e-05', 'epoch': '0.3368'}

 34%|███▎      | 1684/5000 [21:41:02<37:48:42, 41.05s/it][2026-05-22 13:46:49,783] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 633])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:46:50,009] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 639])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 34%|███▎      | 1685/5000 [21:41:51<39:59:51, 43.44s/it]
                                                         
{'loss': '0.097', 'grad_norm': '6.991', 'learning_rate': '1.353e-05', 'epoch': '0.337'}

 34%|███▎      | 1685/5000 [21:41:51<39:59:51, 43.44s/it][2026-05-22 13:47:39,562] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 662])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 13:47:39,576] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 650])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 34%|███▎      | 1686/5000 [21:42:40<41:43:36, 45.33s/it]
                                                         
{'loss': '0.08335', 'grad_norm': '5.685', 'learning_rate': '1.353e-05', 'epoch': '0.3372'}

 34%|███▎      | 1686/5000 [21:42:40<41:43:36, 45.33s/it][2026-05-22 13:48:28,060] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 628])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 86])
[2026-05-22 13:48:29,995] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 673])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 77])

 34%|███▎      | 1687/5000 [21:43:14<38:35:32, 41.94s/it]
                                                         
{'loss': '0.1356', 'grad_norm': '5.128', 'learning_rate': '1.352e-05', 'epoch': '0.3374'}

 34%|███▎      | 1687/5000 [21:43:14<38:35:32, 41.94s/it][2026-05-22 13:49:01,892] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 638])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 91])
[2026-05-22 13:49:03,385] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 637])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 53])

 34%|███▍      | 1688/5000 [21:43:49<36:27:40, 39.63s/it]
                                                         
{'loss': '0.0327', 'grad_norm': '4.989', 'learning_rate': '1.352e-05', 'epoch': '0.3376'}

 34%|███▍      | 1688/5000 [21:43:49<36:27:40, 39.63s/it][2026-05-22 13:49:36,715] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 694])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])
[2026-05-22 13:49:36,804] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 74])

 34%|███▍      | 1689/5000 [21:44:20<34:06:28, 37.09s/it]
                                                         
{'loss': '0.8278', 'grad_norm': '19.08', 'learning_rate': '1.351e-05', 'epoch': '0.3378'}

 34%|███▍      | 1689/5000 [21:44:20<34:06:28, 37.09s/it][2026-05-22 13:50:07,874] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 60])
[2026-05-22 13:50:08,725] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 73])

 34%|███▍      | 1690/5000 [21:44:51<32:27:31, 35.30s/it]
                                                         
{'loss': '0.7649', 'grad_norm': '19.68', 'learning_rate': '1.351e-05', 'epoch': '0.338'}

 34%|███▍      | 1690/5000 [21:44:51<32:27:31, 35.30s/it][2026-05-22 13:50:38,120] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 13:50:39,352] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 34%|███▍      | 1691/5000 [21:45:27<32:37:28, 35.49s/it]
                                                         
{'loss': '1.365', 'grad_norm': '9.307', 'learning_rate': '1.351e-05', 'epoch': '0.3382'}

 34%|███▍      | 1691/5000 [21:45:27<32:37:28, 35.49s/it][2026-05-22 13:51:14,563] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 13:51:14,776] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 643])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 34%|███▍      | 1692/5000 [21:46:06<33:31:17, 36.48s/it]
                                                         
{'loss': '1.224', 'grad_norm': '8.753', 'learning_rate': '1.35e-05', 'epoch': '0.3384'}

 34%|███▍      | 1692/5000 [21:46:06<33:31:17, 36.48s/it][2026-05-22 13:51:53,555] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 642])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 145])
[2026-05-22 13:51:54,078] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 593])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1693/5000 [21:46:40<32:47:36, 35.70s/it]
                                                         
{'loss': '0.558', 'grad_norm': '8.404', 'learning_rate': '1.35e-05', 'epoch': '0.3386'}

 34%|███▍      | 1693/5000 [21:46:40<32:47:36, 35.70s/it][2026-05-22 13:52:26,833] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 694])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:52:28,428] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 577])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1694/5000 [21:47:12<31:53:08, 34.72s/it]
                                                         
{'loss': '0.4578', 'grad_norm': '6.091', 'learning_rate': '1.349e-05', 'epoch': '0.3388'}

 34%|███▍      | 1694/5000 [21:47:12<31:53:08, 34.72s/it][2026-05-22 13:52:58,903] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 296])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:53:00,342] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 987])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1695/5000 [21:47:45<31:28:39, 34.29s/it]
                                                         
{'loss': '2.28', 'grad_norm': '3.515', 'learning_rate': '1.349e-05', 'epoch': '0.339'}

 34%|███▍      | 1695/5000 [21:47:45<31:28:39, 34.29s/it][2026-05-22 13:53:33,275] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 878])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:53:33,958] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1696/5000 [21:48:18<31:06:42, 33.90s/it]
                                                         
{'loss': '2.277', 'grad_norm': '4.009', 'learning_rate': '1.349e-05', 'epoch': '0.3392'}

 34%|███▍      | 1696/5000 [21:48:18<31:06:42, 33.90s/it][2026-05-22 13:54:07,193] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 13:54:07,793] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 145])

 34%|███▍      | 1697/5000 [21:48:57<32:26:19, 35.36s/it]
                                                         
{'loss': '1.698', 'grad_norm': '7.055', 'learning_rate': '1.348e-05', 'epoch': '0.3394'}

 34%|███▍      | 1697/5000 [21:48:57<32:26:19, 35.36s/it][2026-05-22 13:54:45,575] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 13:54:45,831] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1698/5000 [21:49:36<33:23:14, 36.40s/it]
                                                         
{'loss': '1.559', 'grad_norm': '4.009', 'learning_rate': '1.348e-05', 'epoch': '0.3396'}

 34%|███▍      | 1698/5000 [21:49:36<33:23:14, 36.40s/it][2026-05-22 13:55:22,913] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 310])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 13:55:24,120] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 34%|███▍      | 1699/5000 [21:50:12<33:13:33, 36.24s/it]
                                                         
{'loss': '1.318', 'grad_norm': '5.422', 'learning_rate': '1.347e-05', 'epoch': '0.3398'}

 34%|███▍      | 1699/5000 [21:50:12<33:13:33, 36.24s/it][2026-05-22 13:55:59,412] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 315])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 13:56:00,040] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 309])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 34%|███▍      | 1700/5000 [21:50:48<33:18:28, 36.34s/it]
                                                         
{'loss': '1.284', 'grad_norm': '5.589', 'learning_rate': '1.347e-05', 'epoch': '0.34'}

 34%|███▍      | 1700/5000 [21:50:48<33:18:28, 36.34s/it][2026-05-22 13:56:36,240] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 778])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 13:56:36,700] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 34%|███▍      | 1701/5000 [21:51:36<36:29:12, 39.82s/it]
                                                         
{'loss': '0.5544', 'grad_norm': '13.03', 'learning_rate': '1.347e-05', 'epoch': '0.3402'}

 34%|███▍      | 1701/5000 [21:51:36<36:29:12, 39.82s/it][2026-05-22 13:57:23,501] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 346])
[2026-05-22 13:57:24,958] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])

 34%|███▍      | 1702/5000 [21:52:24<38:46:21, 42.32s/it]
                                                         
{'loss': '0.1602', 'grad_norm': '7.484', 'learning_rate': '1.346e-05', 'epoch': '0.3404'}

 34%|███▍      | 1702/5000 [21:52:24<38:46:21, 42.32s/it][2026-05-22 13:58:11,915] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 52])
[2026-05-22 13:58:14,196] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 75])

 34%|███▍      | 1703/5000 [21:53:14<40:47:49, 44.55s/it]
                                                         
{'loss': '2.796', 'grad_norm': '13.2', 'learning_rate': '1.346e-05', 'epoch': '0.3406'}

 34%|███▍      | 1703/5000 [21:53:14<40:47:49, 44.55s/it][2026-05-22 13:59:02,522] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])
[2026-05-22 13:59:03,202] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 73])

 34%|███▍      | 1704/5000 [21:54:04<42:18:36, 46.21s/it]
                                                         
{'loss': '2.754', 'grad_norm': '30.08', 'learning_rate': '1.345e-05', 'epoch': '0.3408'}

 34%|███▍      | 1704/5000 [21:54:04<42:18:36, 46.21s/it][2026-05-22 13:59:51,415] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 13:59:51,536] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 179])

 34%|███▍      | 1705/5000 [21:54:34<37:45:30, 41.25s/it]
                                                         
{'loss': '0.233', 'grad_norm': '9.355', 'learning_rate': '1.345e-05', 'epoch': '0.341'}

 34%|███▍      | 1705/5000 [21:54:34<37:45:30, 41.25s/it][2026-05-22 14:00:20,601] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 338])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 14:00:21,699] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 431])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 174])

 34%|███▍      | 1706/5000 [21:55:02<34:07:11, 37.29s/it]
                                                         
{'loss': '0.1198', 'grad_norm': '6.122', 'learning_rate': '1.344e-05', 'epoch': '0.3412'}

 34%|███▍      | 1706/5000 [21:55:02<34:07:11, 37.29s/it][2026-05-22 14:00:50,003] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 256])
[2026-05-22 14:00:50,512] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 457])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])

 34%|███▍      | 1707/5000 [21:55:50<37:06:02, 40.56s/it]
                                                         
{'loss': '0.1774', 'grad_norm': '5.068', 'learning_rate': '1.344e-05', 'epoch': '0.3414'}

 34%|███▍      | 1707/5000 [21:55:50<37:06:02, 40.56s/it][2026-05-22 14:01:37,897] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 398])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 128])
[2026-05-22 14:01:38,056] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 209])

 34%|███▍      | 1708/5000 [21:56:37<38:51:57, 42.50s/it]
                                                         
{'loss': '0.1757', 'grad_norm': '5.397', 'learning_rate': '1.344e-05', 'epoch': '0.3416'}

 34%|███▍      | 1708/5000 [21:56:37<38:51:57, 42.50s/it][2026-05-22 14:02:28,766] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 185])
[2026-05-22 14:02:29,029] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1709/5000 [21:57:24<40:07:06, 43.89s/it]
                                                         
{'loss': '2.454', 'grad_norm': '9.767', 'learning_rate': '1.343e-05', 'epoch': '0.3418'}

 34%|███▍      | 1709/5000 [21:57:24<40:07:06, 43.89s/it][2026-05-22 14:03:15,193] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 14:03:17,323] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 978])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1710/5000 [21:58:12<41:07:14, 45.00s/it]
                                                         
{'loss': '2.402', 'grad_norm': '14.01', 'learning_rate': '1.343e-05', 'epoch': '0.342'}

 34%|███▍      | 1710/5000 [21:58:12<41:07:14, 45.00s/it][2026-05-22 14:03:59,654] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 264])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 14:04:00,567] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 323])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 34%|███▍      | 1711/5000 [21:58:40<36:35:04, 40.04s/it]
                                                         
{'loss': '0.7632', 'grad_norm': '19.83', 'learning_rate': '1.342e-05', 'epoch': '0.3422'}

 34%|███▍      | 1711/5000 [21:58:40<36:35:04, 40.04s/it][2026-05-22 14:04:28,396] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 308])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 14:04:29,009] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 279])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 34%|███▍      | 1712/5000 [21:59:07<33:00:26, 36.14s/it]
                                                         
{'loss': '0.6545', 'grad_norm': '18.42', 'learning_rate': '1.342e-05', 'epoch': '0.3424'}

 34%|███▍      | 1712/5000 [21:59:07<33:00:26, 36.14s/it][2026-05-22 14:04:54,743] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 249])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 14:04:56,656] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1713/5000 [21:59:38<31:24:08, 34.39s/it]
                                                         
{'loss': '1.149', 'grad_norm': '22', 'learning_rate': '1.342e-05', 'epoch': '0.3426'}

 34%|███▍      | 1713/5000 [21:59:38<31:24:08, 34.39s/it][2026-05-22 14:05:25,257] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 307])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 14:05:27,001] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1714/5000 [22:00:06<29:51:48, 32.72s/it]
                                                         
{'loss': '1.528', 'grad_norm': '26.3', 'learning_rate': '1.341e-05', 'epoch': '0.3428'}

 34%|███▍      | 1714/5000 [22:00:06<29:51:48, 32.72s/it][2026-05-22 14:05:54,345] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 310])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 14:05:54,775] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 433])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1715/5000 [22:00:57<34:43:09, 38.05s/it]
                                                         
{'loss': '0.1269', 'grad_norm': '12.46', 'learning_rate': '1.341e-05', 'epoch': '0.343'}

 34%|███▍      | 1715/5000 [22:00:58<34:43:09, 38.05s/it][2026-05-22 14:06:45,532] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 304])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 14:06:45,905] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 507])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 34%|███▍      | 1716/5000 [22:01:49<38:32:06, 42.24s/it]
                                                         
{'loss': '0.1837', 'grad_norm': '9.883', 'learning_rate': '1.34e-05', 'epoch': '0.3432'}

 34%|███▍      | 1716/5000 [22:01:49<38:32:06, 42.24s/it][2026-05-22 14:07:36,047] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 400])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 228])
[2026-05-22 14:07:37,123] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 428])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 245])

 34%|███▍      | 1717/5000 [22:02:18<34:58:42, 38.36s/it]
                                                         
{'loss': '0.4376', 'grad_norm': '9.794', 'learning_rate': '1.34e-05', 'epoch': '0.3434'}

 34%|███▍      | 1717/5000 [22:02:18<34:58:42, 38.36s/it][2026-05-22 14:08:06,748] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])
[2026-05-22 14:08:07,438] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 171])

 34%|███▍      | 1718/5000 [22:02:48<32:41:51, 35.87s/it]
                                                         
{'loss': '0.02601', 'grad_norm': '4.068', 'learning_rate': '1.34e-05', 'epoch': '0.3436'}

 34%|███▍      | 1718/5000 [22:02:48<32:41:51, 35.87s/it][2026-05-22 14:08:36,168] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 14:08:36,235] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 264])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1719/5000 [22:03:19<31:15:55, 34.31s/it]
                                                         
{'loss': '0.1899', 'grad_norm': '6.536', 'learning_rate': '1.339e-05', 'epoch': '0.3438'}

 34%|███▍      | 1719/5000 [22:03:19<31:15:55, 34.31s/it][2026-05-22 14:09:07,012] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 14:09:07,728] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1720/5000 [22:03:51<30:39:41, 33.65s/it]
                                                         
{'loss': '0.04085', 'grad_norm': '4.717', 'learning_rate': '1.339e-05', 'epoch': '0.344'}

 34%|███▍      | 1720/5000 [22:03:51<30:39:41, 33.65s/it][2026-05-22 14:09:38,903] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 14:09:40,645] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 970])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 34%|███▍      | 1721/5000 [22:04:22<29:58:08, 32.90s/it]
                                                         
{'loss': '0.1454', 'grad_norm': '6.694', 'learning_rate': '1.338e-05', 'epoch': '0.3442'}

 34%|███▍      | 1721/5000 [22:04:22<29:58:08, 32.90s/it][2026-05-22 14:10:10,144] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 975])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 14:10:11,713] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 978])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 34%|███▍      | 1722/5000 [22:04:57<30:20:18, 33.32s/it]
                                                         
{'loss': '0.01399', 'grad_norm': '3.161', 'learning_rate': '1.338e-05', 'epoch': '0.3444'}

 34%|███▍      | 1722/5000 [22:04:57<30:20:18, 33.32s/it][2026-05-22 14:10:47,991] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 19])
[2026-05-22 14:10:48,572] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 18])

 34%|███▍      | 1723/5000 [22:05:26<29:21:47, 32.26s/it]
                                                         
{'loss': '0.863', 'grad_norm': '12.12', 'learning_rate': '1.338e-05', 'epoch': '0.3446'}

 34%|███▍      | 1723/5000 [22:05:26<29:21:47, 32.26s/it][2026-05-22 14:11:17,889] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 16])
[2026-05-22 14:11:18,017] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 18])

 34%|███▍      | 1724/5000 [22:06:09<32:14:59, 35.44s/it]
                                                         
{'loss': '1.098', 'grad_norm': '19', 'learning_rate': '1.337e-05', 'epoch': '0.3448'}

 34%|███▍      | 1724/5000 [22:06:09<32:14:59, 35.44s/it][2026-05-22 14:11:56,478] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 300])
[2026-05-22 14:11:57,564] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 426])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 258])

 34%|███▍      | 1725/5000 [22:06:39<30:49:31, 33.88s/it]
                                                         
{'loss': '0.4736', 'grad_norm': '9.716', 'learning_rate': '1.337e-05', 'epoch': '0.345'}

 34%|███▍      | 1725/5000 [22:06:39<30:49:31, 33.88s/it][2026-05-22 14:12:27,912] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 492])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])
[2026-05-22 14:12:27,994] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 294])

 35%|███▍      | 1726/5000 [22:07:13<30:35:23, 33.64s/it]
                                                         
{'loss': '0.04183', 'grad_norm': '3.757', 'learning_rate': '1.336e-05', 'epoch': '0.3452'}

 35%|███▍      | 1726/5000 [22:07:13<30:35:23, 33.64s/it][2026-05-22 14:13:00,871] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 667])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 55])
[2026-05-22 14:13:01,249] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 651])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 109])

 35%|███▍      | 1727/5000 [22:08:01<34:38:15, 38.10s/it]
                                                         
{'loss': '0.1204', 'grad_norm': '5.869', 'learning_rate': '1.336e-05', 'epoch': '0.3454'}

 35%|███▍      | 1727/5000 [22:08:01<34:38:15, 38.10s/it][2026-05-22 14:13:48,739] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 626])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 58])
[2026-05-22 14:13:50,375] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 630])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 96])

 35%|███▍      | 1728/5000 [22:08:51<37:52:32, 41.67s/it]
                                                         
{'loss': '0.03386', 'grad_norm': '3.609', 'learning_rate': '1.336e-05', 'epoch': '0.3456'}

 35%|███▍      | 1728/5000 [22:08:51<37:52:32, 41.67s/it][2026-05-22 14:14:38,576] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 626])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 14:14:39,178] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 647])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 35%|███▍      | 1729/5000 [22:09:37<39:08:40, 43.08s/it]
                                                         
{'loss': '0.4852', 'grad_norm': '22.65', 'learning_rate': '1.335e-05', 'epoch': '0.3458'}

 35%|███▍      | 1729/5000 [22:09:37<39:08:40, 43.08s/it][2026-05-22 14:15:24,974] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 628])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 14:15:25,587] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 649])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 35%|███▍      | 1730/5000 [22:10:24<39:59:17, 44.02s/it]
                                                         
{'loss': '0.4035', 'grad_norm': '16.44', 'learning_rate': '1.335e-05', 'epoch': '0.346'}

 35%|███▍      | 1730/5000 [22:10:24<39:59:17, 44.02s/it][2026-05-22 14:16:11,024] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 403])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 84])
[2026-05-22 14:16:11,064] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 517])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 134])

 35%|███▍      | 1731/5000 [22:10:52<35:38:20, 39.25s/it]
                                                         
{'loss': '0.4001', 'grad_norm': '13.03', 'learning_rate': '1.334e-05', 'epoch': '0.3462'}

 35%|███▍      | 1731/5000 [22:10:52<35:38:20, 39.25s/it][2026-05-22 14:16:38,521] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 179])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 14:16:39,745] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 438])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 108])

 35%|███▍      | 1732/5000 [22:11:21<32:53:55, 36.24s/it]
                                                         
{'loss': '0.2266', 'grad_norm': '8.675', 'learning_rate': '1.334e-05', 'epoch': '0.3464'}

 35%|███▍      | 1732/5000 [22:11:21<32:53:55, 36.24s/it][2026-05-22 14:17:07,693] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-22 14:17:08,880] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 439])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 35%|███▍      | 1733/5000 [22:11:54<31:59:40, 35.26s/it]
                                                         
{'loss': '1.024', 'grad_norm': '8.78', 'learning_rate': '1.333e-05', 'epoch': '0.3466'}

 35%|███▍      | 1733/5000 [22:11:54<31:59:40, 35.26s/it][2026-05-22 14:17:41,142] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 501])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 14:17:41,269] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 167])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])

 35%|███▍      | 1734/5000 [22:12:26<31:00:03, 34.17s/it]
                                                         
{'loss': '1.061', 'grad_norm': '10.26', 'learning_rate': '1.333e-05', 'epoch': '0.3468'}

 35%|███▍      | 1734/5000 [22:12:26<31:00:03, 34.17s/it][2026-05-22 14:18:13,403] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 64])
[2026-05-22 14:18:14,458] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 495])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 89])

 35%|███▍      | 1735/5000 [22:13:03<31:52:30, 35.15s/it]
                                                         
{'loss': '0.9477', 'grad_norm': '14.65', 'learning_rate': '1.333e-05', 'epoch': '0.347'}

 35%|███▍      | 1735/5000 [22:13:03<31:52:30, 35.15s/it][2026-05-22 14:18:50,165] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 171])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 67])
[2026-05-22 14:18:52,353] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 496])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 97])

 35%|███▍      | 1736/5000 [22:13:40<32:27:25, 35.80s/it]
                                                         
{'loss': '0.9613', 'grad_norm': '13.23', 'learning_rate': '1.332e-05', 'epoch': '0.3472'}

 35%|███▍      | 1736/5000 [22:13:40<32:27:25, 35.80s/it][2026-05-22 14:19:26,997] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 221])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])
[2026-05-22 14:19:28,126] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 479])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 346])

 35%|███▍      | 1737/5000 [22:14:09<30:31:46, 33.68s/it]
                                                         
{'loss': '0.7397', 'grad_norm': '19.7', 'learning_rate': '1.332e-05', 'epoch': '0.3474'}

 35%|███▍      | 1737/5000 [22:14:09<30:31:46, 33.68s/it][2026-05-22 14:19:56,222] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 306])
[2026-05-22 14:19:56,369] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 175])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 35%|███▍      | 1738/5000 [22:14:36<28:35:08, 31.55s/it]
                                                         
{'loss': '0.7145', 'grad_norm': '19.94', 'learning_rate': '1.331e-05', 'epoch': '0.3476'}

 35%|███▍      | 1738/5000 [22:14:36<28:35:08, 31.55s/it][2026-05-22 14:20:23,932] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 14:20:24,331] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 35%|███▍      | 1739/5000 [22:15:03<27:29:33, 30.35s/it]
                                                         
{'loss': '0.2964', 'grad_norm': '8.754', 'learning_rate': '1.331e-05', 'epoch': '0.3478'}

 35%|███▍      | 1739/5000 [22:15:03<27:29:33, 30.35s/it][2026-05-22 14:20:51,344] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 14:20:51,707] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 35%|███▍      | 1740/5000 [22:15:32<26:59:16, 29.80s/it]
                                                         
{'loss': '0.3387', 'grad_norm': '9.261', 'learning_rate': '1.331e-05', 'epoch': '0.348'}

 35%|███▍      | 1740/5000 [22:15:32<26:59:16, 29.80s/it][2026-05-22 14:21:18,786] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 96])
[2026-05-22 14:21:19,735] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 35%|███▍      | 1741/5000 [22:16:00<26:31:09, 29.29s/it]
                                                         
{'loss': '0.05084', 'grad_norm': '5.327', 'learning_rate': '1.33e-05', 'epoch': '0.3482'}

 35%|███▍      | 1741/5000 [22:16:00<26:31:09, 29.29s/it][2026-05-22 14:21:47,383] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 31])
[2026-05-22 14:21:47,740] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 181])

 35%|███▍      | 1742/5000 [22:16:28<26:14:27, 29.00s/it]
                                                         
{'loss': '0.08039', 'grad_norm': '6.171', 'learning_rate': '1.33e-05', 'epoch': '0.3484'}

 35%|███▍      | 1742/5000 [22:16:28<26:14:27, 29.00s/it][2026-05-22 14:22:15,759] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 64])
[2026-05-22 14:22:15,973] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 506])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 136])

 35%|███▍      | 1743/5000 [22:16:58<26:32:35, 29.34s/it]
                                                         
{'loss': '0.3606', 'grad_norm': '9.72', 'learning_rate': '1.329e-05', 'epoch': '0.3486'}

 35%|███▍      | 1743/5000 [22:16:58<26:32:35, 29.34s/it][2026-05-22 14:22:45,242] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 173])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 95])
[2026-05-22 14:22:46,943] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 464])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 70])

 35%|███▍      | 1744/5000 [22:17:28<26:37:22, 29.44s/it]
                                                         
{'loss': '0.3563', 'grad_norm': '9.352', 'learning_rate': '1.329e-05', 'epoch': '0.3488'}

 35%|███▍      | 1744/5000 [22:17:28<26:37:22, 29.44s/it][2026-05-22 14:23:14,768] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 409])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 254])
[2026-05-22 14:23:16,035] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 431])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 35%|███▍      | 1745/5000 [22:18:16<31:39:11, 35.01s/it]
                                                         
{'loss': '1.581', 'grad_norm': '11.77', 'learning_rate': '1.329e-05', 'epoch': '0.349'}

 35%|███▍      | 1745/5000 [22:18:16<31:39:11, 35.01s/it][2026-05-22 14:24:03,415] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 467])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 14:24:03,911] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 35%|███▍      | 1746/5000 [22:19:04<35:12:25, 38.95s/it]
                                                         
{'loss': '1.581', 'grad_norm': '9.512', 'learning_rate': '1.328e-05', 'epoch': '0.3492'}

 35%|███▍      | 1746/5000 [22:19:04<35:12:25, 38.95s/it][2026-05-22 14:24:51,972] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 397])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 14:24:52,447] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 480])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])

 35%|███▍      | 1747/5000 [22:19:48<36:32:41, 40.44s/it]
                                                         
{'loss': '1.807', 'grad_norm': '21.59', 'learning_rate': '1.328e-05', 'epoch': '0.3494'}

 35%|███▍      | 1747/5000 [22:19:48<36:32:41, 40.44s/it][2026-05-22 14:25:35,586] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 302])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 14:25:37,083] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 440])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 35%|███▍      | 1748/5000 [22:20:32<37:36:43, 41.64s/it]
                                                         
{'loss': '1.684', 'grad_norm': '19.58', 'learning_rate': '1.327e-05', 'epoch': '0.3496'}

 35%|███▍      | 1748/5000 [22:20:32<37:36:43, 41.64s/it][2026-05-22 14:26:23,850] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])
[2026-05-22 14:26:24,798] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])

 35%|███▍      | 1749/5000 [22:21:03<34:32:15, 38.25s/it]
                                                         
{'loss': '0.4048', 'grad_norm': '14.33', 'learning_rate': '1.327e-05', 'epoch': '0.3498'}

 35%|███▍      | 1749/5000 [22:21:03<34:32:15, 38.25s/it][2026-05-22 14:26:55,061] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])
[2026-05-22 14:26:55,429] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])

 35%|███▌      | 1750/5000 [22:21:33<32:19:05, 35.80s/it]
                                                         
{'loss': '0.414', 'grad_norm': '14.39', 'learning_rate': '1.327e-05', 'epoch': '0.35'}

 35%|███▌      | 1750/5000 [22:21:33<32:19:05, 35.80s/it][2026-05-22 14:27:18,005] INFO [src.utils.basic_utils:19] Saving model to /workspace/VLM2Vec/outputs/Qwen3.5-08b-continued-mixed-alltasks/checkpoint-1750
[2026-05-22 14:27:19,476] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-22 14:27:19,520] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-22 14:27:20,515] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 430])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])
[2026-05-22 14:27:21,290] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 404])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 183])

 35%|███▌      | 1751/5000 [22:22:04<31:07:20, 34.48s/it]
                                                         
{'loss': '0.5279', 'grad_norm': '12.32', 'learning_rate': '1.326e-05', 'epoch': '0.3502'}

 35%|███▌      | 1751/5000 [22:22:04<31:07:20, 34.48s/it][2026-05-22 14:27:51,048] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 176])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 14:27:52,263] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 479])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 35%|███▌      | 1752/5000 [22:22:36<30:15:17, 33.53s/it]
                                                         
{'loss': '0.2766', 'grad_norm': '12.34', 'learning_rate': '1.326e-05', 'epoch': '0.3504'}

 35%|███▌      | 1752/5000 [22:22:36<30:15:17, 33.53s/it][2026-05-22 14:28:23,463] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 298])
[2026-05-22 14:28:23,739] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 507])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 272])

 35%|███▌      | 1753/5000 [22:23:10<30:23:29, 33.70s/it]
                                                         
{'loss': '0.4492', 'grad_norm': '7.724', 'learning_rate': '1.325e-05', 'epoch': '0.3506'}

 35%|███▌      | 1753/5000 [22:23:10<30:23:29, 33.70s/it][2026-05-22 14:28:57,908] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 296])
[2026-05-22 14:28:57,926] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 292])

 35%|███▌      | 1754/5000 [22:23:42<30:05:29, 33.37s/it]
                                                         
{'loss': '0.01929', 'grad_norm': '2.777', 'learning_rate': '1.325e-05', 'epoch': '0.3508'}

 35%|███▌      | 1754/5000 [22:23:42<30:05:29, 33.37s/it][2026-05-22 14:29:28,627] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 93])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 14:29:29,239] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 143])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 35%|███▌      | 1755/5000 [22:24:28<33:32:59, 37.22s/it]
                                                         
{'loss': '0.09188', 'grad_norm': '6.149', 'learning_rate': '1.324e-05', 'epoch': '0.351'}

 35%|███▌      | 1755/5000 [22:24:28<33:32:59, 37.22s/it][2026-05-22 14:30:14,799] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 144])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 14:30:15,393] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 48])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 35%|███▌      | 1756/5000 [22:25:15<35:59:09, 39.94s/it]
                                                         
{'loss': '0.1141', 'grad_norm': '10.9', 'learning_rate': '1.324e-05', 'epoch': '0.3512'}

 35%|███▌      | 1756/5000 [22:25:15<35:59:09, 39.94s/it][2026-05-22 14:31:02,262] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 511])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 308])
[2026-05-22 14:31:02,526] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 259])

 35%|███▌      | 1757/5000 [22:25:42<32:36:15, 36.19s/it]
                                                         
{'loss': '0.2394', 'grad_norm': '7.915', 'learning_rate': '1.324e-05', 'epoch': '0.3514'}

 35%|███▌      | 1757/5000 [22:25:42<32:36:15, 36.19s/it][2026-05-22 14:31:29,743] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 312])
[2026-05-22 14:31:30,496] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])

 35%|███▌      | 1758/5000 [22:26:10<30:14:18, 33.58s/it]
                                                         
{'loss': '0.2356', 'grad_norm': '7.483', 'learning_rate': '1.323e-05', 'epoch': '0.3516'}

 35%|███▌      | 1758/5000 [22:26:10<30:14:18, 33.58s/it][2026-05-22 14:31:56,781] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-22 14:31:57,946] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 215])

 35%|███▌      | 1759/5000 [22:26:40<29:14:15, 32.48s/it]
                                                         
{'loss': '0.5131', 'grad_norm': '9.952', 'learning_rate': '1.323e-05', 'epoch': '0.3518'}

 35%|███▌      | 1759/5000 [22:26:40<29:14:15, 32.48s/it][2026-05-22 14:32:26,607] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 207])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-22 14:32:27,863] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 480])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

 35%|███▌      | 1760/5000 [22:27:09<28:29:52, 31.66s/it]
                                                         
{'loss': '0.03124', 'grad_norm': '2.949', 'learning_rate': '1.322e-05', 'epoch': '0.352'}

 35%|███▌      | 1760/5000 [22:27:09<28:29:52, 31.66s/it][2026-05-22 14:32:56,791] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 984])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 14:32:57,040] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 978])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 35%|███▌      | 1761/5000 [22:27:33<26:17:29, 29.22s/it]
                                                         
{'loss': '0.8168', 'grad_norm': '19.61', 'learning_rate': '1.322e-05', 'epoch': '0.3522'}

 35%|███▌      | 1761/5000 [22:27:33<26:17:29, 29.22s/it][2026-05-22 14:33:20,339] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 14:33:20,470] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 973])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 35%|███▌      | 1762/5000 [22:27:56<24:37:56, 27.39s/it]
                                                         
{'loss': '0.7534', 'grad_norm': '18.55', 'learning_rate': '1.322e-05', 'epoch': '0.3524'}

 35%|███▌      | 1762/5000 [22:27:56<24:37:56, 27.39s/it][2026-05-22 14:33:47,046] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1010])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 88])
[2026-05-22 14:33:47,672] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 93])

 35%|███▌      | 1763/5000 [22:28:27<25:32:31, 28.41s/it]
                                                         
{'loss': '0.4351', 'grad_norm': '9.485', 'learning_rate': '1.321e-05', 'epoch': '0.3526'}

 35%|███▌      | 1763/5000 [22:28:27<25:32:31, 28.41s/it][2026-05-22 14:34:18,057] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 14:34:19,813] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1017])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 65])

 35%|███▌      | 1764/5000 [22:28:58<26:11:59, 29.15s/it]
                                                         
{'loss': '0.07391', 'grad_norm': '4.876', 'learning_rate': '1.321e-05', 'epoch': '0.3528'}

 35%|███▌      | 1764/5000 [22:28:58<26:11:59, 29.15s/it][2026-05-22 14:34:46,215] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 14:34:46,401] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 35%|███▌      | 1765/5000 [22:29:26<26:01:41, 28.96s/it]
                                                         
{'loss': '0.2108', 'grad_norm': '7.25', 'learning_rate': '1.32e-05', 'epoch': '0.353'}

 35%|███▌      | 1765/5000 [22:29:26<26:01:41, 28.96s/it][2026-05-22 14:35:14,854] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])
[2026-05-22 14:35:15,332] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 35%|███▌      | 1766/5000 [22:29:56<26:11:25, 29.15s/it]
                                                         
{'loss': '0.2475', 'grad_norm': '8.013', 'learning_rate': '1.32e-05', 'epoch': '0.3532'}

 35%|███▌      | 1766/5000 [22:29:56<26:11:25, 29.15s/it][2026-05-22 14:35:42,591] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 319])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])
[2026-05-22 14:35:43,756] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 438])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])

 35%|███▌      | 1767/5000 [22:30:39<30:06:20, 33.52s/it]
                                                         
{'loss': '0.6138', 'grad_norm': '18.18', 'learning_rate': '1.32e-05', 'epoch': '0.3534'}

 35%|███▌      | 1767/5000 [22:30:40<30:06:20, 33.52s/it][2026-05-22 14:36:26,718] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 197])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 14:36:27,357] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 435])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])

 35%|███▌      | 1768/5000 [22:31:24<33:02:46, 36.81s/it]
                                                         
{'loss': '0.5606', 'grad_norm': '17.23', 'learning_rate': '1.319e-05', 'epoch': '0.3536'}

 35%|███▌      | 1768/5000 [22:31:24<33:02:46, 36.81s/it][2026-05-22 14:37:11,880] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 14:37:12,343] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])

 35%|███▌      | 1769/5000 [22:32:11<35:39:12, 39.73s/it]
                                                         
{'loss': '0.5897', 'grad_norm': '19.48', 'learning_rate': '1.319e-05', 'epoch': '0.3538'}

 35%|███▌      | 1769/5000 [22:32:11<35:39:12, 39.73s/it][2026-05-22 14:37:58,393] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 14:37:58,746] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 35%|███▌      | 1770/5000 [22:32:57<37:25:58, 41.72s/it]
                                                         
{'loss': '0.6966', 'grad_norm': '17.19', 'learning_rate': '1.318e-05', 'epoch': '0.354'}

 35%|███▌      | 1770/5000 [22:32:57<37:25:58, 41.72s/it][2026-05-22 14:38:45,661] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 14:38:46,366] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 35%|███▌      | 1771/5000 [22:33:42<38:17:17, 42.69s/it]
                                                         
{'loss': '2.618', 'grad_norm': '14.66', 'learning_rate': '1.318e-05', 'epoch': '0.3542'}

 35%|███▌      | 1771/5000 [22:33:42<38:17:17, 42.69s/it][2026-05-22 14:39:31,468] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 14:39:32,137] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 35%|███▌      | 1772/5000 [22:34:28<39:12:33, 43.73s/it]
                                                         
{'loss': '2.599', 'grad_norm': '30.56', 'learning_rate': '1.318e-05', 'epoch': '0.3544'}

 35%|███▌      | 1772/5000 [22:34:28<39:12:33, 43.73s/it][2026-05-22 14:40:15,815] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])
[2026-05-22 14:40:16,073] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 35%|███▌      | 1773/5000 [22:34:55<34:46:43, 38.80s/it]
                                                         
{'loss': '0.2726', 'grad_norm': '8.499', 'learning_rate': '1.317e-05', 'epoch': '0.3546'}

 35%|███▌      | 1773/5000 [22:34:55<34:46:43, 38.80s/it][2026-05-22 14:40:43,350] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 470])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 14:40:43,873] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 35%|███▌      | 1774/5000 [22:35:23<31:47:35, 35.48s/it]
                                                         
{'loss': '0.2909', 'grad_norm': '8.645', 'learning_rate': '1.317e-05', 'epoch': '0.3548'}

 35%|███▌      | 1774/5000 [22:35:23<31:47:35, 35.48s/it][2026-05-22 14:41:10,292] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 320])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 14:41:11,699] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 36%|███▌      | 1775/5000 [22:36:12<35:32:44, 39.68s/it]
                                                         
{'loss': '0.4204', 'grad_norm': '8.242', 'learning_rate': '1.316e-05', 'epoch': '0.355'}

 36%|███▌      | 1775/5000 [22:36:12<35:32:44, 39.68s/it][2026-05-22 14:42:00,764] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 426])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 14:42:00,974] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 331])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 36%|███▌      | 1776/5000 [22:37:01<37:51:24, 42.27s/it]
                                                         
{'loss': '0.1304', 'grad_norm': '2.064', 'learning_rate': '1.316e-05', 'epoch': '0.3552'}

 36%|███▌      | 1776/5000 [22:37:01<37:51:24, 42.27s/it][2026-05-22 14:42:48,638] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 251])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 14:42:49,088] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 432])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 36%|███▌      | 1777/5000 [22:37:50<39:34:09, 44.20s/it]
                                                         
{'loss': '2.659', 'grad_norm': '46.05', 'learning_rate': '1.316e-05', 'epoch': '0.3554'}

 36%|███▌      | 1777/5000 [22:37:50<39:34:09, 44.20s/it][2026-05-22 14:43:37,568] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 14:43:38,305] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 501])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])

 36%|███▌      | 1778/5000 [22:38:41<41:26:55, 46.31s/it]
                                                         
{'loss': '2.345', 'grad_norm': '12.4', 'learning_rate': '1.315e-05', 'epoch': '0.3556'}

 36%|███▌      | 1778/5000 [22:38:41<41:26:55, 46.31s/it][2026-05-22 14:44:27,608] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 293])
[2026-05-22 14:44:28,581] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 433])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 302])

 36%|███▌      | 1779/5000 [22:39:10<36:57:41, 41.31s/it]
                                                         
{'loss': '3.797', 'grad_norm': '228.2', 'learning_rate': '1.315e-05', 'epoch': '0.3558'}

 36%|███▌      | 1779/5000 [22:39:10<36:57:41, 41.31s/it][2026-05-22 14:44:57,718] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 430])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])
[2026-05-22 14:44:58,336] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 321])

 36%|███▌      | 1780/5000 [22:39:43<34:44:14, 38.84s/it]
                                                         
{'loss': '3.112', 'grad_norm': '224.1', 'learning_rate': '1.314e-05', 'epoch': '0.356'}

 36%|███▌      | 1780/5000 [22:39:43<34:44:14, 38.84s/it][2026-05-22 14:45:31,007] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 307])
[2026-05-22 14:45:31,038] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 241])

 36%|███▌      | 1781/5000 [22:40:16<32:58:34, 36.88s/it]
                                                         
{'loss': '0.2509', 'grad_norm': '8.026', 'learning_rate': '1.314e-05', 'epoch': '0.3562'}

 36%|███▌      | 1781/5000 [22:40:16<32:58:34, 36.88s/it][2026-05-22 14:46:03,453] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 693])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 247])
[2026-05-22 14:46:04,086] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 282])

 36%|███▌      | 1782/5000 [22:40:48<31:36:19, 35.36s/it]
                                                         
{'loss': '0.03031', 'grad_norm': '3.984', 'learning_rate': '1.313e-05', 'epoch': '0.3564'}

 36%|███▌      | 1782/5000 [22:40:48<31:36:19, 35.36s/it][2026-05-22 14:46:34,224] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 106])
[2026-05-22 14:46:34,770] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 82])

 36%|███▌      | 1783/5000 [22:41:18<30:15:04, 33.85s/it]
                                                         
{'loss': '0.2941', 'grad_norm': '7.72', 'learning_rate': '1.313e-05', 'epoch': '0.3566'}

 36%|███▌      | 1783/5000 [22:41:18<30:15:04, 33.85s/it][2026-05-22 14:47:05,702] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 797])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 48])
[2026-05-22 14:47:06,446] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 115])

 36%|███▌      | 1784/5000 [22:41:52<30:17:48, 33.91s/it]
                                                         
{'loss': '0.04762', 'grad_norm': '3.783', 'learning_rate': '1.313e-05', 'epoch': '0.3568'}

 36%|███▌      | 1784/5000 [22:41:52<30:17:48, 33.91s/it][2026-05-22 14:47:39,647] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 222])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 249])
[2026-05-22 14:47:40,109] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 474])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 36%|███▌      | 1785/5000 [22:42:20<28:41:28, 32.13s/it]
                                                         
{'loss': '0.3138', 'grad_norm': '7.177', 'learning_rate': '1.312e-05', 'epoch': '0.357'}

 36%|███▌      | 1785/5000 [22:42:20<28:41:28, 32.13s/it][2026-05-22 14:48:07,644] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 493])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 14:48:08,195] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 36%|███▌      | 1786/5000 [22:42:50<28:06:02, 31.48s/it]
                                                         
{'loss': '0.1335', 'grad_norm': '5.239', 'learning_rate': '1.312e-05', 'epoch': '0.3572'}

 36%|███▌      | 1786/5000 [22:42:50<28:06:02, 31.48s/it][2026-05-22 14:48:37,180] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 233])
[2026-05-22 14:48:37,672] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 36%|███▌      | 1787/5000 [22:43:19<27:30:58, 30.83s/it]
                                                         
{'loss': '0.3598', 'grad_norm': '7.671', 'learning_rate': '1.311e-05', 'epoch': '0.3574'}

 36%|███▌      | 1787/5000 [22:43:19<27:30:58, 30.83s/it][2026-05-22 14:49:06,964] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 14:49:07,797] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 399])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 36%|███▌      | 1788/5000 [22:43:57<29:24:56, 32.97s/it]
                                                         
{'loss': '0.1143', 'grad_norm': '9.481', 'learning_rate': '1.311e-05', 'epoch': '0.3576'}

 36%|███▌      | 1788/5000 [22:43:57<29:24:56, 32.97s/it][2026-05-22 14:49:48,267] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 169])
[2026-05-22 14:49:50,034] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])

 36%|███▌      | 1789/5000 [22:44:24<27:44:46, 31.11s/it]
                                                         
{'loss': '1.17', 'grad_norm': '21.26', 'learning_rate': '1.311e-05', 'epoch': '0.3578'}

 36%|███▌      | 1789/5000 [22:44:24<27:44:46, 31.11s/it][2026-05-22 14:50:14,495] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 51])
[2026-05-22 14:50:16,677] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1019])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 55])

 36%|███▌      | 1790/5000 [22:45:10<31:37:58, 35.48s/it]
                                                         
{'loss': '1.423', 'grad_norm': '21.15', 'learning_rate': '1.31e-05', 'epoch': '0.358'}

 36%|███▌      | 1790/5000 [22:45:10<31:37:58, 35.48s/it][2026-05-22 14:50:56,855] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 417])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 14:50:57,970] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 439])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 36%|███▌      | 1791/5000 [22:45:41<30:37:41, 34.36s/it]
                                                         
{'loss': '0.5818', 'grad_norm': '12.95', 'learning_rate': '1.31e-05', 'epoch': '0.3582'}

 36%|███▌      | 1791/5000 [22:45:41<30:37:41, 34.36s/it][2026-05-22 14:51:29,215] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 280])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 14:51:29,642] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 497])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 366])

 36%|███▌      | 1792/5000 [22:46:13<30:00:04, 33.67s/it]
                                                         
{'loss': '0.2656', 'grad_norm': '11.59', 'learning_rate': '1.309e-05', 'epoch': '0.3584'}

 36%|███▌      | 1792/5000 [22:46:13<30:00:04, 33.67s/it][2026-05-22 14:52:00,814] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 262])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 14:52:00,849] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 232])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 36%|███▌      | 1793/5000 [22:46:45<29:19:12, 32.91s/it]
                                                         
{'loss': '1.123', 'grad_norm': '20.5', 'learning_rate': '1.309e-05', 'epoch': '0.3586'}

 36%|███▌      | 1793/5000 [22:46:45<29:19:12, 32.91s/it][2026-05-22 14:52:31,880] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 275])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 14:52:31,970] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 263])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 36%|███▌      | 1794/5000 [22:47:16<28:53:02, 32.43s/it]
                                                         
{'loss': '1.049', 'grad_norm': '18.82', 'learning_rate': '1.309e-05', 'epoch': '0.3588'}

 36%|███▌      | 1794/5000 [22:47:16<28:53:02, 32.43s/it][2026-05-22 14:53:03,722] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 14:53:03,728] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 257])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])

 36%|███▌      | 1795/5000 [22:48:04<33:01:52, 37.10s/it]
                                                         
{'loss': '1.025', 'grad_norm': '18.69', 'learning_rate': '1.308e-05', 'epoch': '0.359'}

 36%|███▌      | 1795/5000 [22:48:04<33:01:52, 37.10s/it][2026-05-22 14:53:51,831] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 14:53:51,852] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 36%|███▌      | 1796/5000 [22:48:50<35:29:46, 39.88s/it]
                                                         
{'loss': '1.514', 'grad_norm': '27.7', 'learning_rate': '1.308e-05', 'epoch': '0.3592'}

 36%|███▌      | 1796/5000 [22:48:50<35:29:46, 39.88s/it][2026-05-22 14:54:36,926] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])
[2026-05-22 14:54:37,618] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 412])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 306])

 36%|███▌      | 1797/5000 [22:49:21<32:55:40, 37.01s/it]
                                                         
{'loss': '0.1063', 'grad_norm': '6.07', 'learning_rate': '1.307e-05', 'epoch': '0.3594'}

 36%|███▌      | 1797/5000 [22:49:21<32:55:40, 37.01s/it][2026-05-22 14:55:08,092] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 14:55:08,181] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 492])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 326])

 36%|███▌      | 1798/5000 [22:49:52<31:28:23, 35.39s/it]
                                                         
{'loss': '0.05685', 'grad_norm': '5.019', 'learning_rate': '1.307e-05', 'epoch': '0.3596'}

 36%|███▌      | 1798/5000 [22:49:52<31:28:23, 35.39s/it][2026-05-22 14:55:39,474] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 14:55:40,177] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 398])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 36%|███▌      | 1799/5000 [22:50:21<29:37:08, 33.31s/it]
                                                         
{'loss': '1.239', 'grad_norm': '21.74', 'learning_rate': '1.307e-05', 'epoch': '0.3598'}

 36%|███▌      | 1799/5000 [22:50:21<29:37:08, 33.31s/it][2026-05-22 14:56:07,948] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 358])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 14:56:07,960] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 380])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 36%|███▌      | 1800/5000 [22:50:49<28:13:43, 31.76s/it]
                                                         
{'loss': '0.7795', 'grad_norm': '18.83', 'learning_rate': '1.306e-05', 'epoch': '0.36'}

 36%|███▌      | 1800/5000 [22:50:49<28:13:43, 31.76s/it][2026-05-22 14:56:36,253] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 788])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 351])
[2026-05-22 14:56:36,447] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 398])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 174])

 36%|███▌      | 1801/5000 [22:51:18<27:41:01, 31.15s/it]
                                                         
{'loss': '0.3797', 'grad_norm': '7.971', 'learning_rate': '1.306e-05', 'epoch': '0.3602'}

 36%|███▌      | 1801/5000 [22:51:19<27:41:01, 31.15s/it][2026-05-22 14:57:07,220] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 794])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 332])
[2026-05-22 14:57:07,230] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 794])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 23])

 36%|███▌      | 1802/5000 [22:51:52<28:11:57, 31.74s/it]
                                                         
{'loss': '0.04729', 'grad_norm': '6.187', 'learning_rate': '1.305e-05', 'epoch': '0.3604'}

 36%|███▌      | 1802/5000 [22:51:52<28:11:57, 31.74s/it][2026-05-22 14:57:39,200] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])
[2026-05-22 14:57:39,492] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])

 36%|███▌      | 1803/5000 [22:52:20<27:19:24, 30.77s/it]
                                                         
{'loss': '0.7017', 'grad_norm': '16.09', 'learning_rate': '1.305e-05', 'epoch': '0.3606'}

 36%|███▌      | 1803/5000 [22:52:20<27:19:24, 30.77s/it][2026-05-22 14:58:07,623] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])
[2026-05-22 14:58:07,665] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])

 36%|███▌      | 1804/5000 [22:52:52<27:29:24, 30.97s/it]
                                                         
{'loss': '0.6695', 'grad_norm': '16.37', 'learning_rate': '1.304e-05', 'epoch': '0.3608'}

 36%|███▌      | 1804/5000 [22:52:52<27:29:24, 30.97s/it][2026-05-22 14:58:38,753] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 290])
[2026-05-22 14:58:40,473] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 36%|███▌      | 1805/5000 [22:53:22<27:26:22, 30.92s/it]
                                                         
{'loss': '0.529', 'grad_norm': '13.86', 'learning_rate': '1.304e-05', 'epoch': '0.361'}

 36%|███▌      | 1805/5000 [22:53:22<27:26:22, 30.92s/it][2026-05-22 14:59:11,288] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 14:59:11,353] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 430])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 36%|███▌      | 1806/5000 [22:53:53<27:18:50, 30.79s/it]
                                                         
{'loss': '1.409', 'grad_norm': '20.41', 'learning_rate': '1.304e-05', 'epoch': '0.3612'}

 36%|███▌      | 1806/5000 [22:53:53<27:18:50, 30.79s/it][2026-05-22 14:59:40,588] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 300])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 14:59:40,828] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 365])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 36%|███▌      | 1807/5000 [22:54:32<29:36:03, 33.37s/it]
                                                         
{'loss': '1.075', 'grad_norm': '13.34', 'learning_rate': '1.303e-05', 'epoch': '0.3614'}

 36%|███▌      | 1807/5000 [22:54:32<29:36:03, 33.37s/it][2026-05-22 15:00:20,231] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 271])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:00:20,235] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 255])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 36%|███▌      | 1808/5000 [22:55:13<31:33:45, 35.60s/it]
                                                         
{'loss': '1.711', 'grad_norm': '21.11', 'learning_rate': '1.303e-05', 'epoch': '0.3616'}

 36%|███▌      | 1808/5000 [22:55:13<31:33:45, 35.60s/it][2026-05-22 15:00:59,559] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 290])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 185])
[2026-05-22 15:01:00,546] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 255])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 36%|███▌      | 1809/5000 [22:55:42<29:50:26, 33.67s/it]
                                                         
{'loss': '0.7417', 'grad_norm': '20.61', 'learning_rate': '1.302e-05', 'epoch': '0.3618'}

 36%|███▌      | 1809/5000 [22:55:42<29:50:26, 33.67s/it][2026-05-22 15:01:29,664] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 309])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 15:01:29,679] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 358])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 36%|███▌      | 1810/5000 [22:56:11<28:35:03, 32.26s/it]
                                                         
{'loss': '0.5033', 'grad_norm': '16.62', 'learning_rate': '1.302e-05', 'epoch': '0.362'}

 36%|███▌      | 1810/5000 [22:56:11<28:35:03, 32.26s/it][2026-05-22 15:01:58,096] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 156])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 15:01:58,100] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 156])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 36%|███▌      | 1811/5000 [22:56:44<28:38:33, 32.33s/it]
                                                         
{'loss': '0.5305', 'grad_norm': '7.024', 'learning_rate': '1.302e-05', 'epoch': '0.3622'}

 36%|███▌      | 1811/5000 [22:56:44<28:38:33, 32.33s/it][2026-05-22 15:02:30,090] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 184])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 15:02:30,768] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 156])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

 36%|███▌      | 1812/5000 [22:57:15<28:14:35, 31.89s/it]
                                                         
{'loss': '0.05359', 'grad_norm': '3.204', 'learning_rate': '1.301e-05', 'epoch': '0.3624'}

 36%|███▌      | 1812/5000 [22:57:15<28:14:35, 31.89s/it][2026-05-22 15:03:01,635] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 260])
[2026-05-22 15:03:02,152] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])

 36%|███▋      | 1813/5000 [22:57:45<27:52:28, 31.49s/it]
                                                         
{'loss': '0.4416', 'grad_norm': '7.499', 'learning_rate': '1.301e-05', 'epoch': '0.3626'}

 36%|███▋      | 1813/5000 [22:57:45<27:52:28, 31.49s/it][2026-05-22 15:03:32,855] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 351])
[2026-05-22 15:03:33,172] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 310])

 36%|███▋      | 1814/5000 [22:58:15<27:20:26, 30.89s/it]
                                                         
{'loss': '0.06584', 'grad_norm': '4.316', 'learning_rate': '1.3e-05', 'epoch': '0.3628'}

 36%|███▋      | 1814/5000 [22:58:15<27:20:26, 30.89s/it][2026-05-22 15:04:02,532] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])
[2026-05-22 15:04:02,643] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 692])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 204])

 36%|███▋      | 1815/5000 [22:58:42<26:23:53, 29.84s/it]
                                                         
{'loss': '0.6083', 'grad_norm': '9.5', 'learning_rate': '1.3e-05', 'epoch': '0.363'}

 36%|███▋      | 1815/5000 [22:58:42<26:23:53, 29.84s/it][2026-05-22 15:04:29,859] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 788])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 65])
[2026-05-22 15:04:30,411] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 793])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 90])

 36%|███▋      | 1816/5000 [22:59:13<26:37:37, 30.11s/it]
                                                         
{'loss': '0.353', 'grad_norm': '9.146', 'learning_rate': '1.3e-05', 'epoch': '0.3632'}

 36%|███▋      | 1816/5000 [22:59:13<26:37:37, 30.11s/it][2026-05-22 15:04:59,266] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])
[2026-05-22 15:04:59,356] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])

 36%|███▋      | 1817/5000 [22:59:35<24:34:55, 27.80s/it]
                                                         
{'loss': '0.7609', 'grad_norm': '18.14', 'learning_rate': '1.299e-05', 'epoch': '0.3634'}

 36%|███▋      | 1817/5000 [22:59:35<24:34:55, 27.80s/it][2026-05-22 15:05:22,199] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 213])
[2026-05-22 15:05:22,888] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 166])

 36%|███▋      | 1818/5000 [22:59:58<23:23:13, 26.46s/it]
                                                         
{'loss': '0.7606', 'grad_norm': '17.36', 'learning_rate': '1.299e-05', 'epoch': '0.3636'}

 36%|███▋      | 1818/5000 [22:59:58<23:23:13, 26.46s/it][2026-05-22 15:05:46,265] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 299])
[2026-05-22 15:05:47,460] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 501])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 340])

 36%|███▋      | 1819/5000 [23:00:27<23:59:35, 27.15s/it]
                                                         
{'loss': '0.4499', 'grad_norm': '8.953', 'learning_rate': '1.298e-05', 'epoch': '0.3638'}

 36%|███▋      | 1819/5000 [23:00:27<23:59:35, 27.15s/it][2026-05-22 15:06:15,063] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 275])
[2026-05-22 15:06:16,209] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 454])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 273])

 36%|███▋      | 1820/5000 [23:00:56<24:32:14, 27.78s/it]
                                                         
{'loss': '0.1576', 'grad_norm': '6.212', 'learning_rate': '1.298e-05', 'epoch': '0.364'}

 36%|███▋      | 1820/5000 [23:00:56<24:32:14, 27.78s/it][2026-05-22 15:06:43,596] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 270])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 15:06:44,595] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 456])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 36%|███▋      | 1821/5000 [23:01:31<26:12:50, 29.69s/it]
                                                         
{'loss': '1.009', 'grad_norm': '19.6', 'learning_rate': '1.298e-05', 'epoch': '0.3642'}

 36%|███▋      | 1821/5000 [23:01:31<26:12:50, 29.69s/it][2026-05-22 15:07:18,758] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 320])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:07:18,928] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 36%|███▋      | 1822/5000 [23:02:09<28:26:09, 32.21s/it]
                                                         
{'loss': '2.193', 'grad_norm': '29.76', 'learning_rate': '1.297e-05', 'epoch': '0.3644'}

 36%|███▋      | 1822/5000 [23:02:09<28:26:09, 32.21s/it][2026-05-22 15:07:56,518] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 15:07:57,800] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 497])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 36%|███▋      | 1823/5000 [23:02:37<27:22:38, 31.02s/it]
                                                         
{'loss': '0.6396', 'grad_norm': '11.17', 'learning_rate': '1.297e-05', 'epoch': '0.3646'}

 36%|███▋      | 1823/5000 [23:02:37<27:22:38, 31.02s/it][2026-05-22 15:08:24,380] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 285])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:08:26,033] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 437])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 36%|███▋      | 1824/5000 [23:03:15<29:20:10, 33.25s/it]
                                                         
{'loss': '0.653', 'grad_norm': '12.75', 'learning_rate': '1.296e-05', 'epoch': '0.3648'}

 36%|███▋      | 1824/5000 [23:03:15<29:20:10, 33.25s/it][2026-05-22 15:09:02,730] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 402])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 15:09:03,944] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 479])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 36%|███▋      | 1825/5000 [23:03:47<28:49:40, 32.69s/it]
                                                         
{'loss': '0.3633', 'grad_norm': '7.853', 'learning_rate': '1.296e-05', 'epoch': '0.365'}

 36%|███▋      | 1825/5000 [23:03:47<28:49:40, 32.69s/it][2026-05-22 15:09:34,901] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:09:35,496] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 501])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1826/5000 [23:04:17<28:13:09, 32.01s/it]
                                                         
{'loss': '0.02097', 'grad_norm': '2.35', 'learning_rate': '1.296e-05', 'epoch': '0.3652'}

 37%|███▋      | 1826/5000 [23:04:17<28:13:09, 32.01s/it][2026-05-22 15:10:05,103] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])
[2026-05-22 15:10:05,522] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 254])

 37%|███▋      | 1827/5000 [23:04:49<28:04:02, 31.84s/it]
                                                         
{'loss': '0.298', 'grad_norm': '6.886', 'learning_rate': '1.295e-05', 'epoch': '0.3654'}

 37%|███▋      | 1827/5000 [23:04:49<28:04:02, 31.84s/it][2026-05-22 15:10:36,989] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 417])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 267])
[2026-05-22 15:10:37,100] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])

 37%|███▋      | 1828/5000 [23:05:21<28:14:33, 32.05s/it]
                                                         
{'loss': '0.0174', 'grad_norm': '2.266', 'learning_rate': '1.295e-05', 'epoch': '0.3656'}

 37%|███▋      | 1828/5000 [23:05:21<28:14:33, 32.05s/it][2026-05-22 15:11:08,209] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 145])
[2026-05-22 15:11:08,284] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1829/5000 [23:05:53<28:14:27, 32.06s/it]
                                                         
{'loss': '0.2497', 'grad_norm': '6.342', 'learning_rate': '1.294e-05', 'epoch': '0.3658'}

 37%|███▋      | 1829/5000 [23:05:53<28:14:27, 32.06s/it][2026-05-22 15:11:40,387] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 968])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:11:40,424] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1830/5000 [23:06:24<27:50:59, 31.63s/it]
                                                         
{'loss': '0.01682', 'grad_norm': '2.238', 'learning_rate': '1.294e-05', 'epoch': '0.366'}

 37%|███▋      | 1830/5000 [23:06:24<27:50:59, 31.63s/it][2026-05-22 15:12:13,042] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])
[2026-05-22 15:12:14,650] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 37%|███▋      | 1831/5000 [23:06:57<28:07:11, 31.94s/it]
                                                         
{'loss': '0.1176', 'grad_norm': '4.632', 'learning_rate': '1.293e-05', 'epoch': '0.3662'}

 37%|███▋      | 1831/5000 [23:06:57<28:07:11, 31.94s/it][2026-05-22 15:12:46,365] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 996])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 15:12:47,033] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1832/5000 [23:07:31<28:41:08, 32.60s/it]
                                                         
{'loss': '0.06758', 'grad_norm': '4.381', 'learning_rate': '1.293e-05', 'epoch': '0.3664'}

 37%|███▋      | 1832/5000 [23:07:31<28:41:08, 32.60s/it][2026-05-22 15:13:18,866] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 319])
[2026-05-22 15:13:18,896] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 181])

 37%|███▋      | 1833/5000 [23:08:00<27:47:41, 31.60s/it]
                                                         
{'loss': '0.812', 'grad_norm': '16.45', 'learning_rate': '1.293e-05', 'epoch': '0.3666'}

 37%|███▋      | 1833/5000 [23:08:00<27:47:41, 31.60s/it][2026-05-22 15:13:47,518] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1006])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 158])
[2026-05-22 15:13:47,875] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 988])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 183])

 37%|███▋      | 1834/5000 [23:08:30<27:20:02, 31.08s/it]
                                                         
{'loss': '0.3076', 'grad_norm': '10.81', 'learning_rate': '1.292e-05', 'epoch': '0.3668'}

 37%|███▋      | 1834/5000 [23:08:30<27:20:02, 31.08s/it][2026-05-22 15:14:17,514] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 249])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 267])
[2026-05-22 15:14:18,436] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 37%|███▋      | 1835/5000 [23:09:17<31:37:07, 35.96s/it]
                                                         
{'loss': '0.7323', 'grad_norm': '17.05', 'learning_rate': '1.292e-05', 'epoch': '0.367'}

 37%|███▋      | 1835/5000 [23:09:17<31:37:07, 35.96s/it][2026-05-22 15:15:05,713] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:15:05,804] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 37%|███▋      | 1836/5000 [23:10:05<34:46:26, 39.57s/it]
                                                         
{'loss': '0.7095', 'grad_norm': '16.41', 'learning_rate': '1.291e-05', 'epoch': '0.3672'}

 37%|███▋      | 1836/5000 [23:10:05<34:46:26, 39.57s/it][2026-05-22 15:15:53,035] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 15:15:53,711] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 501])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1837/5000 [23:10:55<37:24:55, 42.58s/it]
                                                         
{'loss': '0.6446', 'grad_norm': '15.37', 'learning_rate': '1.291e-05', 'epoch': '0.3674'}

 37%|███▋      | 1837/5000 [23:10:55<37:24:55, 42.58s/it][2026-05-22 15:16:42,046] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 264])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 15:16:43,327] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 522])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1838/5000 [23:11:46<39:38:08, 45.13s/it]
                                                         
{'loss': '0.2582', 'grad_norm': '13.86', 'learning_rate': '1.291e-05', 'epoch': '0.3676'}

 37%|███▋      | 1838/5000 [23:11:46<39:38:08, 45.13s/it][2026-05-22 15:17:32,998] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 15:17:34,576] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 313])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 37%|███▋      | 1839/5000 [23:12:32<39:59:48, 45.55s/it]
                                                         
{'loss': '0.5976', 'grad_norm': '16.17', 'learning_rate': '1.29e-05', 'epoch': '0.3678'}

 37%|███▋      | 1839/5000 [23:12:32<39:59:48, 45.55s/it][2026-05-22 15:18:20,960] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 313])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 15:18:21,133] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 320])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 37%|███▋      | 1840/5000 [23:13:17<39:51:16, 45.40s/it]
                                                         
{'loss': '0.4372', 'grad_norm': '12.04', 'learning_rate': '1.29e-05', 'epoch': '0.368'}

 37%|███▋      | 1840/5000 [23:13:17<39:51:16, 45.40s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:19:04,496] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 15:19:04,854] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 37%|███▋      | 1841/5000 [23:13:48<36:03:08, 41.09s/it]
                                                         
{'loss': '0.6802', 'grad_norm': '68.4', 'learning_rate': '1.289e-05', 'epoch': '0.3682'}

 37%|███▋      | 1841/5000 [23:13:48<36:03:08, 41.09s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:19:35,441] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 243])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:19:35,546] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 497])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1842/5000 [23:14:21<33:43:43, 38.45s/it]
                                                         
{'loss': '0.2237', 'grad_norm': '44.5', 'learning_rate': '1.289e-05', 'epoch': '0.3684'}

 37%|███▋      | 1842/5000 [23:14:21<33:43:43, 38.45s/it][2026-05-22 15:20:07,936] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 424])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-22 15:20:09,418] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 480])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 37%|███▋      | 1843/5000 [23:14:53<32:09:18, 36.67s/it]
                                                         
{'loss': '0.1675', 'grad_norm': '5.406', 'learning_rate': '1.289e-05', 'epoch': '0.3686'}

 37%|███▋      | 1843/5000 [23:14:53<32:09:18, 36.67s/it][2026-05-22 15:20:41,024] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 176])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])
[2026-05-22 15:20:41,961] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 475])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 37%|███▋      | 1844/5000 [23:15:28<31:32:12, 35.97s/it]
                                                         
{'loss': '0.04565', 'grad_norm': '4.393', 'learning_rate': '1.288e-05', 'epoch': '0.3688'}

 37%|███▋      | 1844/5000 [23:15:28<31:32:12, 35.97s/it][2026-05-22 15:21:14,892] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 955])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 15:21:15,000] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 990])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 168])

 37%|███▋      | 1845/5000 [23:15:57<29:45:19, 33.95s/it]
                                                         
{'loss': '0.3407', 'grad_norm': '6.729', 'learning_rate': '1.288e-05', 'epoch': '0.369'}

 37%|███▋      | 1845/5000 [23:15:57<29:45:19, 33.95s/it][2026-05-22 15:21:43,624] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 132])
[2026-05-22 15:21:44,272] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 37%|███▋      | 1846/5000 [23:16:26<28:31:56, 32.57s/it]
                                                         
{'loss': '0.05843', 'grad_norm': '4.086', 'learning_rate': '1.287e-05', 'epoch': '0.3692'}

 37%|███▋      | 1846/5000 [23:16:26<28:31:56, 32.57s/it][2026-05-22 15:22:14,486] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 64])
[2026-05-22 15:22:14,933] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 94])

 37%|███▋      | 1847/5000 [23:16:59<28:30:39, 32.55s/it]
                                                         
{'loss': '0.2166', 'grad_norm': '7.019', 'learning_rate': '1.287e-05', 'epoch': '0.3694'}

 37%|███▋      | 1847/5000 [23:16:59<28:30:39, 32.55s/it][2026-05-22 15:22:47,222] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 50])
[2026-05-22 15:22:47,909] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 81])

 37%|███▋      | 1848/5000 [23:17:31<28:32:14, 32.59s/it]
                                                         
{'loss': '0.06138', 'grad_norm': '6.477', 'learning_rate': '1.287e-05', 'epoch': '0.3696'}

 37%|███▋      | 1848/5000 [23:17:31<28:32:14, 32.59s/it][2026-05-22 15:23:19,241] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 277])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:23:20,291] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 432])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 37%|███▋      | 1849/5000 [23:18:01<27:37:28, 31.56s/it]
                                                         
{'loss': '0.1638', 'grad_norm': '6.592', 'learning_rate': '1.286e-05', 'epoch': '0.3698'}

 37%|███▋      | 1849/5000 [23:18:01<27:37:28, 31.56s/it][2026-05-22 15:23:49,144] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 289])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 15:23:49,391] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 498])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1850/5000 [23:18:30<26:58:48, 30.83s/it]
                                                         
{'loss': '0.1521', 'grad_norm': '6.084', 'learning_rate': '1.286e-05', 'epoch': '0.37'}

 37%|███▋      | 1850/5000 [23:18:30<26:58:48, 30.83s/it][2026-05-22 15:24:16,058] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 115])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])
[2026-05-22 15:24:16,083] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 176])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 37%|███▋      | 1851/5000 [23:19:13<30:17:52, 34.64s/it]
                                                         
{'loss': '1.025', 'grad_norm': '21.41', 'learning_rate': '1.285e-05', 'epoch': '0.3702'}

 37%|███▋      | 1851/5000 [23:19:13<30:17:52, 34.64s/it][2026-05-22 15:24:59,563] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 157])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 15:24:59,623] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 54])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 37%|███▋      | 1852/5000 [23:19:57<32:36:20, 37.29s/it]
                                                         
{'loss': '1.077', 'grad_norm': '19.95', 'learning_rate': '1.285e-05', 'epoch': '0.3704'}

 37%|███▋      | 1852/5000 [23:19:57<32:36:20, 37.29s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:25:44,239] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 298])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 15:25:44,742] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 304])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1853/5000 [23:20:41<34:31:13, 39.49s/it]
                                                         
{'loss': '2.335', 'grad_norm': '7.482', 'learning_rate': '1.284e-05', 'epoch': '0.3706'}

 37%|███▋      | 1853/5000 [23:20:41<34:31:13, 39.49s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:26:29,366] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 277])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:26:29,452] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 279])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 37%|███▋      | 1854/5000 [23:21:26<35:57:12, 41.14s/it]
                                                         
{'loss': '2.385', 'grad_norm': '6.961', 'learning_rate': '1.284e-05', 'epoch': '0.3708'}

 37%|███▋      | 1854/5000 [23:21:26<35:57:12, 41.14s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:27:13,642] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 203])
[2026-05-22 15:27:13,645] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 233])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 237])

 37%|███▋      | 1855/5000 [23:21:58<33:27:30, 38.30s/it]
                                                         
{'loss': '0.3384', 'grad_norm': '10.44', 'learning_rate': '1.284e-05', 'epoch': '0.371'}

 37%|███▋      | 1855/5000 [23:21:58<33:27:30, 38.30s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:27:44,783] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 167])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 240])
[2026-05-22 15:27:45,407] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 463])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 260])

 37%|███▋      | 1856/5000 [23:22:31<32:00:10, 36.64s/it]
                                                         
{'loss': '0.03583', 'grad_norm': '3.551', 'learning_rate': '1.283e-05', 'epoch': '0.3712'}

 37%|███▋      | 1856/5000 [23:22:31<32:00:10, 36.64s/it][2026-05-22 15:28:18,086] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 710])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 15:28:18,739] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 502])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1857/5000 [23:22:54<28:34:57, 32.74s/it]
                                                         
{'loss': '1.653', 'grad_norm': '25.51', 'learning_rate': '1.283e-05', 'epoch': '0.3714'}

 37%|███▋      | 1857/5000 [23:22:54<28:34:57, 32.74s/it][2026-05-22 15:28:43,243] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 791])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:28:43,580] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1858/5000 [23:23:18<26:09:32, 29.97s/it]
                                                         
{'loss': '1.258', 'grad_norm': '23.42', 'learning_rate': '1.282e-05', 'epoch': '0.3716'}

 37%|███▋      | 1858/5000 [23:23:18<26:09:32, 29.97s/it][2026-05-22 15:29:05,496] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 251])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 15:29:06,071] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 301])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1859/5000 [23:23:46<25:36:11, 29.34s/it]
                                                         
{'loss': '0.5215', 'grad_norm': '11.22', 'learning_rate': '1.282e-05', 'epoch': '0.3718'}

 37%|███▋      | 1859/5000 [23:23:46<25:36:11, 29.34s/it][2026-05-22 15:29:33,229] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 251])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:29:33,251] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 268])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 37%|███▋      | 1860/5000 [23:24:15<25:40:25, 29.43s/it]
                                                         
{'loss': '0.2481', 'grad_norm': '9.711', 'learning_rate': '1.282e-05', 'epoch': '0.372'}

 37%|███▋      | 1860/5000 [23:24:15<25:40:25, 29.43s/it][2026-05-22 15:30:02,680] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 425])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 100])
[2026-05-22 15:30:02,684] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])

 37%|███▋      | 1861/5000 [23:24:44<25:30:23, 29.25s/it]
                                                         
{'loss': '0.05473', 'grad_norm': '5.967', 'learning_rate': '1.281e-05', 'epoch': '0.3722'}

 37%|███▋      | 1861/5000 [23:24:44<25:30:23, 29.25s/it][2026-05-22 15:30:32,017] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 97])
[2026-05-22 15:30:32,116] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 90])

 37%|███▋      | 1862/5000 [23:25:13<25:24:50, 29.16s/it]
                                                         
{'loss': '0.0825', 'grad_norm': '6.618', 'learning_rate': '1.281e-05', 'epoch': '0.3724'}

 37%|███▋      | 1862/5000 [23:25:13<25:24:50, 29.16s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:31:00,742] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 393])
[2026-05-22 15:31:01,814] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 501])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 227])

 37%|███▋      | 1863/5000 [23:25:51<27:33:19, 31.62s/it]
                                                         
{'loss': '0.887', 'grad_norm': '12.35', 'learning_rate': '1.28e-05', 'epoch': '0.3726'}

 37%|███▋      | 1863/5000 [23:25:51<27:33:19, 31.62s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:31:37,457] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 201])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:31:38,943] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])

 37%|███▋      | 1864/5000 [23:26:32<30:14:35, 34.72s/it]
                                                         
{'loss': '1.061', 'grad_norm': '17.21', 'learning_rate': '1.28e-05', 'epoch': '0.3728'}

 37%|███▋      | 1864/5000 [23:26:32<30:14:35, 34.72s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:32:19,030] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 209])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])
[2026-05-22 15:32:19,075] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 266])

 37%|███▋      | 1865/5000 [23:27:00<28:16:30, 32.47s/it]
                                                         
{'loss': '0.3747', 'grad_norm': '7.031', 'learning_rate': '1.28e-05', 'epoch': '0.373'}

 37%|███▋      | 1865/5000 [23:27:00<28:16:30, 32.47s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:32:46,312] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])
[2026-05-22 15:32:46,834] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])

 37%|███▋      | 1866/5000 [23:27:29<27:19:46, 31.39s/it]
                                                         
{'loss': '0.1688', 'grad_norm': '5.576', 'learning_rate': '1.279e-05', 'epoch': '0.3732'}

 37%|███▋      | 1866/5000 [23:27:29<27:19:46, 31.39s/it][2026-05-22 15:33:16,282] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])
[2026-05-22 15:33:17,017] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 303])

 37%|███▋      | 1867/5000 [23:27:57<26:25:31, 30.36s/it]
                                                         
{'loss': '0.8917', 'grad_norm': '17.49', 'learning_rate': '1.279e-05', 'epoch': '0.3734'}

 37%|███▋      | 1867/5000 [23:27:57<26:25:31, 30.36s/it][2026-05-22 15:33:45,067] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 431])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 310])
[2026-05-22 15:33:45,161] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 303])

 37%|███▋      | 1868/5000 [23:28:25<26:00:09, 29.89s/it]
                                                         
{'loss': '1.959', 'grad_norm': '30.61', 'learning_rate': '1.278e-05', 'epoch': '0.3736'}

 37%|███▋      | 1868/5000 [23:28:25<26:00:09, 29.89s/it][2026-05-22 15:34:12,715] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 231])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:34:13,398] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1869/5000 [23:28:57<26:20:55, 30.30s/it]
                                                         
{'loss': '0.05587', 'grad_norm': '4.351', 'learning_rate': '1.278e-05', 'epoch': '0.3738'}

 37%|███▋      | 1869/5000 [23:28:57<26:20:55, 30.30s/it][2026-05-22 15:34:44,653] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:34:44,702] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 37%|███▋      | 1870/5000 [23:29:27<26:14:50, 30.19s/it]
                                                         
{'loss': '0.03724', 'grad_norm': '3.673', 'learning_rate': '1.278e-05', 'epoch': '0.374'}

 37%|███▋      | 1870/5000 [23:29:27<26:14:50, 30.19s/it][2026-05-22 15:35:13,645] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 596])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])
[2026-05-22 15:35:13,661] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 970])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

 37%|███▋      | 1871/5000 [23:29:53<25:23:53, 29.22s/it]
                                                         
{'loss': '0.2773', 'grad_norm': '5.951', 'learning_rate': '1.277e-05', 'epoch': '0.3742'}

 37%|███▋      | 1871/5000 [23:29:53<25:23:53, 29.22s/it][2026-05-22 15:35:40,612] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 433])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 183])
[2026-05-22 15:35:40,677] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 37%|███▋      | 1872/5000 [23:30:21<24:52:00, 28.62s/it]
                                                         
{'loss': '0.2746', 'grad_norm': '4.752', 'learning_rate': '1.277e-05', 'epoch': '0.3744'}

 37%|███▋      | 1872/5000 [23:30:21<24:52:00, 28.62s/it][2026-05-22 15:36:07,796] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 53])
[2026-05-22 15:36:08,981] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 953])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 95])

 37%|███▋      | 1873/5000 [23:30:50<25:09:57, 28.97s/it]
                                                         
{'loss': '0.3719', 'grad_norm': '6.608', 'learning_rate': '1.276e-05', 'epoch': '0.3746'}

 37%|███▋      | 1873/5000 [23:30:50<25:09:57, 28.97s/it][2026-05-22 15:36:37,641] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 483])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 59])
[2026-05-22 15:36:38,691] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 72])

 37%|███▋      | 1874/5000 [23:31:20<25:24:09, 29.25s/it]
                                                         
{'loss': '0.06365', 'grad_norm': '3.419', 'learning_rate': '1.276e-05', 'epoch': '0.3748'}

 37%|███▋      | 1874/5000 [23:31:20<25:24:09, 29.25s/it][2026-05-22 15:37:07,816] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 291])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 185])
[2026-05-22 15:37:08,210] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 232])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 38%|███▊      | 1875/5000 [23:31:52<25:57:22, 29.90s/it]
                                                         
{'loss': '0.4981', 'grad_norm': '8.097', 'learning_rate': '1.276e-05', 'epoch': '0.375'}

 38%|███▊      | 1875/5000 [23:31:52<25:57:22, 29.90s/it][2026-05-22 15:37:39,559] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 280])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:37:39,573] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 293])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 38%|███▊      | 1876/5000 [23:32:22<25:58:11, 29.93s/it]
                                                         
{'loss': '0.04353', 'grad_norm': '3.665', 'learning_rate': '1.275e-05', 'epoch': '0.3752'}

 38%|███▊      | 1876/5000 [23:32:22<25:58:11, 29.93s/it][2026-05-22 15:38:13,047] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:38:13,248] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 38%|███▊      | 1877/5000 [23:33:08<30:05:53, 34.70s/it]
                                                         
{'loss': '0.3645', 'grad_norm': '9.45', 'learning_rate': '1.275e-05', 'epoch': '0.3754'}

 38%|███▊      | 1877/5000 [23:33:08<30:05:53, 34.70s/it][2026-05-22 15:38:59,937] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:39:00,282] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 38%|███▊      | 1878/5000 [23:33:53<32:46:19, 37.79s/it]
                                                         
{'loss': '0.5052', 'grad_norm': '12.51', 'learning_rate': '1.274e-05', 'epoch': '0.3756'}

 38%|███▊      | 1878/5000 [23:33:53<32:46:19, 37.79s/it][2026-05-22 15:39:40,042] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 267])
[2026-05-22 15:39:40,517] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])

 38%|███▊      | 1879/5000 [23:34:34<33:44:30, 38.92s/it]
                                                         
{'loss': '0.1719', 'grad_norm': '8.829', 'learning_rate': '1.274e-05', 'epoch': '0.3758'}

 38%|███▊      | 1879/5000 [23:34:34<33:44:30, 38.92s/it][2026-05-22 15:40:21,895] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 290])
[2026-05-22 15:40:22,073] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 279])

 38%|███▊      | 1880/5000 [23:35:10<33:01:53, 38.11s/it]
                                                         
{'loss': '0.1711', 'grad_norm': '7.805', 'learning_rate': '1.273e-05', 'epoch': '0.376'}

 38%|███▊      | 1880/5000 [23:35:10<33:01:53, 38.11s/it][2026-05-22 15:40:56,896] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 15:40:56,962] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 233])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 38%|███▊      | 1881/5000 [23:35:39<30:29:00, 35.18s/it]
                                                         
{'loss': '0.9967', 'grad_norm': '19.38', 'learning_rate': '1.273e-05', 'epoch': '0.3762'}

 38%|███▊      | 1881/5000 [23:35:39<30:29:00, 35.18s/it][2026-05-22 15:41:25,277] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 206])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:41:25,340] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 269])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 38%|███▊      | 1882/5000 [23:36:07<28:47:53, 33.25s/it]
                                                         
{'loss': '0.7396', 'grad_norm': '17.79', 'learning_rate': '1.273e-05', 'epoch': '0.3764'}

 38%|███▊      | 1882/5000 [23:36:08<28:47:53, 33.25s/it][2026-05-22 15:41:55,371] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 229])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 15:41:56,532] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 459])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])

 38%|███▊      | 1883/5000 [23:36:58<33:14:25, 38.39s/it]
                                                         
{'loss': '0.1932', 'grad_norm': '7.396', 'learning_rate': '1.272e-05', 'epoch': '0.3766'}

 38%|███▊      | 1883/5000 [23:36:58<33:14:25, 38.39s/it][2026-05-22 15:42:46,316] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 307])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 15:42:47,290] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 38%|███▊      | 1884/5000 [23:37:51<37:09:17, 42.93s/it]
                                                         
{'loss': '0.263', 'grad_norm': '17.51', 'learning_rate': '1.272e-05', 'epoch': '0.3768'}

 38%|███▊      | 1884/5000 [23:37:51<37:09:17, 42.93s/it][2026-05-22 15:43:38,582] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 323])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 63])
[2026-05-22 15:43:40,748] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 541])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])

 38%|███▊      | 1885/5000 [23:38:21<33:46:38, 39.04s/it]
                                                         
{'loss': '0.6669', 'grad_norm': '14.15', 'learning_rate': '1.271e-05', 'epoch': '0.377'}

 38%|███▊      | 1885/5000 [23:38:21<33:46:38, 39.04s/it][2026-05-22 15:44:08,522] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 200])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])
[2026-05-22 15:44:10,060] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 496])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 91])

 38%|███▊      | 1886/5000 [23:38:52<31:33:29, 36.48s/it]
                                                         
{'loss': '0.266', 'grad_norm': '10.81', 'learning_rate': '1.271e-05', 'epoch': '0.3772'}

 38%|███▊      | 1886/5000 [23:38:52<31:33:29, 36.48s/it][2026-05-22 15:44:38,798] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 294])
[2026-05-22 15:44:39,067] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])

 38%|███▊      | 1887/5000 [23:39:19<29:00:00, 33.54s/it]
                                                         
{'loss': '1.143', 'grad_norm': '23.69', 'learning_rate': '1.271e-05', 'epoch': '0.3774'}

 38%|███▊      | 1887/5000 [23:39:19<29:00:00, 33.54s/it][2026-05-22 15:45:06,364] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 412])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 267])
[2026-05-22 15:45:06,368] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])

 38%|███▊      | 1888/5000 [23:39:46<27:20:23, 31.63s/it]
                                                         
{'loss': '1.111', 'grad_norm': '21.91', 'learning_rate': '1.27e-05', 'epoch': '0.3776'}

 38%|███▊      | 1888/5000 [23:39:46<27:20:23, 31.63s/it][2026-05-22 15:45:33,942] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 288])
[2026-05-22 15:45:34,467] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 967])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 291])

 38%|███▊      | 1889/5000 [23:40:19<27:52:20, 32.25s/it]
                                                         
{'loss': '0.2315', 'grad_norm': '7.265', 'learning_rate': '1.27e-05', 'epoch': '0.3778'}

 38%|███▊      | 1889/5000 [23:40:19<27:52:20, 32.25s/it][2026-05-22 15:46:07,543] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 258])
[2026-05-22 15:46:07,634] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 257])

 38%|███▊      | 1890/5000 [23:40:52<27:54:15, 32.30s/it]
                                                         
{'loss': '0.01529', 'grad_norm': '2.373', 'learning_rate': '1.269e-05', 'epoch': '0.378'}

 38%|███▊      | 1890/5000 [23:40:52<27:54:15, 32.30s/it][2026-05-22 15:46:42,973] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 95])
[2026-05-22 15:46:42,987] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 83])

 38%|███▊      | 1891/5000 [23:41:22<27:25:18, 31.75s/it]
                                                         
{'loss': '0.08578', 'grad_norm': '5.9', 'learning_rate': '1.269e-05', 'epoch': '0.3782'}

 38%|███▊      | 1891/5000 [23:41:22<27:25:18, 31.75s/it][2026-05-22 15:47:13,962] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 91])
[2026-05-22 15:47:14,238] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 33])

 38%|███▊      | 1892/5000 [23:41:52<26:58:21, 31.24s/it]
                                                         
{'loss': '0.05649', 'grad_norm': '5.421', 'learning_rate': '1.269e-05', 'epoch': '0.3784'}

 38%|███▊      | 1892/5000 [23:41:52<26:58:21, 31.24s/it][2026-05-22 15:47:39,440] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 191])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 170])
[2026-05-22 15:47:39,507] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 412])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 179])

 38%|███▊      | 1893/5000 [23:42:19<25:48:34, 29.90s/it]
                                                         
{'loss': '0.6158', 'grad_norm': '10.18', 'learning_rate': '1.268e-05', 'epoch': '0.3786'}

 38%|███▊      | 1893/5000 [23:42:19<25:48:34, 29.90s/it][2026-05-22 15:48:05,691] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 258])
[2026-05-22 15:48:05,725] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 259])

 38%|███▊      | 1894/5000 [23:42:48<25:26:17, 29.48s/it]
                                                         
{'loss': '0.2917', 'grad_norm': '8.532', 'learning_rate': '1.268e-05', 'epoch': '0.3788'}

 38%|███▊      | 1894/5000 [23:42:48<25:26:17, 29.48s/it][2026-05-22 15:48:34,896] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 409])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 221])
[2026-05-22 15:48:35,943] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 456])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 452])

 38%|███▊      | 1895/5000 [23:43:33<29:38:30, 34.37s/it]
                                                         
{'loss': '0.4032', 'grad_norm': '6.909', 'learning_rate': '1.267e-05', 'epoch': '0.379'}

 38%|███▊      | 1895/5000 [23:43:33<29:38:30, 34.37s/it][2026-05-22 15:49:21,454] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 338])
[2026-05-22 15:49:22,243] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 437])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 304])

 38%|███▊      | 1896/5000 [23:44:20<32:41:00, 37.91s/it]
                                                         
{'loss': '0.06298', 'grad_norm': '3.736', 'learning_rate': '1.267e-05', 'epoch': '0.3792'}

 38%|███▊      | 1896/5000 [23:44:20<32:41:00, 37.91s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:50:07,356] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 417])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 281])
[2026-05-22 15:50:08,179] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 299])

 38%|███▊      | 1897/5000 [23:45:06<34:53:51, 40.49s/it]
                                                         
{'loss': '0.7585', 'grad_norm': '19.73', 'learning_rate': '1.267e-05', 'epoch': '0.3794'}

 38%|███▊      | 1897/5000 [23:45:06<34:53:51, 40.49s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:50:53,724] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 283])
[2026-05-22 15:50:54,021] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 501])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 270])

 38%|███▊      | 1898/5000 [23:45:52<36:18:25, 42.14s/it]
                                                         
{'loss': '0.6337', 'grad_norm': '21.88', 'learning_rate': '1.266e-05', 'epoch': '0.3796'}

 38%|███▊      | 1898/5000 [23:45:52<36:18:25, 42.14s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:51:38,922] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 318])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 15:51:38,977] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])

 38%|███▊      | 1899/5000 [23:46:18<32:04:01, 37.23s/it]
                                                         
{'loss': '1.006', 'grad_norm': '20.74', 'learning_rate': '1.266e-05', 'epoch': '0.3798'}

 38%|███▊      | 1899/5000 [23:46:18<32:04:01, 37.23s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 15:52:05,341] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 400])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 234])
[2026-05-22 15:52:05,384] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 398])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 302])

 38%|███▊      | 1900/5000 [23:46:45<29:28:43, 34.23s/it]
                                                         
{'loss': '0.9267', 'grad_norm': '19.66', 'learning_rate': '1.265e-05', 'epoch': '0.38'}

 38%|███▊      | 1900/5000 [23:46:45<29:28:43, 34.23s/it][2026-05-22 15:52:32,570] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 15:52:32,724] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 710])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 289])

 38%|███▊      | 1901/5000 [23:47:18<29:01:16, 33.71s/it]
                                                         
{'loss': '0.3033', 'grad_norm': '7.272', 'learning_rate': '1.265e-05', 'epoch': '0.3802'}

 38%|███▊      | 1901/5000 [23:47:18<29:01:16, 33.71s/it][2026-05-22 15:53:04,984] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 792])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])
[2026-05-22 15:53:04,986] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 445])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 38%|███▊      | 1902/5000 [23:47:46<27:44:12, 32.23s/it]
                                                         
{'loss': '0.03563', 'grad_norm': '3.769', 'learning_rate': '1.264e-05', 'epoch': '0.3804'}

 38%|███▊      | 1902/5000 [23:47:46<27:44:12, 32.23s/it][2026-05-22 15:53:33,670] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 822])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])
[2026-05-22 15:53:33,974] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 458])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 95])

 38%|███▊      | 1903/5000 [23:48:16<27:02:06, 31.43s/it]
                                                         
{'loss': '0.253', 'grad_norm': '7.141', 'learning_rate': '1.264e-05', 'epoch': '0.3806'}

 38%|███▊      | 1903/5000 [23:48:16<27:02:06, 31.43s/it][2026-05-22 15:54:04,015] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])
[2026-05-22 15:54:04,248] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 97])

 38%|███▊      | 1904/5000 [23:48:46<26:48:17, 31.17s/it]
                                                         
{'loss': '0.03651', 'grad_norm': '3.586', 'learning_rate': '1.264e-05', 'epoch': '0.3808'}

 38%|███▊      | 1904/5000 [23:48:46<26:48:17, 31.17s/it][2026-05-22 15:54:33,847] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 15:54:34,305] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 38%|███▊      | 1905/5000 [23:49:14<25:48:49, 30.03s/it]
                                                         
{'loss': '0.4755', 'grad_norm': '14.43', 'learning_rate': '1.263e-05', 'epoch': '0.381'}

 38%|███▊      | 1905/5000 [23:49:14<25:48:49, 30.03s/it][2026-05-22 15:55:01,101] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:55:01,604] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 442])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 38%|███▊      | 1906/5000 [23:49:41<25:07:51, 29.24s/it]
                                                         
{'loss': '0.8811', 'grad_norm': '24.84', 'learning_rate': '1.263e-05', 'epoch': '0.3812'}

 38%|███▊      | 1906/5000 [23:49:41<25:07:51, 29.24s/it][2026-05-22 15:55:28,926] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 650])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 15:55:29,383] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 647])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 38%|███▊      | 1907/5000 [23:50:17<26:49:39, 31.23s/it]
                                                         
{'loss': '0.6067', 'grad_norm': '13.92', 'learning_rate': '1.262e-05', 'epoch': '0.3814'}

 38%|███▊      | 1907/5000 [23:50:17<26:49:39, 31.23s/it][2026-05-22 15:56:05,429] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 630])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:56:05,478] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 634])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 38%|███▊      | 1908/5000 [23:50:56<28:43:36, 33.45s/it]
                                                         
{'loss': '0.7177', 'grad_norm': '13.18', 'learning_rate': '1.262e-05', 'epoch': '0.3816'}

 38%|███▊      | 1908/5000 [23:50:56<28:43:36, 33.45s/it][2026-05-22 15:56:43,840] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 634])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 15:56:43,900] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 629])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 38%|███▊      | 1909/5000 [23:51:33<29:47:52, 34.70s/it]
                                                         
{'loss': '0.7937', 'grad_norm': '16.87', 'learning_rate': '1.262e-05', 'epoch': '0.3818'}

 38%|███▊      | 1909/5000 [23:51:33<29:47:52, 34.70s/it][2026-05-22 15:57:21,578] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 628])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])
[2026-05-22 15:57:21,579] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 649])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 38%|███▊      | 1910/5000 [23:52:12<30:44:06, 35.81s/it]
                                                         
{'loss': '1.025', 'grad_norm': '18.5', 'learning_rate': '1.261e-05', 'epoch': '0.382'}

 38%|███▊      | 1910/5000 [23:52:12<30:44:06, 35.81s/it][2026-05-22 15:57:58,595] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 219])
[2026-05-22 15:57:59,134] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 513])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 344])

 38%|███▊      | 1911/5000 [23:52:41<29:08:24, 33.96s/it]
                                                         
{'loss': '0.2332', 'grad_norm': '7.469', 'learning_rate': '1.261e-05', 'epoch': '0.3822'}

 38%|███▊      | 1911/5000 [23:52:41<29:08:24, 33.96s/it][2026-05-22 15:58:28,714] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 267])
[2026-05-22 15:58:28,780] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 490])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 297])

 38%|███▊      | 1912/5000 [23:53:12<28:09:15, 32.82s/it]
                                                         
{'loss': '0.3014', 'grad_norm': '7.713', 'learning_rate': '1.26e-05', 'epoch': '0.3824'}

 38%|███▊      | 1912/5000 [23:53:12<28:09:15, 32.82s/it][2026-05-22 15:58:59,269] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 235])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 145])
[2026-05-22 15:59:00,444] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 502])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 38%|███▊      | 1913/5000 [23:53:49<29:16:33, 34.14s/it]
                                                         
{'loss': '1.276', 'grad_norm': '10.64', 'learning_rate': '1.26e-05', 'epoch': '0.3826'}

 38%|███▊      | 1913/5000 [23:53:49<29:16:33, 34.14s/it][2026-05-22 15:59:36,435] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 203])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 15:59:37,337] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 38%|███▊      | 1914/5000 [23:54:25<29:44:05, 34.69s/it]
                                                         
{'loss': '0.9877', 'grad_norm': '10.16', 'learning_rate': '1.26e-05', 'epoch': '0.3828'}

 38%|███▊      | 1914/5000 [23:54:25<29:44:05, 34.69s/it][2026-05-22 16:00:12,097] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 254])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 16:00:12,879] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 286])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 38%|███▊      | 1915/5000 [23:55:01<30:01:46, 35.04s/it]
                                                         
{'loss': '0.9414', 'grad_norm': '11.52', 'learning_rate': '1.259e-05', 'epoch': '0.383'}

 38%|███▊      | 1915/5000 [23:55:01<30:01:46, 35.04s/it][2026-05-22 16:00:48,768] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 358])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 16:00:49,010] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 311])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 38%|███▊      | 1916/5000 [23:55:38<30:35:53, 35.72s/it]
                                                         
{'loss': '0.9796', 'grad_norm': '12.42', 'learning_rate': '1.259e-05', 'epoch': '0.3832'}

 38%|███▊      | 1916/5000 [23:55:38<30:35:53, 35.72s/it][2026-05-22 16:01:24,991] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 346])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 16:01:25,202] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 253])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 38%|███▊      | 1917/5000 [23:56:08<29:15:51, 34.17s/it]
                                                         
{'loss': '0.6039', 'grad_norm': '13.61', 'learning_rate': '1.258e-05', 'epoch': '0.3834'}

 38%|███▊      | 1917/5000 [23:56:08<29:15:51, 34.17s/it][2026-05-22 16:01:56,150] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 304])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 16:01:56,198] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 260])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 38%|███▊      | 1918/5000 [23:56:38<27:58:25, 32.68s/it]
                                                         
{'loss': '0.3366', 'grad_norm': '11.61', 'learning_rate': '1.258e-05', 'epoch': '0.3836'}

 38%|███▊      | 1918/5000 [23:56:38<27:58:25, 32.68s/it][2026-05-22 16:02:24,985] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 272])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 275])
[2026-05-22 16:02:26,153] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 38%|███▊      | 1919/5000 [23:57:10<27:45:34, 32.44s/it]
                                                         
{'loss': '0.3637', 'grad_norm': '12.21', 'learning_rate': '1.258e-05', 'epoch': '0.3838'}

 38%|███▊      | 1919/5000 [23:57:10<27:45:34, 32.44s/it][2026-05-22 16:02:57,538] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:02:57,550] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 38%|███▊      | 1920/5000 [23:57:42<27:44:58, 32.43s/it]
                                                         
{'loss': '0.2718', 'grad_norm': '11.4', 'learning_rate': '1.257e-05', 'epoch': '0.384'}

 38%|███▊      | 1920/5000 [23:57:42<27:44:58, 32.43s/it][2026-05-22 16:03:29,208] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])
[2026-05-22 16:03:29,801] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 789])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 232])

 38%|███▊      | 1921/5000 [23:58:11<26:59:09, 31.55s/it]
                                                         
{'loss': '0.4885', 'grad_norm': '7.808', 'learning_rate': '1.257e-05', 'epoch': '0.3842'}

 38%|███▊      | 1921/5000 [23:58:11<26:59:09, 31.55s/it][2026-05-22 16:04:00,516] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])
[2026-05-22 16:04:00,745] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 855])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 302])

 38%|███▊      | 1922/5000 [23:58:43<27:05:27, 31.69s/it]
                                                         
{'loss': '0.0299', 'grad_norm': '3.05', 'learning_rate': '1.256e-05', 'epoch': '0.3844'}

 38%|███▊      | 1922/5000 [23:58:43<27:05:27, 31.69s/it][2026-05-22 16:04:31,312] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])
[2026-05-22 16:04:32,296] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 497])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 276])

 38%|███▊      | 1923/5000 [23:59:12<26:23:13, 30.87s/it]
                                                         
{'loss': '0.4245', 'grad_norm': '8.834', 'learning_rate': '1.256e-05', 'epoch': '0.3846'}

 38%|███▊      | 1923/5000 [23:59:12<26:23:13, 30.87s/it][2026-05-22 16:05:00,319] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 348])
[2026-05-22 16:05:00,962] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 498])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 301])

 38%|███▊      | 1924/5000 [23:59:41<25:53:24, 30.30s/it]
                                                         
{'loss': '0.2859', 'grad_norm': '7.468', 'learning_rate': '1.256e-05', 'epoch': '0.3848'}

 38%|███▊      | 1924/5000 [23:59:41<25:53:24, 30.30s/it][2026-05-22 16:05:28,596] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 252])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 244])
[2026-05-22 16:05:29,683] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 482])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 38%|███▊      | 1925/5000 [24:00:12<25:57:04, 30.38s/it]
                                                         
{'loss': '0.3447', 'grad_norm': '7.267', 'learning_rate': '1.255e-05', 'epoch': '0.385'}

 38%|███▊      | 1925/5000 [24:00:12<25:57:04, 30.38s/it][2026-05-22 16:06:00,421] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:06:00,789] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▊      | 1926/5000 [24:00:44<26:24:14, 30.92s/it]
                                                         
{'loss': '0.03567', 'grad_norm': '3.633', 'learning_rate': '1.255e-05', 'epoch': '0.3852'}

 39%|███▊      | 1926/5000 [24:00:44<26:24:14, 30.92s/it][2026-05-22 16:06:31,877] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 412])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 256])
[2026-05-22 16:06:32,645] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▊      | 1927/5000 [24:01:22<28:11:11, 33.02s/it]
                                                         
{'loss': '1.099', 'grad_norm': '16.03', 'learning_rate': '1.254e-05', 'epoch': '0.3854'}

 39%|███▊      | 1927/5000 [24:01:22<28:11:11, 33.02s/it][2026-05-22 16:07:10,034] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 16:07:10,060] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 39%|███▊      | 1928/5000 [24:02:05<30:39:47, 35.93s/it]
                                                         
{'loss': '1.284', 'grad_norm': '19.05', 'learning_rate': '1.254e-05', 'epoch': '0.3856'}

 39%|███▊      | 1928/5000 [24:02:05<30:39:47, 35.93s/it][2026-05-22 16:07:51,983] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 254])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])
[2026-05-22 16:07:52,743] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 452])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 39%|███▊      | 1929/5000 [24:02:34<29:02:36, 34.05s/it]
                                                         
{'loss': '0.2472', 'grad_norm': '6.405', 'learning_rate': '1.253e-05', 'epoch': '0.3858'}

 39%|███▊      | 1929/5000 [24:02:34<29:02:36, 34.05s/it][2026-05-22 16:08:23,231] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:08:23,302] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 39%|███▊      | 1930/5000 [24:03:05<28:15:14, 33.13s/it]
                                                         
{'loss': '0.04783', 'grad_norm': '3.902', 'learning_rate': '1.253e-05', 'epoch': '0.386'}

 39%|███▊      | 1930/5000 [24:03:05<28:15:14, 33.13s/it][2026-05-22 16:08:53,307] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 279])
[2026-05-22 16:08:53,838] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])

 39%|███▊      | 1931/5000 [24:03:38<27:58:33, 32.82s/it]
                                                         
{'loss': '0.3991', 'grad_norm': '8.203', 'learning_rate': '1.253e-05', 'epoch': '0.3862'}

 39%|███▊      | 1931/5000 [24:03:38<27:58:33, 32.82s/it][2026-05-22 16:09:25,251] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 471])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 313])
[2026-05-22 16:09:25,923] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 453])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 279])

 39%|███▊      | 1932/5000 [24:04:10<27:51:03, 32.68s/it]
                                                         
{'loss': '0.01355', 'grad_norm': '2.11', 'learning_rate': '1.252e-05', 'epoch': '0.3864'}

 39%|███▊      | 1932/5000 [24:04:10<27:51:03, 32.68s/it][2026-05-22 16:09:56,885] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 576])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 189])
[2026-05-22 16:09:57,539] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 655])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 16])

 39%|███▊      | 1933/5000 [24:04:41<27:20:33, 32.09s/it]
                                                         
{'loss': '0.3627', 'grad_norm': '7.484', 'learning_rate': '1.252e-05', 'epoch': '0.3866'}

 39%|███▊      | 1933/5000 [24:04:41<27:20:33, 32.09s/it][2026-05-22 16:10:28,536] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 986])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 313])
[2026-05-22 16:10:29,138] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1003])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 51])

 39%|███▊      | 1934/5000 [24:05:14<27:39:57, 32.48s/it]
                                                         
{'loss': '0.009684', 'grad_norm': '1.306', 'learning_rate': '1.251e-05', 'epoch': '0.3868'}

 39%|███▊      | 1934/5000 [24:05:14<27:39:57, 32.48s/it][2026-05-22 16:11:01,202] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 212])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 244])
[2026-05-22 16:11:01,535] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 446])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 253])

 39%|███▊      | 1935/5000 [24:05:45<27:17:19, 32.05s/it]
                                                         
{'loss': '0.372', 'grad_norm': '7.648', 'learning_rate': '1.251e-05', 'epoch': '0.387'}

 39%|███▊      | 1935/5000 [24:05:45<27:17:19, 32.05s/it][2026-05-22 16:11:32,137] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 228])
[2026-05-22 16:11:32,706] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 254])

 39%|███▊      | 1936/5000 [24:06:17<27:09:46, 31.91s/it]
                                                         
{'loss': '0.03388', 'grad_norm': '3.909', 'learning_rate': '1.251e-05', 'epoch': '0.3872'}

 39%|███▊      | 1936/5000 [24:06:17<27:09:46, 31.91s/it][2026-05-22 16:12:03,808] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 236])
[2026-05-22 16:12:05,150] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 242])

 39%|███▊      | 1937/5000 [24:06:45<26:20:16, 30.96s/it]
                                                         
{'loss': '1.011', 'grad_norm': '23.75', 'learning_rate': '1.25e-05', 'epoch': '0.3874'}

 39%|███▊      | 1937/5000 [24:06:45<26:20:16, 30.96s/it][2026-05-22 16:12:32,769] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])
[2026-05-22 16:12:34,576] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 496])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 232])

 39%|███▉      | 1938/5000 [24:07:16<26:14:36, 30.85s/it]
                                                         
{'loss': '0.8258', 'grad_norm': '17.15', 'learning_rate': '1.25e-05', 'epoch': '0.3876'}

 39%|███▉      | 1938/5000 [24:07:16<26:14:36, 30.85s/it][2026-05-22 16:13:03,356] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 272])
[2026-05-22 16:13:03,502] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 346])

 39%|███▉      | 1939/5000 [24:07:49<26:48:59, 31.54s/it]
                                                         
{'loss': '1.288', 'grad_norm': '22.59', 'learning_rate': '1.249e-05', 'epoch': '0.3878'}

 39%|███▉      | 1939/5000 [24:07:49<26:48:59, 31.54s/it][2026-05-22 16:13:36,661] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 272])
[2026-05-22 16:13:36,749] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])

 39%|███▉      | 1940/5000 [24:08:35<30:34:22, 35.97s/it]
                                                         
{'loss': '0.9315', 'grad_norm': '16.54', 'learning_rate': '1.249e-05', 'epoch': '0.388'}

 39%|███▉      | 1940/5000 [24:08:35<30:34:22, 35.97s/it][2026-05-22 16:14:22,164] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 211])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 250])
[2026-05-22 16:14:22,368] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 266])

 39%|███▉      | 1941/5000 [24:09:02<28:12:53, 33.20s/it]
                                                         
{'loss': '0.5355', 'grad_norm': '10.04', 'learning_rate': '1.249e-05', 'epoch': '0.3882'}

 39%|███▉      | 1941/5000 [24:09:02<28:12:53, 33.20s/it][2026-05-22 16:14:49,208] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 321])
[2026-05-22 16:14:49,303] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])

 39%|███▉      | 1942/5000 [24:09:30<26:57:08, 31.73s/it]
                                                         
{'loss': '0.2695', 'grad_norm': '7.338', 'learning_rate': '1.248e-05', 'epoch': '0.3884'}

 39%|███▉      | 1942/5000 [24:09:30<26:57:08, 31.73s/it][2026-05-22 16:15:18,331] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])
[2026-05-22 16:15:19,385] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])

 39%|███▉      | 1943/5000 [24:10:00<26:25:39, 31.12s/it]
                                                         
{'loss': '0.1366', 'grad_norm': '5.003', 'learning_rate': '1.248e-05', 'epoch': '0.3886'}

 39%|███▉      | 1943/5000 [24:10:00<26:25:39, 31.12s/it][2026-05-22 16:15:48,653] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 319])
[2026-05-22 16:15:49,048] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 495])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])

 39%|███▉      | 1944/5000 [24:10:30<26:10:45, 30.84s/it]
                                                         
{'loss': '0.0528', 'grad_norm': '4.916', 'learning_rate': '1.247e-05', 'epoch': '0.3888'}

 39%|███▉      | 1944/5000 [24:10:30<26:10:45, 30.84s/it][2026-05-22 16:16:17,760] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 233])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 285])
[2026-05-22 16:16:18,836] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 498])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 281])

 39%|███▉      | 1945/5000 [24:10:59<25:41:42, 30.28s/it]
                                                         
{'loss': '0.3997', 'grad_norm': '8.881', 'learning_rate': '1.247e-05', 'epoch': '0.389'}

 39%|███▉      | 1945/5000 [24:10:59<25:41:42, 30.28s/it][2026-05-22 16:16:47,324] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 327])
[2026-05-22 16:16:47,688] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 275])

 39%|███▉      | 1946/5000 [24:11:29<25:26:11, 29.98s/it]
                                                         
{'loss': '0.3595', 'grad_norm': '8.675', 'learning_rate': '1.247e-05', 'epoch': '0.3892'}

 39%|███▉      | 1946/5000 [24:11:29<25:26:11, 29.98s/it][2026-05-22 16:17:16,499] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 92])
[2026-05-22 16:17:18,123] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 496])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 91])

 39%|███▉      | 1947/5000 [24:11:57<24:54:57, 29.38s/it]
                                                         
{'loss': '1.116', 'grad_norm': '20.72', 'learning_rate': '1.246e-05', 'epoch': '0.3894'}

 39%|███▉      | 1947/5000 [24:11:57<24:54:57, 29.38s/it][2026-05-22 16:17:44,560] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 202])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 58])
[2026-05-22 16:17:45,987] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 504])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 64])

 39%|███▉      | 1948/5000 [24:12:25<24:45:55, 29.21s/it]
                                                         
{'loss': '0.7645', 'grad_norm': '17.75', 'learning_rate': '1.246e-05', 'epoch': '0.3896'}

 39%|███▉      | 1948/5000 [24:12:25<24:45:55, 29.21s/it][2026-05-22 16:18:13,049] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 67])
[2026-05-22 16:18:15,101] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 971])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 89])

 39%|███▉      | 1949/5000 [24:12:57<25:15:35, 29.80s/it]
                                                         
{'loss': '0.3593', 'grad_norm': '7.756', 'learning_rate': '1.245e-05', 'epoch': '0.3898'}

 39%|███▉      | 1949/5000 [24:12:57<25:15:35, 29.80s/it][2026-05-22 16:18:44,183] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 993])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 59])
[2026-05-22 16:18:45,674] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 974])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 92])

 39%|███▉      | 1950/5000 [24:13:27<25:22:15, 29.95s/it]
                                                         
{'loss': '0.05182', 'grad_norm': '4.321', 'learning_rate': '1.245e-05', 'epoch': '0.39'}

 39%|███▉      | 1950/5000 [24:13:27<25:22:15, 29.95s/it][2026-05-22 16:19:18,179] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 991])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:19:18,738] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 39%|███▉      | 1951/5000 [24:13:58<25:39:44, 30.30s/it]
                                                         
{'loss': '0.2949', 'grad_norm': '7.191', 'learning_rate': '1.244e-05', 'epoch': '0.3902'}

 39%|███▉      | 1951/5000 [24:13:58<25:39:44, 30.30s/it][2026-05-22 16:19:50,342] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:19:50,665] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 39%|███▉      | 1952/5000 [24:14:31<26:16:49, 31.04s/it]
                                                         
{'loss': '0.05923', 'grad_norm': '5.102', 'learning_rate': '1.244e-05', 'epoch': '0.3904'}

 39%|███▉      | 1952/5000 [24:14:31<26:16:49, 31.04s/it][2026-05-22 16:20:18,172] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])
[2026-05-22 16:20:19,116] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 412])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▉      | 1953/5000 [24:14:59<25:36:36, 30.26s/it]
                                                         
{'loss': '0.06053', 'grad_norm': '5.796', 'learning_rate': '1.244e-05', 'epoch': '0.3906'}

 39%|███▉      | 1953/5000 [24:14:59<25:36:36, 30.26s/it][2026-05-22 16:20:46,771] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:20:46,825] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▉      | 1954/5000 [24:15:28<25:06:48, 29.68s/it]
                                                         
{'loss': '0.07999', 'grad_norm': '4.515', 'learning_rate': '1.243e-05', 'epoch': '0.3908'}

 39%|███▉      | 1954/5000 [24:15:28<25:06:48, 29.68s/it][2026-05-22 16:21:15,445] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 220])
[2026-05-22 16:21:15,938] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 509])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 295])

 39%|███▉      | 1955/5000 [24:16:12<28:53:43, 34.16s/it]
                                                         
{'loss': '0.08628', 'grad_norm': '10.52', 'learning_rate': '1.243e-05', 'epoch': '0.391'}

 39%|███▉      | 1955/5000 [24:16:12<28:53:43, 34.16s/it][2026-05-22 16:22:00,464] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 284])
[2026-05-22 16:22:01,267] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 236])

 39%|███▉      | 1956/5000 [24:16:57<31:29:00, 37.23s/it]
                                                         
{'loss': '0.05771', 'grad_norm': '5.195', 'learning_rate': '1.242e-05', 'epoch': '0.3912'}

 39%|███▉      | 1956/5000 [24:16:57<31:29:00, 37.23s/it][2026-05-22 16:22:43,709] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 16:22:45,216] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 442])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 39%|███▉      | 1957/5000 [24:17:47<34:47:45, 41.17s/it]
                                                         
{'loss': '0.2114', 'grad_norm': '5.388', 'learning_rate': '1.242e-05', 'epoch': '0.3914'}

 39%|███▉      | 1957/5000 [24:17:47<34:47:45, 41.17s/it][2026-05-22 16:23:34,565] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:23:35,000] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 425])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▉      | 1958/5000 [24:18:38<37:12:16, 44.03s/it]
                                                         
{'loss': '0.1403', 'grad_norm': '14.4', 'learning_rate': '1.242e-05', 'epoch': '0.3916'}

 39%|███▉      | 1958/5000 [24:18:38<37:12:16, 44.03s/it][2026-05-22 16:24:25,406] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 240])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 283])
[2026-05-22 16:24:25,789] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▉      | 1959/5000 [24:19:08<33:50:28, 40.06s/it]
                                                         
{'loss': '0.7327', 'grad_norm': '19.24', 'learning_rate': '1.241e-05', 'epoch': '0.3918'}

 39%|███▉      | 1959/5000 [24:19:08<33:50:28, 40.06s/it][2026-05-22 16:24:57,172] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:24:57,230] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▉      | 1960/5000 [24:19:40<31:34:28, 37.39s/it]
                                                         
{'loss': '2.195', 'grad_norm': '31.57', 'learning_rate': '1.241e-05', 'epoch': '0.392'}

 39%|███▉      | 1960/5000 [24:19:40<31:34:28, 37.39s/it][2026-05-22 16:25:27,525] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 639])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 225])
[2026-05-22 16:25:28,332] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 669])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▉      | 1961/5000 [24:20:10<29:51:14, 35.37s/it]
                                                         
{'loss': '0.3145', 'grad_norm': '7.398', 'learning_rate': '1.24e-05', 'epoch': '0.3922'}

 39%|███▉      | 1961/5000 [24:20:10<29:51:14, 35.37s/it][2026-05-22 16:25:58,430] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 658])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 16:25:58,553] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 653])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▉      | 1962/5000 [24:20:41<28:34:31, 33.86s/it]
                                                         
{'loss': '0.02833', 'grad_norm': '3.201', 'learning_rate': '1.24e-05', 'epoch': '0.3924'}

 39%|███▉      | 1962/5000 [24:20:41<28:34:31, 33.86s/it][2026-05-22 16:26:27,509] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 431])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 274])
[2026-05-22 16:26:27,812] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 248])

 39%|███▉      | 1963/5000 [24:21:11<27:38:41, 32.77s/it]
                                                         
{'loss': '0.3535', 'grad_norm': '7.994', 'learning_rate': '1.24e-05', 'epoch': '0.3926'}

 39%|███▉      | 1963/5000 [24:21:11<27:38:41, 32.77s/it][2026-05-22 16:26:58,522] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 272])
[2026-05-22 16:26:58,547] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 278])

 39%|███▉      | 1964/5000 [24:21:41<27:01:13, 32.04s/it]
                                                         
{'loss': '0.02906', 'grad_norm': '2.832', 'learning_rate': '1.239e-05', 'epoch': '0.3928'}

 39%|███▉      | 1964/5000 [24:21:41<27:01:13, 32.04s/it][2026-05-22 16:27:29,038] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 659])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:27:29,912] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 612])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])

 39%|███▉      | 1965/5000 [24:22:13<26:55:55, 31.95s/it]
                                                         
{'loss': '0.3684', 'grad_norm': '7.744', 'learning_rate': '1.239e-05', 'epoch': '0.393'}

 39%|███▉      | 1965/5000 [24:22:13<26:55:55, 31.95s/it][2026-05-22 16:28:00,984] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 639])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 16:28:01,453] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 615])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▉      | 1966/5000 [24:22:45<27:04:39, 32.13s/it]
                                                         
{'loss': '0.0242', 'grad_norm': '2.837', 'learning_rate': '1.238e-05', 'epoch': '0.3932'}

 39%|███▉      | 1966/5000 [24:22:45<27:04:39, 32.13s/it][2026-05-22 16:28:33,486] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 625])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 48])
[2026-05-22 16:28:34,002] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 663])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 61])

 39%|███▉      | 1967/5000 [24:23:23<28:33:32, 33.90s/it]
                                                         
{'loss': '2.185', 'grad_norm': '9.942', 'learning_rate': '1.238e-05', 'epoch': '0.3934'}

 39%|███▉      | 1967/5000 [24:23:23<28:33:32, 33.90s/it][2026-05-22 16:29:11,392] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 634])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])
[2026-05-22 16:29:12,609] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 638])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 108])

 39%|███▉      | 1968/5000 [24:24:02<29:48:55, 35.40s/it]
                                                         
{'loss': '2.098', 'grad_norm': '10.68', 'learning_rate': '1.238e-05', 'epoch': '0.3936'}

 39%|███▉      | 1968/5000 [24:24:02<29:48:55, 35.40s/it][2026-05-22 16:29:50,179] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 222])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])
[2026-05-22 16:29:51,280] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 310])

 39%|███▉      | 1969/5000 [24:24:30<27:57:09, 33.20s/it]
                                                         
{'loss': '0.3702', 'grad_norm': '7.734', 'learning_rate': '1.237e-05', 'epoch': '0.3938'}

 39%|███▉      | 1969/5000 [24:24:30<27:57:09, 33.20s/it][2026-05-22 16:30:18,809] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 326])
[2026-05-22 16:30:18,935] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 446])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])

 39%|███▉      | 1970/5000 [24:25:01<27:22:29, 32.52s/it]
                                                         
{'loss': '0.2902', 'grad_norm': '7.05', 'learning_rate': '1.237e-05', 'epoch': '0.394'}

 39%|███▉      | 1970/5000 [24:25:01<27:22:29, 32.52s/it][2026-05-22 16:30:48,073] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 293])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 16:30:48,353] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 373])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 39%|███▉      | 1971/5000 [24:25:39<28:35:19, 33.98s/it]
                                                         
{'loss': '0.3625', 'grad_norm': '7.192', 'learning_rate': '1.236e-05', 'epoch': '0.3942'}

 39%|███▉      | 1971/5000 [24:25:39<28:35:19, 33.98s/it][2026-05-22 16:31:25,833] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 236])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 16:31:26,365] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 243])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▉      | 1972/5000 [24:26:14<28:54:56, 34.38s/it]
                                                         
{'loss': '0.07853', 'grad_norm': '4.746', 'learning_rate': '1.236e-05', 'epoch': '0.3944'}

 39%|███▉      | 1972/5000 [24:26:14<28:54:56, 34.38s/it][2026-05-22 16:32:01,802] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 238])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 294])
[2026-05-22 16:32:02,413] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▉      | 1973/5000 [24:26:47<28:40:12, 34.10s/it]
                                                         
{'loss': '1.929', 'grad_norm': '9.233', 'learning_rate': '1.236e-05', 'epoch': '0.3946'}

 39%|███▉      | 1973/5000 [24:26:47<28:40:12, 34.10s/it][2026-05-22 16:32:36,092] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 16:32:36,251] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 39%|███▉      | 1974/5000 [24:27:20<28:19:56, 33.71s/it]
                                                         
{'loss': '1.841', 'grad_norm': '9.733', 'learning_rate': '1.235e-05', 'epoch': '0.3948'}

 39%|███▉      | 1974/5000 [24:27:20<28:19:56, 33.71s/it][2026-05-22 16:33:07,088] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 299])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:33:07,560] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 284])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 40%|███▉      | 1975/5000 [24:27:50<27:18:20, 32.50s/it]
                                                         
{'loss': '0.2608', 'grad_norm': '6.823', 'learning_rate': '1.235e-05', 'epoch': '0.395'}

 40%|███▉      | 1975/5000 [24:27:50<27:18:20, 32.50s/it][2026-05-22 16:33:37,333] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 288])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 16:33:37,385] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 309])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 40%|███▉      | 1976/5000 [24:28:20<26:42:58, 31.81s/it]
                                                         
{'loss': '0.03909', 'grad_norm': '3.573', 'learning_rate': '1.234e-05', 'epoch': '0.3952'}

 40%|███▉      | 1976/5000 [24:28:20<26:42:58, 31.81s/it][2026-05-22 16:34:07,788] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 212])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 267])
[2026-05-22 16:34:08,301] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])

 40%|███▉      | 1977/5000 [24:28:51<26:27:20, 31.51s/it]
                                                         
{'loss': '0.3425', 'grad_norm': '5.613', 'learning_rate': '1.234e-05', 'epoch': '0.3954'}

 40%|███▉      | 1977/5000 [24:28:51<26:27:20, 31.51s/it][2026-05-22 16:34:39,128] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 288])
[2026-05-22 16:34:39,153] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 390])

 40%|███▉      | 1978/5000 [24:29:20<25:53:55, 30.85s/it]
                                                         
{'loss': '0.2031', 'grad_norm': '3.913', 'learning_rate': '1.233e-05', 'epoch': '0.3956'}

 40%|███▉      | 1978/5000 [24:29:20<25:53:55, 30.85s/it][2026-05-22 16:35:06,643] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 244])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])
[2026-05-22 16:35:06,658] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 172])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 40%|███▉      | 1979/5000 [24:29:49<25:27:51, 30.34s/it]
                                                         
{'loss': '0.3384', 'grad_norm': '7.375', 'learning_rate': '1.233e-05', 'epoch': '0.3958'}

 40%|███▉      | 1979/5000 [24:29:49<25:27:51, 30.34s/it][2026-05-22 16:35:35,809] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 236])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 16:35:35,816] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 148])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 40%|███▉      | 1980/5000 [24:30:22<26:03:04, 31.05s/it]
                                                         
{'loss': '0.01908', 'grad_norm': '2.324', 'learning_rate': '1.233e-05', 'epoch': '0.396'}

 40%|███▉      | 1980/5000 [24:30:22<26:03:04, 31.05s/it][2026-05-22 16:36:10,684] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 642])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 16:36:10,801] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 645])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 40%|███▉      | 1981/5000 [24:30:51<25:28:25, 30.38s/it]
                                                         
{'loss': '0.2777', 'grad_norm': '7.157', 'learning_rate': '1.232e-05', 'epoch': '0.3962'}

 40%|███▉      | 1981/5000 [24:30:51<25:28:25, 30.38s/it][2026-05-22 16:36:39,447] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 664])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 16:36:39,497] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 854])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])

 40%|███▉      | 1982/5000 [24:31:21<25:27:18, 30.36s/it]
                                                         
{'loss': '0.1181', 'grad_norm': '4.6', 'learning_rate': '1.232e-05', 'epoch': '0.3964'}

 40%|███▉      | 1982/5000 [24:31:21<25:27:18, 30.36s/it][2026-05-22 16:37:08,522] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 231])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 294])
[2026-05-22 16:37:10,380] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 504])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])

 40%|███▉      | 1983/5000 [24:31:51<25:14:05, 30.11s/it]
                                                         
{'loss': '0.4126', 'grad_norm': '8.884', 'learning_rate': '1.231e-05', 'epoch': '0.3966'}

 40%|███▉      | 1983/5000 [24:31:51<25:14:05, 30.11s/it][2026-05-22 16:37:38,479] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 280])
[2026-05-22 16:37:39,420] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 498])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])

 40%|███▉      | 1984/5000 [24:32:23<25:41:01, 30.66s/it]
                                                         
{'loss': '0.05979', 'grad_norm': '4.242', 'learning_rate': '1.231e-05', 'epoch': '0.3968'}

 40%|███▉      | 1984/5000 [24:32:23<25:41:01, 30.66s/it][2026-05-22 16:38:09,790] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:38:10,143] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 292])

 40%|███▉      | 1985/5000 [24:32:46<23:52:21, 28.50s/it]
                                                         
{'loss': '0.4367', 'grad_norm': '9.645', 'learning_rate': '1.231e-05', 'epoch': '0.397'}

 40%|███▉      | 1985/5000 [24:32:46<23:52:21, 28.50s/it][2026-05-22 16:38:33,876] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 16:38:33,921] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 40%|███▉      | 1986/5000 [24:33:10<22:36:45, 27.01s/it]
                                                         
{'loss': '0.3946', 'grad_norm': '9.172', 'learning_rate': '1.23e-05', 'epoch': '0.3972'}

 40%|███▉      | 1986/5000 [24:33:10<22:36:45, 27.01s/it][2026-05-22 16:38:56,821] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 304])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 16:38:58,174] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 320])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 40%|███▉      | 1987/5000 [24:33:43<24:17:27, 29.02s/it]
                                                         
{'loss': '2.133', 'grad_norm': '11.02', 'learning_rate': '1.23e-05', 'epoch': '0.3974'}

 40%|███▉      | 1987/5000 [24:33:43<24:17:27, 29.02s/it][2026-05-22 16:39:31,259] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 336])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:39:31,443] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 272])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 40%|███▉      | 1988/5000 [24:34:24<27:05:58, 32.39s/it]
                                                         
{'loss': '2.115', 'grad_norm': '10.29', 'learning_rate': '1.229e-05', 'epoch': '0.3976'}

 40%|███▉      | 1988/5000 [24:34:24<27:05:58, 32.39s/it][2026-05-22 16:40:10,927] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 480])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:40:11,651] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 786])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 40%|███▉      | 1989/5000 [24:34:53<26:25:45, 31.60s/it]
                                                         
{'loss': '0.1849', 'grad_norm': '6.802', 'learning_rate': '1.229e-05', 'epoch': '0.3978'}

 40%|███▉      | 1989/5000 [24:34:53<26:25:45, 31.60s/it][2026-05-22 16:40:41,329] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:40:42,765] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 855])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 40%|███▉      | 1990/5000 [24:35:25<26:19:35, 31.49s/it]
                                                         
{'loss': '0.04839', 'grad_norm': '5.056', 'learning_rate': '1.229e-05', 'epoch': '0.398'}

 40%|███▉      | 1990/5000 [24:35:25<26:19:35, 31.49s/it][2026-05-22 16:41:11,949] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 317])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 16:41:13,885] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 40%|███▉      | 1991/5000 [24:35:55<25:58:51, 31.08s/it]
                                                         
{'loss': '1.147', 'grad_norm': '23.17', 'learning_rate': '1.228e-05', 'epoch': '0.3982'}

 40%|███▉      | 1991/5000 [24:35:55<25:58:51, 31.08s/it][2026-05-22 16:41:42,247] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 298])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 16:41:43,183] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 430])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 40%|███▉      | 1992/5000 [24:36:27<26:11:36, 31.35s/it]
                                                         
{'loss': '0.9602', 'grad_norm': '17.84', 'learning_rate': '1.228e-05', 'epoch': '0.3984'}

 40%|███▉      | 1992/5000 [24:36:27<26:11:36, 31.35s/it][2026-05-22 16:42:14,716] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 231])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 16:42:15,358] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 504])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 40%|███▉      | 1993/5000 [24:36:57<25:53:23, 31.00s/it]
                                                         
{'loss': '0.3356', 'grad_norm': '8.071', 'learning_rate': '1.227e-05', 'epoch': '0.3986'}

 40%|███▉      | 1993/5000 [24:36:57<25:53:23, 31.00s/it][2026-05-22 16:42:45,008] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 16:42:46,194] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 476])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 40%|███▉      | 1994/5000 [24:37:27<25:38:01, 30.70s/it]
                                                         
{'loss': '0.03175', 'grad_norm': '3.785', 'learning_rate': '1.227e-05', 'epoch': '0.3988'}

 40%|███▉      | 1994/5000 [24:37:27<25:38:01, 30.70s/it][2026-05-22 16:43:13,627] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 242])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])
[2026-05-22 16:43:14,170] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 192])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 40%|███▉      | 1995/5000 [24:38:06<27:47:20, 33.29s/it]
                                                         
{'loss': '0.912', 'grad_norm': '11.93', 'learning_rate': '1.227e-05', 'epoch': '0.399'}

 40%|███▉      | 1995/5000 [24:38:06<27:47:20, 33.29s/it][2026-05-22 16:43:52,963] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 185])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 16:43:53,514] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 162])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 295])

 40%|███▉      | 1996/5000 [24:38:52<31:00:12, 37.15s/it]
                                                         
{'loss': '0.9978', 'grad_norm': '16.01', 'learning_rate': '1.226e-05', 'epoch': '0.3992'}

 40%|███▉      | 1996/5000 [24:38:52<31:00:12, 37.15s/it][2026-05-22 16:44:44,102] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])
[2026-05-22 16:44:44,650] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

 40%|███▉      | 1997/5000 [24:39:26<30:07:15, 36.11s/it]
                                                         
{'loss': '0.2203', 'grad_norm': '7.238', 'learning_rate': '1.226e-05', 'epoch': '0.3994'}

 40%|███▉      | 1997/5000 [24:39:26<30:07:15, 36.11s/it][2026-05-22 16:45:17,444] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1008])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 175])
[2026-05-22 16:45:18,316] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 40%|███▉      | 1998/5000 [24:40:01<29:44:09, 35.66s/it]
                                                         
{'loss': '0.04155', 'grad_norm': '5.003', 'learning_rate': '1.225e-05', 'epoch': '0.3996'}

 40%|███▉      | 1998/5000 [24:40:01<29:44:09, 35.66s/it][2026-05-22 16:45:47,648] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 11])
[2026-05-22 16:45:48,934] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 604])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 6])

 40%|███▉      | 1999/5000 [24:40:33<28:56:10, 34.71s/it]
                                                         
{'loss': '0.113', 'grad_norm': '5.266', 'learning_rate': '1.225e-05', 'epoch': '0.3998'}

 40%|███▉      | 1999/5000 [24:40:33<28:56:10, 34.71s/it][2026-05-22 16:46:21,036] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 976])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 8])
[2026-05-22 16:46:21,626] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 7])

 40%|████      | 2000/5000 [24:41:07<28:43:27, 34.47s/it]
                                                         
{'loss': '0.02653', 'grad_norm': '2.921', 'learning_rate': '1.224e-05', 'epoch': '0.4'}

 40%|████      | 2000/5000 [24:41:07<28:43:27, 34.47s/it][2026-05-22 16:46:52,288] INFO [src.utils.basic_utils:19] Saving model to /workspace/VLM2Vec/outputs/Qwen3.5-08b-continued-mixed-alltasks/checkpoint-2000
[2026-05-22 16:46:53,262] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-22 16:46:53,272] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-22 16:46:55,420] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 432])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 79])
[2026-05-22 16:46:56,311] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 396])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 63])

 40%|████      | 2001/5000 [24:41:33<26:40:03, 32.01s/it]
                                                         
{'loss': '0.4496', 'grad_norm': '9.912', 'learning_rate': '1.224e-05', 'epoch': '0.4002'}

 40%|████      | 2001/5000 [24:41:33<26:40:03, 32.01s/it][2026-05-22 16:47:20,643] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 77])
[2026-05-22 16:47:22,410] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 73])

 40%|████      | 2002/5000 [24:41:59<25:00:47, 30.04s/it]
                                                         
{'loss': '0.4309', 'grad_norm': '9.239', 'learning_rate': '1.224e-05', 'epoch': '0.4004'}

 40%|████      | 2002/5000 [24:41:59<25:00:47, 30.04s/it][2026-05-22 16:47:45,828] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 9])
[2026-05-22 16:47:46,456] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 9])

 40%|████      | 2003/5000 [24:42:46<29:17:33, 35.19s/it]
                                                         
{'loss': '0.1988', 'grad_norm': '7.162', 'learning_rate': '1.223e-05', 'epoch': '0.4006'}

 40%|████      | 2003/5000 [24:42:46<29:17:33, 35.19s/it][2026-05-22 16:48:33,593] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 7])
[2026-05-22 16:48:33,714] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 9])

 40%|████      | 2004/5000 [24:43:34<32:21:41, 38.89s/it]
                                                         
{'loss': '0.203', 'grad_norm': '5.717', 'learning_rate': '1.223e-05', 'epoch': '0.4008'}

 40%|████      | 2004/5000 [24:43:34<32:21:41, 38.89s/it][2026-05-22 16:49:21,215] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:49:21,546] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 793])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 285])

 40%|████      | 2005/5000 [24:44:04<30:21:03, 36.48s/it]
                                                         
{'loss': '1.242', 'grad_norm': '25.67', 'learning_rate': '1.222e-05', 'epoch': '0.401'}

 40%|████      | 2005/5000 [24:44:04<30:21:03, 36.48s/it][2026-05-22 16:49:52,036] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 16:49:52,873] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 788])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 40%|████      | 2006/5000 [24:44:49<32:14:57, 38.78s/it]
                                                         
{'loss': '0.4985', 'grad_norm': '13.08', 'learning_rate': '1.222e-05', 'epoch': '0.4012'}

 40%|████      | 2006/5000 [24:44:49<32:14:57, 38.78s/it][2026-05-22 16:50:36,341] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 295])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 259])
[2026-05-22 16:50:37,183] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 40%|████      | 2007/5000 [24:45:16<29:19:57, 35.28s/it]
                                                         
{'loss': '0.2206', 'grad_norm': '11.11', 'learning_rate': '1.222e-05', 'epoch': '0.4014'}

 40%|████      | 2007/5000 [24:45:16<29:19:57, 35.28s/it][2026-05-22 16:51:03,938] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 16:51:04,360] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 40%|████      | 2008/5000 [24:45:44<27:39:13, 33.27s/it]
                                                         
{'loss': '0.2393', 'grad_norm': '11.39', 'learning_rate': '1.221e-05', 'epoch': '0.4016'}

 40%|████      | 2008/5000 [24:45:44<27:39:13, 33.27s/it][2026-05-22 16:51:31,550] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 268])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 279])
[2026-05-22 16:51:32,927] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 40%|████      | 2009/5000 [24:46:11<25:54:20, 31.18s/it]
                                                         
{'loss': '1.044', 'grad_norm': '22.82', 'learning_rate': '1.221e-05', 'epoch': '0.4018'}

 40%|████      | 2009/5000 [24:46:11<25:54:20, 31.18s/it][2026-05-22 16:51:58,497] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:51:59,296] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 40%|████      | 2010/5000 [24:46:48<27:23:21, 32.98s/it]
                                                         
{'loss': '0.9376', 'grad_norm': '19.95', 'learning_rate': '1.22e-05', 'epoch': '0.402'}

 40%|████      | 2010/5000 [24:46:48<27:23:21, 32.98s/it][2026-05-22 16:52:35,089] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 259])
[2026-05-22 16:52:35,327] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 240])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])

 40%|████      | 2011/5000 [24:47:27<29:03:35, 35.00s/it]
                                                         
{'loss': '0.5033', 'grad_norm': '10.18', 'learning_rate': '1.22e-05', 'epoch': '0.4022'}

 40%|████      | 2011/5000 [24:47:27<29:03:35, 35.00s/it][2026-05-22 16:53:15,445] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 400])
[2026-05-22 16:53:15,517] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 259])

 40%|████      | 2012/5000 [24:48:09<30:47:30, 37.10s/it]
                                                         
{'loss': '0.3153', 'grad_norm': '10.36', 'learning_rate': '1.22e-05', 'epoch': '0.4024'}

 40%|████      | 2012/5000 [24:48:09<30:47:30, 37.10s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 16:53:56,224] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 409])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 16:53:57,057] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 40%|████      | 2013/5000 [24:48:39<28:55:15, 34.86s/it]
                                                         
{'loss': '2.425', 'grad_norm': '262.5', 'learning_rate': '1.219e-05', 'epoch': '0.4026'}

 40%|████      | 2013/5000 [24:48:39<28:55:15, 34.86s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 16:54:25,999] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 306])
[2026-05-22 16:54:26,183] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 467])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 306])

 40%|████      | 2014/5000 [24:49:12<28:25:29, 34.27s/it]
                                                         
{'loss': '1.763', 'grad_norm': '237.6', 'learning_rate': '1.219e-05', 'epoch': '0.4028'}

 40%|████      | 2014/5000 [24:49:12<28:25:29, 34.27s/it][2026-05-22 16:54:58,925] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 265])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 16:54:59,528] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 244])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 40%|████      | 2015/5000 [24:49:43<27:39:02, 33.35s/it]
                                                         
{'loss': '0.5589', 'grad_norm': '9.796', 'learning_rate': '1.218e-05', 'epoch': '0.403'}

 40%|████      | 2015/5000 [24:49:43<27:39:02, 33.35s/it][2026-05-22 16:55:30,620] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 239])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])
[2026-05-22 16:55:30,648] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 263])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 40%|████      | 2016/5000 [24:50:16<27:34:29, 33.27s/it]
                                                         
{'loss': '0.04073', 'grad_norm': '3.289', 'learning_rate': '1.218e-05', 'epoch': '0.4032'}

 40%|████      | 2016/5000 [24:50:16<27:34:29, 33.27s/it][2026-05-22 16:56:03,153] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 289])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 16:56:04,386] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 258])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 40%|████      | 2017/5000 [24:50:44<26:14:32, 31.67s/it]
                                                         
{'loss': '0.4396', 'grad_norm': '8.988', 'learning_rate': '1.218e-05', 'epoch': '0.4034'}

 40%|████      | 2017/5000 [24:50:44<26:14:32, 31.67s/it][2026-05-22 16:56:31,666] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 299])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 16:56:31,762] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 302])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 40%|████      | 2018/5000 [24:51:16<26:11:17, 31.62s/it]
                                                         
{'loss': '0.2669', 'grad_norm': '7.437', 'learning_rate': '1.217e-05', 'epoch': '0.4036'}

 40%|████      | 2018/5000 [24:51:16<26:11:17, 31.62s/it][2026-05-22 16:57:03,304] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 16:57:03,400] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 40%|████      | 2019/5000 [24:51:46<25:48:45, 31.17s/it]
                                                         
{'loss': '0.2272', 'grad_norm': '8.408', 'learning_rate': '1.217e-05', 'epoch': '0.4038'}

 40%|████      | 2019/5000 [24:51:46<25:48:45, 31.17s/it][2026-05-22 16:57:34,155] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 424])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 16:57:34,184] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 40%|████      | 2020/5000 [24:52:15<25:21:03, 30.63s/it]
                                                         
{'loss': '0.09019', 'grad_norm': '6.595', 'learning_rate': '1.216e-05', 'epoch': '0.404'}

 40%|████      | 2020/5000 [24:52:15<25:21:03, 30.63s/it][2026-05-22 16:58:02,455] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 253])
[2026-05-22 16:58:03,784] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 283])

 40%|████      | 2021/5000 [24:52:43<24:44:45, 29.90s/it]
                                                         
{'loss': '0.2957', 'grad_norm': '8.728', 'learning_rate': '1.216e-05', 'epoch': '0.4042'}

 40%|████      | 2021/5000 [24:52:43<24:44:45, 29.90s/it][2026-05-22 16:58:31,190] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 326])
[2026-05-22 16:58:31,324] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 427])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])

 40%|████      | 2022/5000 [24:53:12<24:20:21, 29.42s/it]
                                                         
{'loss': '0.325', 'grad_norm': '9.481', 'learning_rate': '1.216e-05', 'epoch': '0.4044'}

 40%|████      | 2022/5000 [24:53:12<24:20:21, 29.42s/it][2026-05-22 16:58:59,057] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 439])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 16:58:59,127] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 223])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])

 40%|████      | 2023/5000 [24:53:41<24:19:44, 29.42s/it]
                                                         
{'loss': '0.4558', 'grad_norm': '7.911', 'learning_rate': '1.215e-05', 'epoch': '0.4046'}

 40%|████      | 2023/5000 [24:53:41<24:19:44, 29.42s/it][2026-05-22 16:59:28,540] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 180])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 215])
[2026-05-22 16:59:28,980] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 481])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 40%|████      | 2024/5000 [24:54:11<24:19:24, 29.42s/it]
                                                         
{'loss': '0.1132', 'grad_norm': '5.098', 'learning_rate': '1.215e-05', 'epoch': '0.4048'}

 40%|████      | 2024/5000 [24:54:11<24:19:24, 29.42s/it][2026-05-22 16:59:57,759] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 698])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 55])
[2026-05-22 16:59:58,543] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 425])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 72])

 40%|████      | 2025/5000 [24:54:41<24:37:12, 29.79s/it]
                                                         
{'loss': '0.4818', 'grad_norm': '11.28', 'learning_rate': '1.214e-05', 'epoch': '0.405'}

 40%|████      | 2025/5000 [24:54:41<24:37:12, 29.79s/it][2026-05-22 17:00:28,827] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 378])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 46])
[2026-05-22 17:00:29,111] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 759])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 48])

 41%|████      | 2026/5000 [24:55:13<25:10:58, 30.48s/it]
                                                         
{'loss': '0.2786', 'grad_norm': '10.94', 'learning_rate': '1.214e-05', 'epoch': '0.4052'}

 41%|████      | 2026/5000 [24:55:13<25:10:58, 30.48s/it][2026-05-22 17:01:03,116] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 20])
[2026-05-22 17:01:03,692] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1013])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 16])

 41%|████      | 2027/5000 [24:55:43<24:55:32, 30.18s/it]
                                                         
{'loss': '0.5293', 'grad_norm': '10.24', 'learning_rate': '1.213e-05', 'epoch': '0.4054'}

 41%|████      | 2027/5000 [24:55:43<24:55:32, 30.18s/it][2026-05-22 17:01:32,563] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1009])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 19])
[2026-05-22 17:01:33,618] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 15])

 41%|████      | 2028/5000 [24:56:13<24:53:30, 30.15s/it]
                                                         
{'loss': '0.2656', 'grad_norm': '10.05', 'learning_rate': '1.213e-05', 'epoch': '0.4056'}

 41%|████      | 2028/5000 [24:56:13<24:53:30, 30.15s/it][2026-05-22 17:02:01,397] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 193])
[2026-05-22 17:02:01,956] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 198])

 41%|████      | 2029/5000 [24:56:41<24:18:50, 29.46s/it]
                                                         
{'loss': '0.1663', 'grad_norm': '5.793', 'learning_rate': '1.213e-05', 'epoch': '0.4058'}

 41%|████      | 2029/5000 [24:56:41<24:18:50, 29.46s/it][2026-05-22 17:02:29,389] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 206])
[2026-05-22 17:02:29,880] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 180])

 41%|████      | 2030/5000 [24:57:10<24:16:42, 29.43s/it]
                                                         
{'loss': '0.2594', 'grad_norm': '7.314', 'learning_rate': '1.212e-05', 'epoch': '0.406'}

 41%|████      | 2030/5000 [24:57:10<24:16:42, 29.43s/it][2026-05-22 17:02:57,442] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 315])
[2026-05-22 17:02:57,650] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 276])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])

 41%|████      | 2031/5000 [24:57:43<25:01:54, 30.35s/it]
                                                         
{'loss': '1.016', 'grad_norm': '14.12', 'learning_rate': '1.212e-05', 'epoch': '0.4062'}

 41%|████      | 2031/5000 [24:57:43<25:01:54, 30.35s/it][2026-05-22 17:03:29,957] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 413])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 282])
[2026-05-22 17:03:30,715] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 251])

 41%|████      | 2032/5000 [24:58:17<26:09:02, 31.72s/it]
                                                         
{'loss': '0.8291', 'grad_norm': '12.77', 'learning_rate': '1.211e-05', 'epoch': '0.4064'}

 41%|████      | 2032/5000 [24:58:17<26:09:02, 31.72s/it][2026-05-22 17:04:04,345] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 263])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 345])
[2026-05-22 17:04:05,633] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 41%|████      | 2033/5000 [24:59:03<29:30:26, 35.80s/it]
                                                         
{'loss': '0.5424', 'grad_norm': '17.47', 'learning_rate': '1.211e-05', 'epoch': '0.4066'}

 41%|████      | 2033/5000 [24:59:03<29:30:26, 35.80s/it][2026-05-22 17:04:50,093] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 17:04:50,905] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 41%|████      | 2034/5000 [24:59:48<31:49:38, 38.63s/it]
                                                         
{'loss': '0.5634', 'grad_norm': '25', 'learning_rate': '1.211e-05', 'epoch': '0.4068'}

 41%|████      | 2034/5000 [24:59:48<31:49:38, 38.63s/it][2026-05-22 17:05:35,813] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 248])
[2026-05-22 17:05:36,905] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 486])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 258])

 41%|████      | 2035/5000 [25:00:34<33:33:48, 40.75s/it]
                                                         
{'loss': '0.3185', 'grad_norm': '7.674', 'learning_rate': '1.21e-05', 'epoch': '0.407'}

 41%|████      | 2035/5000 [25:00:34<33:33:48, 40.75s/it][2026-05-22 17:06:21,336] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 466])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])
[2026-05-22 17:06:22,724] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 518])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 259])

 41%|████      | 2036/5000 [25:01:20<34:52:06, 42.35s/it]
                                                         
{'loss': '0.3297', 'grad_norm': '9.213', 'learning_rate': '1.21e-05', 'epoch': '0.4072'}

 41%|████      | 2036/5000 [25:01:20<34:52:06, 42.35s/it][2026-05-22 17:07:06,993] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 257])
[2026-05-22 17:07:08,274] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 496])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])

 41%|████      | 2037/5000 [25:01:48<31:22:18, 38.12s/it]
                                                         
{'loss': '0.368', 'grad_norm': '8.268', 'learning_rate': '1.209e-05', 'epoch': '0.4074'}

 41%|████      | 2037/5000 [25:01:48<31:22:18, 38.12s/it][2026-05-22 17:07:36,345] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 482])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 310])
[2026-05-22 17:07:36,571] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 569])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 270])

 41%|████      | 2038/5000 [25:02:17<29:01:45, 35.28s/it]
                                                         
{'loss': '0.1669', 'grad_norm': '5.737', 'learning_rate': '1.209e-05', 'epoch': '0.4076'}

 41%|████      | 2038/5000 [25:02:17<29:01:45, 35.28s/it][2026-05-22 17:08:04,633] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 51])
[2026-05-22 17:08:05,169] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 437])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])

 41%|████      | 2039/5000 [25:02:47<27:51:11, 33.86s/it]
                                                         
{'loss': '0.4203', 'grad_norm': '7.455', 'learning_rate': '1.209e-05', 'epoch': '0.4078'}

 41%|████      | 2039/5000 [25:02:47<27:51:11, 33.86s/it][2026-05-22 17:08:34,435] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 63])
[2026-05-22 17:08:35,712] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])

 41%|████      | 2040/5000 [25:03:18<27:02:32, 32.89s/it]
                                                         
{'loss': '0.1444', 'grad_norm': '9.457', 'learning_rate': '1.208e-05', 'epoch': '0.408'}

 41%|████      | 2040/5000 [25:03:18<27:02:32, 32.89s/it][2026-05-22 17:09:05,177] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 17:09:06,628] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 497])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 111])

 41%|████      | 2041/5000 [25:03:48<26:21:15, 32.06s/it]
                                                         
{'loss': '0.237', 'grad_norm': '5.816', 'learning_rate': '1.208e-05', 'epoch': '0.4082'}

 41%|████      | 2041/5000 [25:03:48<26:21:15, 32.06s/it][2026-05-22 17:09:35,737] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 199])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 103])
[2026-05-22 17:09:36,753] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 41%|████      | 2042/5000 [25:04:19<26:11:20, 31.87s/it]
                                                         
{'loss': '0.06854', 'grad_norm': '4.228', 'learning_rate': '1.207e-05', 'epoch': '0.4084'}

 41%|████      | 2042/5000 [25:04:19<26:11:20, 31.87s/it][2026-05-22 17:10:05,884] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 88])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 17:10:06,542] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 96])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 41%|████      | 2043/5000 [25:04:48<25:24:46, 30.94s/it]
                                                         
{'loss': '0.3737', 'grad_norm': '7.359', 'learning_rate': '1.207e-05', 'epoch': '0.4086'}

 41%|████      | 2043/5000 [25:04:48<25:24:46, 30.94s/it][2026-05-22 17:10:34,623] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 141])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-22 17:10:35,307] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 70])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

 41%|████      | 2044/5000 [25:05:18<25:12:49, 30.71s/it]
                                                         
{'loss': '0.05591', 'grad_norm': '4.652', 'learning_rate': '1.207e-05', 'epoch': '0.4088'}

 41%|████      | 2044/5000 [25:05:18<25:12:49, 30.71s/it][2026-05-22 17:11:09,748] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 984])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 17:11:10,323] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 41%|████      | 2045/5000 [25:05:47<24:40:11, 30.05s/it]
                                                         
{'loss': '0.4369', 'grad_norm': '16.53', 'learning_rate': '1.206e-05', 'epoch': '0.409'}

 41%|████      | 2045/5000 [25:05:47<24:40:11, 30.05s/it][2026-05-22 17:11:38,455] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 183])
[2026-05-22 17:11:38,489] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 215])

 41%|████      | 2046/5000 [25:06:16<24:22:00, 29.70s/it]
                                                         
{'loss': '0.3795', 'grad_norm': '14.06', 'learning_rate': '1.206e-05', 'epoch': '0.4092'}

 41%|████      | 2046/5000 [25:06:16<24:22:00, 29.70s/it][2026-05-22 17:12:02,944] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 933])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 204])
[2026-05-22 17:12:03,707] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 257])

 41%|████      | 2047/5000 [25:06:44<24:00:01, 29.26s/it]
                                                         
{'loss': '0.07345', 'grad_norm': '4.735', 'learning_rate': '1.205e-05', 'epoch': '0.4094'}

 41%|████      | 2047/5000 [25:06:44<24:00:01, 29.26s/it][2026-05-22 17:12:31,551] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 263])
[2026-05-22 17:12:32,042] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 282])

 41%|████      | 2048/5000 [25:07:13<23:49:13, 29.05s/it]
                                                         
{'loss': '0.04884', 'grad_norm': '3.891', 'learning_rate': '1.205e-05', 'epoch': '0.4096'}

 41%|████      | 2048/5000 [25:07:13<23:49:13, 29.05s/it][2026-05-22 17:13:00,597] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 621])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])
[2026-05-22 17:13:00,982] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 624])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 375])

 41%|████      | 2049/5000 [25:07:36<22:24:55, 27.35s/it]
                                                         
{'loss': '0.8944', 'grad_norm': '17.13', 'learning_rate': '1.204e-05', 'epoch': '0.4098'}

 41%|████      | 2049/5000 [25:07:36<22:24:55, 27.35s/it][2026-05-22 17:13:24,256] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 631])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 357])
[2026-05-22 17:13:24,735] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 629])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])

 41%|████      | 2050/5000 [25:08:00<21:30:40, 26.25s/it]
                                                         
{'loss': '0.7137', 'grad_norm': '16.38', 'learning_rate': '1.204e-05', 'epoch': '0.41'}

 41%|████      | 2050/5000 [25:08:00<21:30:40, 26.25s/it][2026-05-22 17:13:47,555] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 639])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])
[2026-05-22 17:13:48,248] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 632])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 41%|████      | 2051/5000 [25:08:47<26:36:20, 32.48s/it]
                                                         
{'loss': '0.1903', 'grad_norm': '6.308', 'learning_rate': '1.204e-05', 'epoch': '0.4102'}

 41%|████      | 2051/5000 [25:08:47<26:36:20, 32.48s/it][2026-05-22 17:14:34,619] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 645])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 17:14:35,220] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 645])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 41%|████      | 2052/5000 [25:09:34<30:15:49, 36.96s/it]
                                                         
{'loss': '0.1907', 'grad_norm': '7.091', 'learning_rate': '1.203e-05', 'epoch': '0.4104'}

 41%|████      | 2052/5000 [25:09:34<30:15:49, 36.96s/it][2026-05-22 17:15:21,806] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 17:15:22,409] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 506])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 41%|████      | 2053/5000 [25:10:19<32:16:55, 39.44s/it]
                                                         
{'loss': '0.5967', 'grad_norm': '12.97', 'learning_rate': '1.203e-05', 'epoch': '0.4106'}

 41%|████      | 2053/5000 [25:10:19<32:16:55, 39.44s/it][2026-05-22 17:16:06,417] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 207])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 366])
[2026-05-22 17:16:08,134] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 41%|████      | 2054/5000 [25:11:05<33:47:57, 41.30s/it]
                                                         
{'loss': '0.4242', 'grad_norm': '11.39', 'learning_rate': '1.202e-05', 'epoch': '0.4108'}

 41%|████      | 2054/5000 [25:11:06<33:47:57, 41.30s/it][2026-05-22 17:16:52,584] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 233])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 312])
[2026-05-22 17:16:52,886] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 253])

 41%|████      | 2055/5000 [25:11:39<32:04:39, 39.21s/it]
                                                         
{'loss': '0.475', 'grad_norm': '8.332', 'learning_rate': '1.202e-05', 'epoch': '0.411'}

 41%|████      | 2055/5000 [25:11:39<32:04:39, 39.21s/it][2026-05-22 17:17:26,653] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 248])
[2026-05-22 17:17:27,281] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 276])

 41%|████      | 2056/5000 [25:12:13<30:47:09, 37.65s/it]
                                                         
{'loss': '0.1493', 'grad_norm': '5.148', 'learning_rate': '1.202e-05', 'epoch': '0.4112'}

 41%|████      | 2056/5000 [25:12:13<30:47:09, 37.65s/it][2026-05-22 17:18:00,331] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 267])
[2026-05-22 17:18:00,537] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 219])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])

 41%|████      | 2057/5000 [25:12:49<30:18:59, 37.08s/it]
                                                         
{'loss': '2.071', 'grad_norm': '14.13', 'learning_rate': '1.201e-05', 'epoch': '0.4114'}

 41%|████      | 2057/5000 [25:12:49<30:18:59, 37.08s/it][2026-05-22 17:18:36,074] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 402])
[2026-05-22 17:18:36,756] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 492])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 253])

 41%|████      | 2058/5000 [25:13:26<30:18:46, 37.09s/it]
                                                         
{'loss': '2.004', 'grad_norm': '12.63', 'learning_rate': '1.201e-05', 'epoch': '0.4116'}

 41%|████      | 2058/5000 [25:13:26<30:18:46, 37.09s/it][2026-05-22 17:19:13,397] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 248])
[2026-05-22 17:19:15,390] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 478])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 295])

 41%|████      | 2059/5000 [25:13:59<29:14:22, 35.79s/it]
                                                         
{'loss': '0.1437', 'grad_norm': '7.937', 'learning_rate': '1.2e-05', 'epoch': '0.4118'}

 41%|████      | 2059/5000 [25:13:59<29:14:22, 35.79s/it][2026-05-22 17:19:47,033] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])
[2026-05-22 17:19:47,439] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 257])

 41%|████      | 2060/5000 [25:14:32<28:30:06, 34.90s/it]
                                                         
{'loss': '0.05994', 'grad_norm': '5.077', 'learning_rate': '1.2e-05', 'epoch': '0.412'}

 41%|████      | 2060/5000 [25:14:32<28:30:06, 34.90s/it][2026-05-22 17:20:19,385] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 281])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])
[2026-05-22 17:20:19,658] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 41%|████      | 2061/5000 [25:15:00<26:53:14, 32.93s/it]
                                                         
{'loss': '0.3818', 'grad_norm': '9.086', 'learning_rate': '1.2e-05', 'epoch': '0.4122'}

 41%|████      | 2061/5000 [25:15:00<26:53:14, 32.93s/it][2026-05-22 17:20:47,948] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 17:20:48,786] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 41%|████      | 2062/5000 [25:15:30<26:08:55, 32.04s/it]
                                                         
{'loss': '0.3324', 'grad_norm': '7.716', 'learning_rate': '1.199e-05', 'epoch': '0.4124'}

 41%|████      | 2062/5000 [25:15:30<26:08:55, 32.04s/it][2026-05-22 17:21:17,034] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 422])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 17:21:18,718] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 41%|████▏     | 2063/5000 [25:15:58<25:08:07, 30.81s/it]
                                                         
{'loss': '0.5004', 'grad_norm': '9.205', 'learning_rate': '1.199e-05', 'epoch': '0.4126'}

 41%|████▏     | 2063/5000 [25:15:58<25:08:07, 30.81s/it][2026-05-22 17:21:45,523] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 199])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 17:21:45,878] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 522])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])

 41%|████▏     | 2064/5000 [25:16:29<25:13:26, 30.93s/it]
                                                         
{'loss': '0.3272', 'grad_norm': '8.372', 'learning_rate': '1.198e-05', 'epoch': '0.4128'}

 41%|████▏     | 2064/5000 [25:16:29<25:13:26, 30.93s/it][2026-05-22 17:22:16,981] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 218])
[2026-05-22 17:22:17,013] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 275])

 41%|████▏     | 2065/5000 [25:16:59<24:52:44, 30.52s/it]
                                                         
{'loss': '0.2307', 'grad_norm': '6.52', 'learning_rate': '1.198e-05', 'epoch': '0.413'}

 41%|████▏     | 2065/5000 [25:16:59<24:52:44, 30.52s/it][2026-05-22 17:22:45,882] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 370])
[2026-05-22 17:22:46,398] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 151])

 41%|████▏     | 2066/5000 [25:17:28<24:40:05, 30.27s/it]
                                                         
{'loss': '0.04572', 'grad_norm': '4.404', 'learning_rate': '1.198e-05', 'epoch': '0.4132'}

 41%|████▏     | 2066/5000 [25:17:28<24:40:05, 30.27s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 17:23:14,917] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 80])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 198])
[2026-05-22 17:23:14,919] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 80])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 203])

 41%|████▏     | 2067/5000 [25:17:59<24:51:08, 30.50s/it]
                                                         
{'loss': '0.3374', 'grad_norm': '7.073', 'learning_rate': '1.197e-05', 'epoch': '0.4134'}

 41%|████▏     | 2067/5000 [25:17:59<24:51:08, 30.50s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 17:23:46,010] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 80])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 214])
[2026-05-22 17:23:46,599] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 80])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 221])

 41%|████▏     | 2068/5000 [25:18:30<24:58:16, 30.66s/it]
                                                         
{'loss': '0.01153', 'grad_norm': '1.937', 'learning_rate': '1.197e-05', 'epoch': '0.4136'}

 41%|████▏     | 2068/5000 [25:18:31<24:58:16, 30.66s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 17:24:18,240] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 228])
[2026-05-22 17:24:19,381] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 482])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 283])

 41%|████▏     | 2069/5000 [25:19:00<24:38:57, 30.28s/it]
                                                         
{'loss': '0.1154', 'grad_norm': '5.954', 'learning_rate': '1.196e-05', 'epoch': '0.4138'}

 41%|████▏     | 2069/5000 [25:19:00<24:38:57, 30.28s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 17:24:46,962] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 170])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 212])
[2026-05-22 17:24:48,751] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 192])

 41%|████▏     | 2070/5000 [25:19:30<24:37:37, 30.26s/it]
                                                         
{'loss': '0.1281', 'grad_norm': '5.932', 'learning_rate': '1.196e-05', 'epoch': '0.414'}

 41%|████▏     | 2070/5000 [25:19:30<24:37:37, 30.26s/it][2026-05-22 17:25:17,389] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 17:25:18,617] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 480])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 41%|████▏     | 2071/5000 [25:19:59<24:16:41, 29.84s/it]
                                                         
{'loss': '1.382', 'grad_norm': '32.51', 'learning_rate': '1.196e-05', 'epoch': '0.4142'}

 41%|████▏     | 2071/5000 [25:19:59<24:16:41, 29.84s/it][2026-05-22 17:25:47,131] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 249])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 17:25:47,547] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 436])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 41%|████▏     | 2072/5000 [25:20:29<24:19:39, 29.91s/it]
                                                         
{'loss': '2.258', 'grad_norm': '27.28', 'learning_rate': '1.195e-05', 'epoch': '0.4144'}

 41%|████▏     | 2072/5000 [25:20:29<24:19:39, 29.91s/it][2026-05-22 17:26:16,934] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 258])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 17:26:18,280] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 432])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 41%|████▏     | 2073/5000 [25:20:53<22:53:49, 28.16s/it]
                                                         
{'loss': '0.2272', 'grad_norm': '5.117', 'learning_rate': '1.195e-05', 'epoch': '0.4146'}

 41%|████▏     | 2073/5000 [25:20:53<22:53:49, 28.16s/it][2026-05-22 17:26:40,520] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 242])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 17:26:42,219] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 505])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 41%|████▏     | 2074/5000 [25:21:18<21:58:18, 27.03s/it]
                                                         
{'loss': '0.2307', 'grad_norm': '7.216', 'learning_rate': '1.194e-05', 'epoch': '0.4148'}

 41%|████▏     | 2074/5000 [25:21:18<21:58:18, 27.03s/it][2026-05-22 17:27:04,176] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 256])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 17:27:04,562] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 267])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 42%|████▏     | 2075/5000 [25:21:47<22:28:36, 27.66s/it]
                                                         
{'loss': '0.05791', 'grad_norm': '4.556', 'learning_rate': '1.194e-05', 'epoch': '0.415'}

 42%|████▏     | 2075/5000 [25:21:47<22:28:36, 27.66s/it][2026-05-22 17:27:33,700] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 311])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 17:27:34,408] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 238])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 565])

 42%|████▏     | 2076/5000 [25:22:15<22:44:40, 28.00s/it]
                                                         
{'loss': '0.0345', 'grad_norm': '3.652', 'learning_rate': '1.193e-05', 'epoch': '0.4152'}

 42%|████▏     | 2076/5000 [25:22:15<22:44:40, 28.00s/it][2026-05-22 17:28:03,249] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 17:28:04,713] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 42%|████▏     | 2077/5000 [25:22:48<23:45:32, 29.26s/it]
                                                         
{'loss': '0.1451', 'grad_norm': '4.823', 'learning_rate': '1.193e-05', 'epoch': '0.4154'}

 42%|████▏     | 2077/5000 [25:22:48<23:45:32, 29.26s/it][2026-05-22 17:28:35,677] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 17:28:37,029] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 42%|████▏     | 2078/5000 [25:23:20<24:35:40, 30.30s/it]
                                                         
{'loss': '0.07643', 'grad_norm': '5.316', 'learning_rate': '1.193e-05', 'epoch': '0.4156'}

 42%|████▏     | 2078/5000 [25:23:20<24:35:40, 30.30s/it][2026-05-22 17:29:08,309] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 79])
[2026-05-22 17:29:09,305] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 462])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 81])

 42%|████▏     | 2079/5000 [25:23:53<25:09:47, 31.01s/it]
                                                         
{'loss': '0.176', 'grad_norm': '7.177', 'learning_rate': '1.192e-05', 'epoch': '0.4158'}

 42%|████▏     | 2079/5000 [25:23:53<25:09:47, 31.01s/it][2026-05-22 17:29:40,217] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 199])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 86])
[2026-05-22 17:29:41,888] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 483])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 75])

 42%|████▏     | 2080/5000 [25:24:27<25:48:00, 31.81s/it]
                                                         
{'loss': '0.04582', 'grad_norm': '6.281', 'learning_rate': '1.192e-05', 'epoch': '0.416'}

 42%|████▏     | 2080/5000 [25:24:27<25:48:00, 31.81s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 17:30:14,007] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 225])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 17:30:15,546] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 259])

 42%|████▏     | 2081/5000 [25:24:55<24:51:03, 30.65s/it]
                                                         
{'loss': '1.117', 'grad_norm': '18.77', 'learning_rate': '1.191e-05', 'epoch': '0.4162'}

 42%|████▏     | 2081/5000 [25:24:55<24:51:03, 30.65s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 17:30:43,105] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 327])
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 17:30:43,610] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 292])

 42%|████▏     | 2082/5000 [25:25:26<24:54:42, 30.73s/it]
                                                         
{'loss': '0.9298', 'grad_norm': '17.66', 'learning_rate': '1.191e-05', 'epoch': '0.4164'}

 42%|████▏     | 2082/5000 [25:25:26<24:54:42, 30.73s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 17:31:12,727] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 452])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 312])
[2026-05-22 17:31:12,966] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 410])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 245])

 42%|████▏     | 2083/5000 [25:25:57<25:08:54, 31.04s/it]
                                                         
{'loss': '0.1388', 'grad_norm': '4.96', 'learning_rate': '1.191e-05', 'epoch': '0.4166'}

 42%|████▏     | 2083/5000 [25:25:57<25:08:54, 31.04s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 17:31:44,374] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 293])
[2026-05-22 17:31:44,487] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])

 42%|████▏     | 2084/5000 [25:26:29<25:15:37, 31.19s/it]
                                                         
{'loss': '0.03216', 'grad_norm': '3.009', 'learning_rate': '1.19e-05', 'epoch': '0.4168'}

 42%|████▏     | 2084/5000 [25:26:29<25:15:37, 31.19s/it][2026-05-22 17:32:15,994] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 692])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 285])
[2026-05-22 17:32:16,650] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 244])

 42%|████▏     | 2085/5000 [25:26:57<24:26:49, 30.19s/it]
                                                         
{'loss': '0.07974', 'grad_norm': '5.67', 'learning_rate': '1.19e-05', 'epoch': '0.417'}

 42%|████▏     | 2085/5000 [25:26:57<24:26:49, 30.19s/it][2026-05-22 17:32:44,048] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 788])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 303])
[2026-05-22 17:32:44,609] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 446])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 235])

 42%|████▏     | 2086/5000 [25:27:25<23:55:37, 29.56s/it]
                                                         
{'loss': '0.1161', 'grad_norm': '6.686', 'learning_rate': '1.189e-05', 'epoch': '0.4172'}

 42%|████▏     | 2086/5000 [25:27:25<23:55:37, 29.56s/it][2026-05-22 17:33:11,843] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 267])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 17:33:12,188] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])

 42%|████▏     | 2087/5000 [25:27:54<23:52:40, 29.51s/it]
                                                         
{'loss': '0.3716', 'grad_norm': '8.051', 'learning_rate': '1.189e-05', 'epoch': '0.4174'}

 42%|████▏     | 2087/5000 [25:27:54<23:52:40, 29.51s/it][2026-05-22 17:33:41,315] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 17:33:41,919] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 292])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 42%|████▏     | 2088/5000 [25:28:24<23:54:17, 29.55s/it]
                                                         
{'loss': '0.01847', 'grad_norm': '2.15', 'learning_rate': '1.189e-05', 'epoch': '0.4176'}

 42%|████▏     | 2088/5000 [25:28:24<23:54:17, 29.55s/it][2026-05-22 17:34:10,521] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 315])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 17:34:11,527] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 42%|████▏     | 2089/5000 [25:28:52<23:40:12, 29.27s/it]
                                                         
{'loss': '0.2808', 'grad_norm': '8.512', 'learning_rate': '1.188e-05', 'epoch': '0.4178'}

 42%|████▏     | 2089/5000 [25:28:52<23:40:12, 29.27s/it][2026-05-22 17:34:39,430] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 17:34:40,001] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 42%|████▏     | 2090/5000 [25:29:22<23:37:06, 29.22s/it]
                                                         
{'loss': '0.08741', 'grad_norm': '6.361', 'learning_rate': '1.188e-05', 'epoch': '0.418'}

 42%|████▏     | 2090/5000 [25:29:22<23:37:06, 29.22s/it][2026-05-22 17:35:10,087] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 639])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 286])
[2026-05-22 17:35:10,584] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 652])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])

 42%|████▏     | 2091/5000 [25:29:56<24:55:24, 30.84s/it]
                                                         
{'loss': '0.9896', 'grad_norm': '12.81', 'learning_rate': '1.187e-05', 'epoch': '0.4182'}

 42%|████▏     | 2091/5000 [25:29:56<24:55:24, 30.84s/it][2026-05-22 17:35:44,487] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 653])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 276])
[2026-05-22 17:35:45,072] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 633])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 306])

 42%|████▏     | 2092/5000 [25:30:35<26:47:20, 33.16s/it]
                                                         
{'loss': '1.001', 'grad_norm': '12.47', 'learning_rate': '1.187e-05', 'epoch': '0.4184'}

 42%|████▏     | 2092/5000 [25:30:35<26:47:20, 33.16s/it][2026-05-22 17:36:21,908] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 667])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 17:36:21,924] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 637])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 223])

 42%|████▏     | 2093/5000 [25:31:03<25:29:19, 31.56s/it]
                                                         
{'loss': '1.02', 'grad_norm': '17.56', 'learning_rate': '1.187e-05', 'epoch': '0.4186'}

 42%|████▏     | 2093/5000 [25:31:03<25:29:19, 31.56s/it][2026-05-22 17:36:49,743] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 622])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 17:36:50,283] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 681])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 42%|████▏     | 2094/5000 [25:31:32<24:53:32, 30.84s/it]
                                                         
{'loss': '0.6902', 'grad_norm': '16.37', 'learning_rate': '1.186e-05', 'epoch': '0.4188'}

 42%|████▏     | 2094/5000 [25:31:32<24:53:32, 30.84s/it][2026-05-22 17:37:20,474] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 225])
[2026-05-22 17:37:20,839] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 65])

 42%|████▏     | 2095/5000 [25:32:03<24:55:44, 30.89s/it]
                                                         
{'loss': '0.3379', 'grad_norm': '8.119', 'learning_rate': '1.186e-05', 'epoch': '0.419'}

 42%|████▏     | 2095/5000 [25:32:03<24:55:44, 30.89s/it][2026-05-22 17:37:51,170] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 189])
[2026-05-22 17:37:51,623] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 113])

 42%|████▏     | 2096/5000 [25:32:36<25:29:35, 31.60s/it]
                                                         
{'loss': '0.2337', 'grad_norm': '7.353', 'learning_rate': '1.185e-05', 'epoch': '0.4192'}

 42%|████▏     | 2096/5000 [25:32:36<25:29:35, 31.60s/it][2026-05-22 17:38:24,466] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 17:38:24,869] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 42%|████▏     | 2097/5000 [25:33:12<26:31:23, 32.89s/it]
                                                         
{'loss': '1.381', 'grad_norm': '11.44', 'learning_rate': '1.185e-05', 'epoch': '0.4194'}

 42%|████▏     | 2097/5000 [25:33:12<26:31:23, 32.89s/it][2026-05-22 17:39:00,553] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 17:39:01,083] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 42%|████▏     | 2098/5000 [25:33:48<27:16:17, 33.83s/it]
                                                         
{'loss': '1.052', 'grad_norm': '8.112', 'learning_rate': '1.184e-05', 'epoch': '0.4196'}

 42%|████▏     | 2098/5000 [25:33:48<27:16:17, 33.83s/it][2026-05-22 17:39:35,533] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 292])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 17:39:36,212] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 300])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 42%|████▏     | 2099/5000 [25:34:22<27:14:47, 33.81s/it]
                                                         
{'loss': '1.535', 'grad_norm': '18.07', 'learning_rate': '1.184e-05', 'epoch': '0.4198'}

 42%|████▏     | 2099/5000 [25:34:22<27:14:47, 33.81s/it][2026-05-22 17:40:09,435] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 266])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])
[2026-05-22 17:40:09,981] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 259])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 42%|████▏     | 2100/5000 [25:34:55<27:07:48, 33.68s/it]
                                                         
{'loss': '1.563', 'grad_norm': '19.01', 'learning_rate': '1.184e-05', 'epoch': '0.42'}

 42%|████▏     | 2100/5000 [25:34:55<27:07:48, 33.68s/it][2026-05-22 17:40:42,235] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 428])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])
[2026-05-22 17:40:42,289] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 426])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])

 42%|████▏     | 2101/5000 [25:35:41<30:05:59, 37.38s/it]
                                                         
{'loss': '2.64', 'grad_norm': '28.1', 'learning_rate': '1.183e-05', 'epoch': '0.4202'}

 42%|████▏     | 2101/5000 [25:35:41<30:05:59, 37.38s/it][2026-05-22 17:41:28,325] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 429])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 17:41:28,968] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 42%|████▏     | 2102/5000 [25:36:27<32:03:17, 39.82s/it]
                                                         
{'loss': '3.513', 'grad_norm': '23.33', 'learning_rate': '1.183e-05', 'epoch': '0.4204'}

 42%|████▏     | 2102/5000 [25:36:27<32:03:17, 39.82s/it][2026-05-22 17:42:14,452] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 322])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 225])
[2026-05-22 17:42:15,048] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 42%|████▏     | 2103/5000 [25:37:16<34:14:02, 42.54s/it]
                                                         
{'loss': '3.062', 'grad_norm': '10.53', 'learning_rate': '1.182e-05', 'epoch': '0.4206'}

 42%|████▏     | 2103/5000 [25:37:16<34:14:02, 42.54s/it][2026-05-22 17:43:03,450] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 447])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 17:43:04,073] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 471])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 42%|████▏     | 2104/5000 [25:38:04<35:45:52, 44.46s/it]
                                                         
{'loss': '3.185', 'grad_norm': '12.43', 'learning_rate': '1.182e-05', 'epoch': '0.4208'}

 42%|████▏     | 2104/5000 [25:38:04<35:45:52, 44.46s/it][2026-05-22 17:43:51,532] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 185])
[2026-05-22 17:43:51,812] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 42%|████▏     | 2105/5000 [25:38:34<32:12:19, 40.05s/it]
                                                         
{'loss': '0.3097', 'grad_norm': '8.198', 'learning_rate': '1.182e-05', 'epoch': '0.421'}

 42%|████▏     | 2105/5000 [25:38:34<32:12:19, 40.05s/it][2026-05-22 17:44:21,478] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 960])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 17:44:22,169] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 632])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 42%|████▏     | 2106/5000 [25:39:04<29:43:50, 36.98s/it]
                                                         
{'loss': '0.08895', 'grad_norm': '5.669', 'learning_rate': '1.181e-05', 'epoch': '0.4212'}

 42%|████▏     | 2106/5000 [25:39:04<29:43:50, 36.98s/it][2026-05-22 17:44:51,438] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])
[2026-05-22 17:44:52,031] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 458])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 159])

 42%|████▏     | 2107/5000 [25:39:34<27:54:53, 34.74s/it]
                                                         
{'loss': '0.465', 'grad_norm': '17.5', 'learning_rate': '1.181e-05', 'epoch': '0.4214'}

 42%|████▏     | 2107/5000 [25:39:34<27:54:53, 34.74s/it][2026-05-22 17:45:20,840] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 171])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])
[2026-05-22 17:45:21,491] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 447])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 42%|████▏     | 2108/5000 [25:40:04<26:49:39, 33.40s/it]
                                                         
{'loss': '0.4522', 'grad_norm': '16.09', 'learning_rate': '1.18e-05', 'epoch': '0.4216'}

 42%|████▏     | 2108/5000 [25:40:04<26:49:39, 33.40s/it][2026-05-22 17:45:50,967] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 236])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 17:45:51,404] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 42%|████▏     | 2109/5000 [25:40:36<26:28:08, 32.96s/it]
                                                         
{'loss': '0.4267', 'grad_norm': '7.18', 'learning_rate': '1.18e-05', 'epoch': '0.4218'}

 42%|████▏     | 2109/5000 [25:40:36<26:28:08, 32.96s/it][2026-05-22 17:46:22,967] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 210])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 17:46:23,356] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 481])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 42%|████▏     | 2110/5000 [25:41:09<26:26:21, 32.93s/it]
                                                         
{'loss': '0.05191', 'grad_norm': '3.844', 'learning_rate': '1.18e-05', 'epoch': '0.422'}

 42%|████▏     | 2110/5000 [25:41:09<26:26:21, 32.93s/it][2026-05-22 17:46:55,300] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 219])
[2026-05-22 17:46:56,300] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 268])

 42%|████▏     | 2111/5000 [25:41:56<29:52:09, 37.22s/it]
                                                         
{'loss': '0.5064', 'grad_norm': '11.72', 'learning_rate': '1.179e-05', 'epoch': '0.4222'}

 42%|████▏     | 2111/5000 [25:41:56<29:52:09, 37.22s/it][2026-05-22 17:47:43,509] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 493])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 264])
[2026-05-22 17:47:43,577] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 509])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])

 42%|████▏     | 2112/5000 [25:42:43<32:10:02, 40.10s/it]
                                                         
{'loss': '0.4202', 'grad_norm': '11.58', 'learning_rate': '1.179e-05', 'epoch': '0.4224'}

 42%|████▏     | 2112/5000 [25:42:43<32:10:02, 40.10s/it][2026-05-22 17:48:30,462] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 184])
[2026-05-22 17:48:31,304] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])

 42%|████▏     | 2113/5000 [25:43:11<29:22:27, 36.63s/it]
                                                         
{'loss': '0.8566', 'grad_norm': '18.17', 'learning_rate': '1.178e-05', 'epoch': '0.4226'}

 42%|████▏     | 2113/5000 [25:43:11<29:22:27, 36.63s/it][2026-05-22 17:48:58,709] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 239])
[2026-05-22 17:48:59,691] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])

 42%|████▏     | 2114/5000 [25:43:39<27:21:03, 34.12s/it]
                                                         
{'loss': '0.8412', 'grad_norm': '16.92', 'learning_rate': '1.178e-05', 'epoch': '0.4228'}

 42%|████▏     | 2114/5000 [25:43:39<27:21:03, 34.12s/it][2026-05-22 17:49:26,110] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 17:49:26,856] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])

 42%|████▏     | 2115/5000 [25:44:11<26:40:20, 33.28s/it]
                                                         
{'loss': '0.241', 'grad_norm': '9.557', 'learning_rate': '1.178e-05', 'epoch': '0.423'}

 42%|████▏     | 2115/5000 [25:44:11<26:40:20, 33.28s/it][2026-05-22 17:49:58,035] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 483])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])
[2026-05-22 17:49:58,040] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 968])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 42%|████▏     | 2116/5000 [25:44:42<26:13:07, 32.73s/it]
                                                         
{'loss': '0.3078', 'grad_norm': '10.89', 'learning_rate': '1.177e-05', 'epoch': '0.4232'}

 42%|████▏     | 2116/5000 [25:44:42<26:13:07, 32.73s/it][2026-05-22 17:50:29,980] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 17:50:29,986] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 416])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 279])

 42%|████▏     | 2117/5000 [25:45:10<25:03:27, 31.29s/it]
                                                         
{'loss': '0.9729', 'grad_norm': '17.05', 'learning_rate': '1.177e-05', 'epoch': '0.4234'}

 42%|████▏     | 2117/5000 [25:45:10<25:03:27, 31.29s/it][2026-05-22 17:50:57,930] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 17:50:58,043] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 569])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 42%|████▏     | 2118/5000 [25:45:40<24:40:40, 30.83s/it]
                                                         
{'loss': '0.8446', 'grad_norm': '16.98', 'learning_rate': '1.176e-05', 'epoch': '0.4236'}

 42%|████▏     | 2118/5000 [25:45:40<24:40:40, 30.83s/it][2026-05-22 17:51:27,065] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 272])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 211])
[2026-05-22 17:51:28,409] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 42%|████▏     | 2119/5000 [25:46:09<24:22:04, 30.45s/it]
                                                         
{'loss': '0.1965', 'grad_norm': '6.053', 'learning_rate': '1.176e-05', 'epoch': '0.4238'}

 42%|████▏     | 2119/5000 [25:46:09<24:22:04, 30.45s/it][2026-05-22 17:51:58,046] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 17:51:58,138] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 42%|████▏     | 2120/5000 [25:46:40<24:20:19, 30.42s/it]
                                                         
{'loss': '0.02803', 'grad_norm': '3.083', 'learning_rate': '1.176e-05', 'epoch': '0.424'}

 42%|████▏     | 2120/5000 [25:46:40<24:20:19, 30.42s/it][2026-05-22 17:52:31,521] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 265])
[2026-05-22 17:52:31,917] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 995])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])

 42%|████▏     | 2121/5000 [25:47:26<28:00:21, 35.02s/it]
                                                         
{'loss': '0.4814', 'grad_norm': '10.66', 'learning_rate': '1.175e-05', 'epoch': '0.4242'}

 42%|████▏     | 2121/5000 [25:47:26<28:00:21, 35.02s/it][2026-05-22 17:53:17,347] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 336])
[2026-05-22 17:53:18,456] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 276])

 42%|████▏     | 2122/5000 [25:48:10<30:21:18, 37.97s/it]
                                                         
{'loss': '0.3946', 'grad_norm': '10.11', 'learning_rate': '1.175e-05', 'epoch': '0.4244'}

 42%|████▏     | 2122/5000 [25:48:10<30:21:18, 37.97s/it][2026-05-22 17:53:56,826] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 132])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 111])
[2026-05-22 17:53:56,846] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 284])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 199])

 42%|████▏     | 2123/5000 [25:48:50<30:40:56, 38.39s/it]
                                                         
{'loss': '0.4205', 'grad_norm': '9.473', 'learning_rate': '1.174e-05', 'epoch': '0.4246'}

 42%|████▏     | 2123/5000 [25:48:50<30:40:56, 38.39s/it][2026-05-22 17:54:36,218] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 156])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])
[2026-05-22 17:54:36,828] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 188])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])

 42%|████▏     | 2124/5000 [25:49:29<30:57:27, 38.75s/it]
                                                         
{'loss': '0.1345', 'grad_norm': '6.413', 'learning_rate': '1.174e-05', 'epoch': '0.4248'}

 42%|████▏     | 2124/5000 [25:49:29<30:57:27, 38.75s/it][2026-05-22 17:55:16,446] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 186])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 145])
[2026-05-22 17:55:16,543] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 172])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 42%|████▎     | 2125/5000 [25:49:59<28:41:06, 35.92s/it]
                                                         
{'loss': '0.401', 'grad_norm': '7.992', 'learning_rate': '1.173e-05', 'epoch': '0.425'}

 42%|████▎     | 2125/5000 [25:49:59<28:41:06, 35.92s/it][2026-05-22 17:55:45,254] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 246])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 17:55:45,909] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 264])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 43%|████▎     | 2126/5000 [25:50:31<27:50:58, 34.88s/it]
                                                         
{'loss': '0.0183', 'grad_norm': '2.508', 'learning_rate': '1.173e-05', 'epoch': '0.4252'}

 43%|████▎     | 2126/5000 [25:50:31<27:50:58, 34.88s/it][2026-05-22 17:56:18,331] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 265])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 17:56:19,285] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 581])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 43%|████▎     | 2127/5000 [25:51:19<31:02:15, 38.89s/it]
                                                         
{'loss': '0.3191', 'grad_norm': '7.492', 'learning_rate': '1.173e-05', 'epoch': '0.4254'}

 43%|████▎     | 2127/5000 [25:51:19<31:02:15, 38.89s/it][2026-05-22 17:57:07,674] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 337])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 17:57:07,743] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 457])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 43%|████▎     | 2128/5000 [25:52:08<33:25:57, 41.91s/it]
                                                         
{'loss': '0.2439', 'grad_norm': '20.62', 'learning_rate': '1.172e-05', 'epoch': '0.4256'}

 43%|████▎     | 2128/5000 [25:52:08<33:25:57, 41.91s/it][2026-05-22 17:57:55,678] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 81])
[2026-05-22 17:57:56,231] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 439])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 106])

 43%|████▎     | 2129/5000 [25:52:33<29:19:58, 36.78s/it]
                                                         
{'loss': '0.4208', 'grad_norm': '8.409', 'learning_rate': '1.172e-05', 'epoch': '0.4258'}

 43%|████▎     | 2129/5000 [25:52:33<29:19:58, 36.78s/it][2026-05-22 17:58:19,906] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 166])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 126])
[2026-05-22 17:58:21,041] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 462])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 70])

 43%|████▎     | 2130/5000 [25:52:56<26:00:41, 32.63s/it]
                                                         
{'loss': '0.3814', 'grad_norm': '8.832', 'learning_rate': '1.171e-05', 'epoch': '0.426'}

 43%|████▎     | 2130/5000 [25:52:56<26:00:41, 32.63s/it][2026-05-22 17:58:43,384] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 64])
[2026-05-22 17:58:44,248] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 540])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 72])

 43%|████▎     | 2131/5000 [25:53:24<24:58:38, 31.34s/it]
                                                         
{'loss': '0.8281', 'grad_norm': '15.74', 'learning_rate': '1.171e-05', 'epoch': '0.4262'}

 43%|████▎     | 2131/5000 [25:53:24<24:58:38, 31.34s/it][2026-05-22 17:59:12,127] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 540])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 80])
[2026-05-22 17:59:12,492] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 575])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 65])

 43%|████▎     | 2132/5000 [25:53:52<24:09:35, 30.33s/it]
                                                         
{'loss': '1.102', 'grad_norm': '18.7', 'learning_rate': '1.171e-05', 'epoch': '0.4264'}

 43%|████▎     | 2132/5000 [25:53:52<24:09:35, 30.33s/it][2026-05-22 17:59:40,164] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 866])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 53])
[2026-05-22 17:59:41,317] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 790])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 68])

 43%|████▎     | 2133/5000 [25:54:27<25:14:01, 31.69s/it]
                                                         
{'loss': '0.2589', 'grad_norm': '6.933', 'learning_rate': '1.17e-05', 'epoch': '0.4266'}

 43%|████▎     | 2133/5000 [25:54:27<25:14:01, 31.69s/it][2026-05-22 18:00:14,520] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 76])
[2026-05-22 18:00:16,213] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 628])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 73])

 43%|████▎     | 2134/5000 [25:55:00<25:34:08, 32.12s/it]
                                                         
{'loss': '0.01932', 'grad_norm': '2.625', 'learning_rate': '1.17e-05', 'epoch': '0.4268'}

 43%|████▎     | 2134/5000 [25:55:00<25:34:08, 32.12s/it][2026-05-22 18:00:47,599] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 82])
[2026-05-22 18:00:48,555] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 932])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 85])

 43%|████▎     | 2135/5000 [25:55:27<24:16:32, 30.50s/it]
                                                         
{'loss': '0.7629', 'grad_norm': '15.79', 'learning_rate': '1.169e-05', 'epoch': '0.427'}

 43%|████▎     | 2135/5000 [25:55:27<24:16:32, 30.50s/it][2026-05-22 18:01:14,848] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 56])
[2026-05-22 18:01:15,204] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])

 43%|████▎     | 2136/5000 [25:55:55<23:34:31, 29.63s/it]
                                                         
{'loss': '0.9717', 'grad_norm': '18.15', 'learning_rate': '1.169e-05', 'epoch': '0.4272'}

 43%|████▎     | 2136/5000 [25:55:55<23:34:31, 29.63s/it][2026-05-22 18:01:42,461] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 483])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 294])
[2026-05-22 18:01:43,041] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 951])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 293])

 43%|████▎     | 2137/5000 [25:56:38<26:42:54, 33.59s/it]
                                                         
{'loss': '0.1653', 'grad_norm': '7.806', 'learning_rate': '1.169e-05', 'epoch': '0.4274'}

 43%|████▎     | 2137/5000 [25:56:38<26:42:54, 33.59s/it][2026-05-22 18:02:25,988] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 680])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 276])
[2026-05-22 18:02:26,101] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 340])

 43%|████▎     | 2138/5000 [25:57:07<25:49:39, 32.49s/it]
                                                         
{'loss': '0.2569', 'grad_norm': '9.502', 'learning_rate': '1.168e-05', 'epoch': '0.4276'}

 43%|████▎     | 2138/5000 [25:57:07<25:49:39, 32.49s/it][2026-05-22 18:02:54,676] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 417])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 94])
[2026-05-22 18:02:56,296] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 67])

 43%|████▎     | 2139/5000 [25:57:47<27:23:31, 34.47s/it]
                                                         
{'loss': '0.1962', 'grad_norm': '8.127', 'learning_rate': '1.168e-05', 'epoch': '0.4278'}

 43%|████▎     | 2139/5000 [25:57:47<27:23:31, 34.47s/it][2026-05-22 18:03:33,738] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 199])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 18:03:35,274] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 98])

 43%|████▎     | 2140/5000 [25:58:30<29:25:25, 37.04s/it]
                                                         
{'loss': '0.167', 'grad_norm': '8.045', 'learning_rate': '1.167e-05', 'epoch': '0.428'}

 43%|████▎     | 2140/5000 [25:58:30<29:25:25, 37.04s/it][2026-05-22 18:04:16,789] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 79])
[2026-05-22 18:04:18,872] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 498])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 76])

 43%|████▎     | 2141/5000 [25:59:14<31:08:21, 39.21s/it]
                                                         
{'loss': '0.2482', 'grad_norm': '10.06', 'learning_rate': '1.167e-05', 'epoch': '0.4282'}

 43%|████▎     | 2141/5000 [25:59:14<31:08:21, 39.21s/it][2026-05-22 18:05:01,523] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 365])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 65])
[2026-05-22 18:05:03,125] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 502])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 65])

 43%|████▎     | 2142/5000 [25:59:59<32:27:06, 40.88s/it]
                                                         
{'loss': '0.2429', 'grad_norm': '9.507', 'learning_rate': '1.167e-05', 'epoch': '0.4284'}

 43%|████▎     | 2142/5000 [25:59:59<32:27:06, 40.88s/it][2026-05-22 18:05:46,103] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 18:05:46,435] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 416])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 43%|████▎     | 2143/5000 [26:00:42<33:01:59, 41.62s/it]
                                                         
{'loss': '0.3845', 'grad_norm': '10.01', 'learning_rate': '1.166e-05', 'epoch': '0.4286'}

 43%|████▎     | 2143/5000 [26:00:42<33:01:59, 41.62s/it][2026-05-22 18:06:29,463] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 18:06:29,519] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 43%|████▎     | 2144/5000 [26:01:17<31:24:15, 39.59s/it]
                                                         
{'loss': '0.1473', 'grad_norm': '7.587', 'learning_rate': '1.166e-05', 'epoch': '0.4288'}

 43%|████▎     | 2144/5000 [26:01:17<31:24:15, 39.59s/it][2026-05-22 18:07:04,003] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 295])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 18:07:04,839] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 412])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 43%|████▎     | 2145/5000 [26:01:45<28:40:50, 36.16s/it]
                                                         
{'loss': '0.07233', 'grad_norm': '6.933', 'learning_rate': '1.165e-05', 'epoch': '0.429'}

 43%|████▎     | 2145/5000 [26:01:45<28:40:50, 36.16s/it][2026-05-22 18:07:33,226] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 18:07:33,247] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 411])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 43%|████▎     | 2146/5000 [26:02:14<26:56:45, 33.99s/it]
                                                         
{'loss': '0.08824', 'grad_norm': '5.069', 'learning_rate': '1.165e-05', 'epoch': '0.4292'}

 43%|████▎     | 2146/5000 [26:02:14<26:56:45, 33.99s/it][2026-05-22 18:08:01,896] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 424])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 261])
[2026-05-22 18:08:02,507] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 349])

 43%|████▎     | 2147/5000 [26:02:43<25:44:47, 32.49s/it]
                                                         
{'loss': '0.05708', 'grad_norm': '5.223', 'learning_rate': '1.164e-05', 'epoch': '0.4294'}

 43%|████▎     | 2147/5000 [26:02:43<25:44:47, 32.49s/it][2026-05-22 18:08:30,663] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 240])
[2026-05-22 18:08:30,746] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 298])

 43%|████▎     | 2148/5000 [26:03:12<24:49:57, 31.35s/it]
                                                         
{'loss': '0.0776', 'grad_norm': '6.299', 'learning_rate': '1.164e-05', 'epoch': '0.4296'}

 43%|████▎     | 2148/5000 [26:03:12<24:49:57, 31.35s/it][2026-05-22 18:08:58,871] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 274])
[2026-05-22 18:08:59,494] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 494])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 43%|████▎     | 2149/5000 [26:03:42<24:41:17, 31.17s/it]
                                                         
{'loss': '0.7914', 'grad_norm': '17.79', 'learning_rate': '1.164e-05', 'epoch': '0.4298'}

 43%|████▎     | 2149/5000 [26:03:42<24:41:17, 31.17s/it][2026-05-22 18:09:30,768] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 18:09:30,888] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 43%|████▎     | 2150/5000 [26:04:14<24:53:22, 31.44s/it]
                                                         
{'loss': '0.7203', 'grad_norm': '19.03', 'learning_rate': '1.163e-05', 'epoch': '0.43'}

 43%|████▎     | 2150/5000 [26:04:14<24:53:22, 31.44s/it][2026-05-22 18:10:02,371] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 317])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 18:10:03,347] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 509])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 43%|████▎     | 2151/5000 [26:04:45<24:42:46, 31.23s/it]
                                                         
{'loss': '0.3445', 'grad_norm': '10.33', 'learning_rate': '1.163e-05', 'epoch': '0.4302'}

 43%|████▎     | 2151/5000 [26:04:45<24:42:46, 31.23s/it][2026-05-22 18:10:32,850] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 286])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 18:10:34,159] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 437])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 43%|████▎     | 2152/5000 [26:05:16<24:38:50, 31.16s/it]
                                                         
{'loss': '0.2071', 'grad_norm': '8.486', 'learning_rate': '1.162e-05', 'epoch': '0.4304'}

 43%|████▎     | 2152/5000 [26:05:16<24:38:50, 31.16s/it][2026-05-22 18:11:03,465] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 232])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 270])
[2026-05-22 18:11:04,551] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 491])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 317])

 43%|████▎     | 2153/5000 [26:05:47<24:36:12, 31.11s/it]
                                                         
{'loss': '0.3537', 'grad_norm': '12.05', 'learning_rate': '1.162e-05', 'epoch': '0.4306'}

 43%|████▎     | 2153/5000 [26:05:47<24:36:12, 31.11s/it][2026-05-22 18:11:36,426] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 434])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])
[2026-05-22 18:11:36,503] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 315])

 43%|████▎     | 2154/5000 [26:06:18<24:24:50, 30.88s/it]
                                                         
{'loss': '0.1953', 'grad_norm': '8.619', 'learning_rate': '1.162e-05', 'epoch': '0.4308'}

 43%|████▎     | 2154/5000 [26:06:18<24:24:50, 30.88s/it][2026-05-22 18:12:04,518] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 136])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 18:12:04,530] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 130])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])

 43%|████▎     | 2155/5000 [26:06:48<24:23:20, 30.86s/it]
                                                         
{'loss': '0.422', 'grad_norm': '11.45', 'learning_rate': '1.161e-05', 'epoch': '0.431'}

 43%|████▎     | 2155/5000 [26:06:48<24:23:20, 30.86s/it][2026-05-22 18:12:34,721] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 131])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 143])
[2026-05-22 18:12:35,401] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 88])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 43%|████▎     | 2156/5000 [26:07:21<24:48:51, 31.41s/it]
                                                         
{'loss': '0.1328', 'grad_norm': '7.65', 'learning_rate': '1.161e-05', 'epoch': '0.4312'}

 43%|████▎     | 2156/5000 [26:07:21<24:48:51, 31.41s/it][2026-05-22 18:13:08,035] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 319])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 176])
[2026-05-22 18:13:09,048] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 461])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 165])

 43%|████▎     | 2157/5000 [26:07:53<24:56:45, 31.59s/it]
                                                         
{'loss': '0.2903', 'grad_norm': '7.201', 'learning_rate': '1.16e-05', 'epoch': '0.4314'}

 43%|████▎     | 2157/5000 [26:07:53<24:56:45, 31.59s/it][2026-05-22 18:13:40,760] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 212])
[2026-05-22 18:13:41,199] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 460])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 276])

 43%|████▎     | 2158/5000 [26:08:25<25:07:14, 31.82s/it]
                                                         
{'loss': '0.01885', 'grad_norm': '2.714', 'learning_rate': '1.16e-05', 'epoch': '0.4316'}

 43%|████▎     | 2158/5000 [26:08:25<25:07:14, 31.82s/it][2026-05-22 18:14:12,878] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 218])
[2026-05-22 18:14:13,483] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 287])

 43%|████▎     | 2159/5000 [26:08:55<24:31:59, 31.09s/it]
                                                         
{'loss': '0.2807', 'grad_norm': '7.254', 'learning_rate': '1.16e-05', 'epoch': '0.4318'}

 43%|████▎     | 2159/5000 [26:08:55<24:31:59, 31.09s/it][2026-05-22 18:14:42,241] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 532])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 306])
[2026-05-22 18:14:42,931] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])

 43%|████▎     | 2160/5000 [26:09:24<24:10:54, 30.65s/it]
                                                         
{'loss': '0.03478', 'grad_norm': '2.742', 'learning_rate': '1.159e-05', 'epoch': '0.432'}

 43%|████▎     | 2160/5000 [26:09:24<24:10:54, 30.65s/it][2026-05-22 18:15:11,379] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 241])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 225])
[2026-05-22 18:15:11,841] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 43%|████▎     | 2161/5000 [26:09:48<22:25:07, 28.43s/it]
                                                         
{'loss': '0.914', 'grad_norm': '16.88', 'learning_rate': '1.159e-05', 'epoch': '0.4322'}

 43%|████▎     | 2161/5000 [26:09:48<22:25:07, 28.43s/it][2026-05-22 18:15:35,140] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 252])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])
[2026-05-22 18:15:35,527] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 339])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 43%|████▎     | 2162/5000 [26:10:11<21:11:10, 26.87s/it]
                                                         
{'loss': '0.8826', 'grad_norm': '17.06', 'learning_rate': '1.158e-05', 'epoch': '0.4324'}

 43%|████▎     | 2162/5000 [26:10:11<21:11:10, 26.87s/it][2026-05-22 18:15:58,277] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 229])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 18:15:58,954] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 278])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 43%|████▎     | 2163/5000 [26:10:39<21:29:52, 27.28s/it]
                                                         
{'loss': '0.1577', 'grad_norm': '6.229', 'learning_rate': '1.158e-05', 'epoch': '0.4326'}

 43%|████▎     | 2163/5000 [26:10:39<21:29:52, 27.28s/it][2026-05-22 18:16:26,426] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 18:16:27,054] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 272])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 505])

 43%|████▎     | 2164/5000 [26:11:09<22:02:30, 27.98s/it]
                                                         
{'loss': '0.06784', 'grad_norm': '4.021', 'learning_rate': '1.158e-05', 'epoch': '0.4328'}

 43%|████▎     | 2164/5000 [26:11:09<22:02:30, 27.98s/it][2026-05-22 18:16:55,711] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 85])
[2026-05-22 18:16:56,776] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 43%|████▎     | 2165/5000 [26:11:38<22:23:40, 28.44s/it]
                                                         
{'loss': '0.4196', 'grad_norm': '7.223', 'learning_rate': '1.157e-05', 'epoch': '0.433'}

 43%|████▎     | 2165/5000 [26:11:38<22:23:40, 28.44s/it][2026-05-22 18:17:25,945] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 201])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 18:17:26,414] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 499])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 43%|████▎     | 2166/5000 [26:12:08<22:48:38, 28.98s/it]
                                                         
{'loss': '0.135', 'grad_norm': '5.458', 'learning_rate': '1.157e-05', 'epoch': '0.4332'}

 43%|████▎     | 2166/5000 [26:12:08<22:48:38, 28.98s/it][2026-05-22 18:17:55,944] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 18:17:56,554] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 477])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 43%|████▎     | 2167/5000 [26:12:38<22:54:13, 29.10s/it]
                                                         
{'loss': '0.4039', 'grad_norm': '7.352', 'learning_rate': '1.156e-05', 'epoch': '0.4334'}

 43%|████▎     | 2167/5000 [26:12:38<22:54:13, 29.10s/it][2026-05-22 18:18:25,294] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 166])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 90])
[2026-05-22 18:18:25,908] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 484])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 172])

 43%|████▎     | 2168/5000 [26:13:08<23:02:09, 29.28s/it]
                                                         
{'loss': '0.1213', 'grad_norm': '5.202', 'learning_rate': '1.156e-05', 'epoch': '0.4336'}

 43%|████▎     | 2168/5000 [26:13:08<23:02:09, 29.28s/it][2026-05-22 18:18:55,204] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 18:18:55,568] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 975])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 43%|████▎     | 2169/5000 [26:13:36<22:43:13, 28.89s/it]
                                                         
{'loss': '0.3921', 'grad_norm': '7.755', 'learning_rate': '1.156e-05', 'epoch': '0.4338'}

 43%|████▎     | 2169/5000 [26:13:36<22:43:13, 28.89s/it][2026-05-22 18:19:23,206] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 989])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 18:19:23,585] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 973])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])

 43%|████▎     | 2170/5000 [26:14:05<22:56:06, 29.18s/it]
                                                         
{'loss': '0.1344', 'grad_norm': '5.249', 'learning_rate': '1.155e-05', 'epoch': '0.434'}

 43%|████▎     | 2170/5000 [26:14:05<22:56:06, 29.18s/it][2026-05-22 18:19:57,104] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 21])
[2026-05-22 18:19:57,355] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 19])

 43%|████▎     | 2171/5000 [26:14:35<23:05:33, 29.39s/it]
                                                         
{'loss': '0.162', 'grad_norm': '5.856', 'learning_rate': '1.155e-05', 'epoch': '0.4342'}

 43%|████▎     | 2171/5000 [26:14:35<23:05:33, 29.39s/it][2026-05-22 18:20:27,227] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 21])
[2026-05-22 18:20:27,461] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 982])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 19])

 43%|████▎     | 2172/5000 [26:15:07<23:33:53, 30.00s/it]
                                                         
{'loss': '0.1369', 'grad_norm': '5.981', 'learning_rate': '1.154e-05', 'epoch': '0.4344'}

 43%|████▎     | 2172/5000 [26:15:07<23:33:53, 30.00s/it][2026-05-22 18:20:54,027] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 420])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 227])
[2026-05-22 18:20:54,463] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 451])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 240])

 43%|████▎     | 2173/5000 [26:15:35<23:05:57, 29.42s/it]
                                                         
{'loss': '0.2281', 'grad_norm': '4.401', 'learning_rate': '1.154e-05', 'epoch': '0.4346'}

 43%|████▎     | 2173/5000 [26:15:35<23:05:57, 29.42s/it][2026-05-22 18:21:22,605] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 242])
[2026-05-22 18:21:22,634] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 214])

 43%|████▎     | 2174/5000 [26:16:04<22:57:55, 29.26s/it]
                                                         
{'loss': '0.1869', 'grad_norm': '4.749', 'learning_rate': '1.153e-05', 'epoch': '0.4348'}

 43%|████▎     | 2174/5000 [26:16:04<22:57:55, 29.26s/it][2026-05-22 18:21:51,392] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 294])
[2026-05-22 18:21:51,757] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 44%|████▎     | 2175/5000 [26:16:52<27:20:54, 34.85s/it]
                                                         
{'loss': '0.3022', 'grad_norm': '11.09', 'learning_rate': '1.153e-05', 'epoch': '0.435'}

 44%|████▎     | 2175/5000 [26:16:52<27:20:54, 34.85s/it][2026-05-22 18:22:39,667] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 18:22:39,670] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 44%|████▎     | 2176/5000 [26:17:39<30:22:39, 38.73s/it]
                                                         
{'loss': '0.2101', 'grad_norm': '9.109', 'learning_rate': '1.153e-05', 'epoch': '0.4352'}

 44%|████▎     | 2176/5000 [26:17:39<30:22:39, 38.73s/it][2026-05-22 18:23:26,335] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 289])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 18:23:27,632] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 423])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 44%|████▎     | 2177/5000 [26:18:25<32:06:49, 40.95s/it]
                                                         
{'loss': '0.3968', 'grad_norm': '13.32', 'learning_rate': '1.152e-05', 'epoch': '0.4354'}

 44%|████▎     | 2177/5000 [26:18:25<32:06:49, 40.95s/it][2026-05-22 18:24:13,213] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 346])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 18:24:13,243] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 424])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 465])

 44%|████▎     | 2178/5000 [26:19:12<33:19:28, 42.51s/it]
                                                         
{'loss': '0.3724', 'grad_norm': '12.33', 'learning_rate': '1.152e-05', 'epoch': '0.4356'}

 44%|████▎     | 2178/5000 [26:19:12<33:19:28, 42.51s/it][2026-05-22 18:24:59,295] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 454])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])
[2026-05-22 18:25:00,038] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 78])

 44%|████▎     | 2179/5000 [26:19:41<30:12:49, 38.56s/it]
                                                         
{'loss': '0.51', 'grad_norm': '11.36', 'learning_rate': '1.151e-05', 'epoch': '0.4358'}

 44%|████▎     | 2179/5000 [26:19:41<30:12:49, 38.56s/it][2026-05-22 18:25:29,191] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 968])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 67])
[2026-05-22 18:25:30,135] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 984])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 59])

 44%|████▎     | 2180/5000 [26:20:11<28:17:56, 36.13s/it]
                                                         
{'loss': '0.221', 'grad_norm': '9.353', 'learning_rate': '1.151e-05', 'epoch': '0.436'}

 44%|████▎     | 2180/5000 [26:20:11<28:17:56, 36.13s/it][2026-05-22 18:25:58,757] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 485])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 18:25:59,974] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 464])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 44%|████▎     | 2181/5000 [26:20:42<27:06:36, 34.62s/it]
                                                         
{'loss': '1.279', 'grad_norm': '28.46', 'learning_rate': '1.151e-05', 'epoch': '0.4362'}

 44%|████▎     | 2181/5000 [26:20:42<27:06:36, 34.62s/it][2026-05-22 18:26:30,411] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 994])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 18:26:30,561] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 946])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 44%|████▎     | 2182/5000 [26:21:14<26:18:34, 33.61s/it]
                                                         
{'loss': '2.246', 'grad_norm': '25.61', 'learning_rate': '1.15e-05', 'epoch': '0.4364'}

 44%|████▎     | 2182/5000 [26:21:14<26:18:34, 33.61s/it][2026-05-22 18:27:00,752] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 277])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 386])
[2026-05-22 18:27:00,939] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 692])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 44%|████▎     | 2183/5000 [26:21:44<25:24:11, 32.46s/it]
                                                         
{'loss': '0.3083', 'grad_norm': '7.935', 'learning_rate': '1.15e-05', 'epoch': '0.4366'}

 44%|████▎     | 2183/5000 [26:21:44<25:24:11, 32.46s/it][2026-05-22 18:27:31,167] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 274])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 18:27:31,216] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 786])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 44%|████▎     | 2184/5000 [26:22:15<25:09:18, 32.16s/it]
                                                         
{'loss': '0.04945', 'grad_norm': '4.213', 'learning_rate': '1.149e-05', 'epoch': '0.4368'}

 44%|████▎     | 2184/5000 [26:22:15<25:09:18, 32.16s/it][2026-05-22 18:28:02,922] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 18:28:03,013] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 379])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 44%|████▎     | 2185/5000 [26:22:44<24:19:49, 31.12s/it]
                                                         
{'loss': '0.3597', 'grad_norm': '14.03', 'learning_rate': '1.149e-05', 'epoch': '0.437'}

 44%|████▎     | 2185/5000 [26:22:44<24:19:49, 31.12s/it][2026-05-22 18:28:30,988] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 18:28:30,996] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])

 44%|████▎     | 2186/5000 [26:23:28<27:27:19, 35.12s/it]
                                                         
{'loss': '0.1527', 'grad_norm': '6.571', 'learning_rate': '1.149e-05', 'epoch': '0.4372'}

 44%|████▎     | 2186/5000 [26:23:28<27:27:19, 35.12s/it][2026-05-22 18:29:15,234] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 117])
[2026-05-22 18:29:15,394] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 696])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 174])

 44%|████▎     | 2187/5000 [26:24:00<26:34:14, 34.00s/it]
                                                         
{'loss': '0.4093', 'grad_norm': '9.405', 'learning_rate': '1.148e-05', 'epoch': '0.4374'}

 44%|████▎     | 2187/5000 [26:24:00<26:34:14, 34.00s/it][2026-05-22 18:29:47,121] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 63])
[2026-05-22 18:29:47,315] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 600])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 21])

 44%|████▍     | 2188/5000 [26:24:49<30:09:59, 38.62s/it]
                                                         
{'loss': '0.09067', 'grad_norm': '6.281', 'learning_rate': '1.148e-05', 'epoch': '0.4376'}

 44%|████▍     | 2188/5000 [26:24:49<30:09:59, 38.62s/it][2026-05-22 18:30:37,828] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1000])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 57])
[2026-05-22 18:30:38,105] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 977])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 65])

 44%|████▍     | 2189/5000 [26:25:27<29:56:58, 38.36s/it]
                                                         
{'loss': '0.7913', 'grad_norm': '12.92', 'learning_rate': '1.147e-05', 'epoch': '0.4378'}

 44%|████▍     | 2189/5000 [26:25:27<29:56:58, 38.36s/it][2026-05-22 18:31:14,903] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 987])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 33])
[2026-05-22 18:31:15,070] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 961])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 174])

 44%|████▍     | 2190/5000 [26:26:08<30:31:23, 39.10s/it]
                                                         
{'loss': '0.7751', 'grad_norm': '10.32', 'learning_rate': '1.147e-05', 'epoch': '0.438'}

 44%|████▍     | 2190/5000 [26:26:08<30:31:23, 39.10s/it][2026-05-22 18:31:57,976] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 28])
[2026-05-22 18:31:58,161] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1004])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 30])

 44%|████▍     | 2191/5000 [26:26:40<28:52:32, 37.01s/it]
                                                         
{'loss': '1.516', 'grad_norm': '24.75', 'learning_rate': '1.147e-05', 'epoch': '0.4382'}

 44%|████▍     | 2191/5000 [26:26:40<28:52:32, 37.01s/it][2026-05-22 18:32:31,129] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 997])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 23])
[2026-05-22 18:32:32,923] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 28])

 44%|████▍     | 2192/5000 [26:27:13<28:00:17, 35.90s/it]
                                                         
{'loss': '2.21', 'grad_norm': '23.49', 'learning_rate': '1.146e-05', 'epoch': '0.4384'}

 44%|████▍     | 2192/5000 [26:27:13<28:00:17, 35.90s/it][2026-05-22 18:33:00,057] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 690])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 145])
[2026-05-22 18:33:00,143] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 44%|████▍     | 2193/5000 [26:27:46<27:18:39, 35.03s/it]
                                                         
{'loss': '1.6', 'grad_norm': '18.28', 'learning_rate': '1.146e-05', 'epoch': '0.4386'}

 44%|████▍     | 2193/5000 [26:27:46<27:18:39, 35.03s/it][2026-05-22 18:33:32,612] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 721])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 18:33:32,689] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 44%|████▍     | 2194/5000 [26:28:16<26:05:29, 33.47s/it]
                                                         
{'loss': '2.09', 'grad_norm': '21.97', 'learning_rate': '1.145e-05', 'epoch': '0.4388'}

 44%|████▍     | 2194/5000 [26:28:16<26:05:29, 33.47s/it][2026-05-22 18:34:03,375] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 644])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 18:34:03,838] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 644])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 44%|████▍     | 2195/5000 [26:29:02<28:57:55, 37.17s/it]
                                                         
{'loss': '2.559', 'grad_norm': '15.29', 'learning_rate': '1.145e-05', 'epoch': '0.439'}

 44%|████▍     | 2195/5000 [26:29:02<28:57:55, 37.17s/it][2026-05-22 18:34:50,178] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 651])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 18:34:50,238] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 646])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 44%|████▍     | 2196/5000 [26:29:47<30:57:20, 39.74s/it]
                                                         
{'loss': '2.993', 'grad_norm': '18.22', 'learning_rate': '1.144e-05', 'epoch': '0.4392'}

 44%|████▍     | 2196/5000 [26:29:47<30:57:20, 39.74s/it][2026-05-22 18:35:39,542] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 257])
[2026-05-22 18:35:39,642] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 158])

 44%|████▍     | 2197/5000 [26:30:37<33:15:09, 42.71s/it]
                                                         
{'loss': '0.6943', 'grad_norm': '18.49', 'learning_rate': '1.144e-05', 'epoch': '0.4394'}

 44%|████▍     | 2197/5000 [26:30:37<33:15:09, 42.71s/it][2026-05-22 18:36:28,782] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1002])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])
[2026-05-22 18:36:29,749] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 191])

 44%|████▍     | 2198/5000 [26:31:28<35:10:27, 45.19s/it]
                                                         
{'loss': '0.6023', 'grad_norm': '15.76', 'learning_rate': '1.144e-05', 'epoch': '0.4396'}

 44%|████▍     | 2198/5000 [26:31:28<35:10:27, 45.19s/it][2026-05-22 18:37:15,113] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 226])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 260])
[2026-05-22 18:37:16,373] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 502])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 287])

 44%|████▍     | 2199/5000 [26:32:15<35:33:01, 45.69s/it]
                                                         
{'loss': '0.4683', 'grad_norm': '12.08', 'learning_rate': '1.143e-05', 'epoch': '0.4398'}

 44%|████▍     | 2199/5000 [26:32:15<35:33:01, 45.69s/it][2026-05-22 18:38:03,155] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])
[2026-05-22 18:38:03,295] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 441])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])

 44%|████▍     | 2200/5000 [26:33:02<35:50:08, 46.07s/it]
                                                         
{'loss': '0.5493', 'grad_norm': '10.73', 'learning_rate': '1.143e-05', 'epoch': '0.44'}

 44%|████▍     | 2200/5000 [26:33:02<35:50:08, 46.07s/it][2026-05-22 18:38:50,388] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 648])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 173])
[2026-05-22 18:38:50,855] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 653])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 212])

 44%|████▍     | 2201/5000 [26:33:41<34:12:16, 43.99s/it]
                                                         
{'loss': '1.175', 'grad_norm': '9.487', 'learning_rate': '1.142e-05', 'epoch': '0.4402'}

 44%|████▍     | 2201/5000 [26:33:41<34:12:16, 43.99s/it][2026-05-22 18:39:29,111] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 640])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 161])
[2026-05-22 18:39:30,129] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 649])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 207])

 44%|████▍     | 2202/5000 [26:34:21<33:11:12, 42.70s/it]
                                                         
{'loss': '1.108', 'grad_norm': '9.358', 'learning_rate': '1.142e-05', 'epoch': '0.4404'}

 44%|████▍     | 2202/5000 [26:34:21<33:11:12, 42.70s/it][2026-05-22 18:40:07,886] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 378])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 236])
[2026-05-22 18:40:08,315] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 254])

 44%|████▍     | 2203/5000 [26:35:10<34:38:20, 44.58s/it]
                                                         
{'loss': '0.9098', 'grad_norm': '11.98', 'learning_rate': '1.142e-05', 'epoch': '0.4406'}

 44%|████▍     | 2203/5000 [26:35:10<34:38:20, 44.58s/it][2026-05-22 18:40:57,426] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 550])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 236])
[2026-05-22 18:40:58,051] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 255])

 44%|████▍     | 2204/5000 [26:35:58<35:29:04, 45.69s/it]
                                                         
{'loss': '0.7672', 'grad_norm': '9.916', 'learning_rate': '1.141e-05', 'epoch': '0.4408'}

 44%|████▍     | 2204/5000 [26:35:58<35:29:04, 45.69s/it][2026-05-22 18:41:45,758] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 269])
[2026-05-22 18:41:46,140] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 421])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 304])

 44%|████▍     | 2205/5000 [26:36:28<31:45:07, 40.90s/it]
                                                         
{'loss': '0.2027', 'grad_norm': '6.632', 'learning_rate': '1.141e-05', 'epoch': '0.441'}

 44%|████▍     | 2205/5000 [26:36:28<31:45:07, 40.90s/it][2026-05-22 18:42:15,486] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])
[2026-05-22 18:42:16,327] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 274])

 44%|████▍     | 2206/5000 [26:36:57<29:00:18, 37.37s/it]
                                                         
{'loss': '0.0282', 'grad_norm': '3.229', 'learning_rate': '1.14e-05', 'epoch': '0.4412'}

 44%|████▍     | 2206/5000 [26:36:57<29:00:18, 37.37s/it][2026-05-22 18:42:43,889] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 70])
[2026-05-22 18:42:45,593] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 496])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 73])

 44%|████▍     | 2207/5000 [26:37:32<28:37:22, 36.89s/it]
                                                         
{'loss': '2.184', 'grad_norm': '10.87', 'learning_rate': '1.14e-05', 'epoch': '0.4414'}

 44%|████▍     | 2207/5000 [26:37:32<28:37:22, 36.89s/it][2026-05-22 18:43:20,222] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 200])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 97])
[2026-05-22 18:43:21,458] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 503])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 84])

 44%|████▍     | 2208/5000 [26:38:08<28:21:31, 36.57s/it]
                                                         
{'loss': '2.075', 'grad_norm': '9.95', 'learning_rate': '1.14e-05', 'epoch': '0.4416'}

 44%|████▍     | 2208/5000 [26:38:08<28:21:31, 36.57s/it][2026-05-22 18:43:56,119] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 266])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 18:43:56,725] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 477])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])

 44%|████▍     | 2209/5000 [26:38:36<26:22:39, 34.02s/it]
                                                         
{'loss': '1.111', 'grad_norm': '27.79', 'learning_rate': '1.139e-05', 'epoch': '0.4418'}

 44%|████▍     | 2209/5000 [26:38:36<26:22:39, 34.02s/it][2026-05-22 18:44:24,198] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 241])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 18:44:24,731] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 44%|████▍     | 2210/5000 [26:39:07<25:40:48, 33.14s/it]
                                                         
{'loss': '1.438', 'grad_norm': '27.01', 'learning_rate': '1.139e-05', 'epoch': '0.442'}

 44%|████▍     | 2210/5000 [26:39:07<25:40:48, 33.14s/it][2026-05-22 18:44:54,567] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 66])
[2026-05-22 18:44:55,581] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 558])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 61])

 44%|████▍     | 2211/5000 [26:39:37<24:53:49, 32.14s/it]
                                                         
{'loss': '0.6874', 'grad_norm': '15.95', 'learning_rate': '1.138e-05', 'epoch': '0.4422'}

 44%|████▍     | 2211/5000 [26:39:37<24:53:49, 32.14s/it][2026-05-22 18:45:24,515] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 62])
[2026-05-22 18:45:25,583] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 74])

 44%|████▍     | 2212/5000 [26:40:09<24:51:07, 32.09s/it]
                                                         
{'loss': '0.8495', 'grad_norm': '16.51', 'learning_rate': '1.138e-05', 'epoch': '0.4424'}

 44%|████▍     | 2212/5000 [26:40:09<24:51:07, 32.09s/it][2026-05-22 18:45:57,149] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 305])
[2026-05-22 18:45:57,839] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 776])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 44%|████▍     | 2213/5000 [26:40:39<24:24:32, 31.53s/it]
                                                         
{'loss': '0.05962', 'grad_norm': '5.241', 'learning_rate': '1.138e-05', 'epoch': '0.4426'}

 44%|████▍     | 2213/5000 [26:40:39<24:24:32, 31.53s/it][2026-05-22 18:46:27,708] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 840])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])
[2026-05-22 18:46:27,843] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 968])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 44%|████▍     | 2214/5000 [26:41:08<23:47:56, 30.75s/it]
                                                         
{'loss': '0.04046', 'grad_norm': '4.251', 'learning_rate': '1.137e-05', 'epoch': '0.4428'}

 44%|████▍     | 2214/5000 [26:41:08<23:47:56, 30.75s/it][2026-05-22 18:46:55,676] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 792])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 60])
[2026-05-22 18:46:56,936] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 506])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 58])

 44%|████▍     | 2215/5000 [26:41:40<24:02:22, 31.07s/it]
                                                         
{'loss': '0.1678', 'grad_norm': '7.678', 'learning_rate': '1.137e-05', 'epoch': '0.443'}

 44%|████▍     | 2215/5000 [26:41:40<24:02:22, 31.07s/it][2026-05-22 18:47:27,337] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 228])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 82])
[2026-05-22 18:47:28,339] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 693])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 59])

 44%|████▍     | 2216/5000 [26:42:13<24:21:06, 31.49s/it]
                                                         
{'loss': '0.3968', 'grad_norm': '11.27', 'learning_rate': '1.136e-05', 'epoch': '0.4432'}

 44%|████▍     | 2216/5000 [26:42:13<24:21:06, 31.49s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 18:48:00,325] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 220])
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 18:48:00,813] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 498])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 184])

 44%|████▍     | 2217/5000 [26:42:49<25:21:20, 32.80s/it]
                                                         
{'loss': '0.1845', 'grad_norm': '7.928', 'learning_rate': '1.136e-05', 'epoch': '0.4434'}

 44%|████▍     | 2217/5000 [26:42:49<25:21:20, 32.80s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 18:48:36,205] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 273])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 170])
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 18:48:37,222] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 501])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 161])

 44%|████▍     | 2218/5000 [26:43:32<27:47:16, 35.96s/it]
                                                         
{'loss': '0.1442', 'grad_norm': '7.73', 'learning_rate': '1.136e-05', 'epoch': '0.4436'}

 44%|████▍     | 2218/5000 [26:43:32<27:47:16, 35.96s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 18:49:21,432] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 998])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 197])
[2026-05-22 18:49:21,548] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1001])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 204])

 44%|████▍     | 2219/5000 [26:44:15<29:20:49, 37.99s/it]
                                                         
{'loss': '0.3575', 'grad_norm': '9.287', 'learning_rate': '1.135e-05', 'epoch': '0.4438'}

 44%|████▍     | 2219/5000 [26:44:15<29:20:49, 37.99s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 18:50:03,748] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1005])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 236])
[2026-05-22 18:50:03,933] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 983])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 189])

 44%|████▍     | 2220/5000 [26:45:05<32:14:09, 41.74s/it]
                                                         
{'loss': '0.08211', 'grad_norm': '6.437', 'learning_rate': '1.135e-05', 'epoch': '0.444'}

 44%|████▍     | 2220/5000 [26:45:05<32:14:09, 41.74s/it][2026-05-22 18:50:54,076] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 999])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 106])
[2026-05-22 18:50:55,263] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 985])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 77])

 44%|████▍     | 2221/5000 [26:45:42<31:04:59, 40.27s/it]
                                                         
{'loss': '0.4146', 'grad_norm': '9.773', 'learning_rate': '1.134e-05', 'epoch': '0.4442'}

 44%|████▍     | 2221/5000 [26:45:42<31:04:59, 40.27s/it][2026-05-22 18:51:31,571] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 1011])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 74])
[2026-05-22 18:51:33,011] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 87])

 44%|████▍     | 2222/5000 [26:46:16<29:35:32, 38.35s/it]
                                                         
{'loss': '0.3578', 'grad_norm': '9.265', 'learning_rate': '1.134e-05', 'epoch': '0.4444'}

 44%|████▍     | 2222/5000 [26:46:16<29:35:32, 38.35s/it][2026-05-22 18:52:03,349] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 80])
[2026-05-22 18:52:04,263] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 496])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 74])

 44%|████▍     | 2223/5000 [26:46:45<27:31:10, 35.68s/it]
                                                         
{'loss': '0.2717', 'grad_norm': '7.174', 'learning_rate': '1.133e-05', 'epoch': '0.4446'}

 44%|████▍     | 2223/5000 [26:46:46<27:31:10, 35.68s/it][2026-05-22 18:52:32,057] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 169])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])
[2026-05-22 18:52:33,742] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 505])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 71])

 44%|████▍     | 2224/5000 [26:47:16<26:23:24, 34.22s/it]
                                                         
{'loss': '0.2807', 'grad_norm': '9.779', 'learning_rate': '1.133e-05', 'epoch': '0.4448'}

 44%|████▍     | 2224/5000 [26:47:16<26:23:24, 34.22s/it][2026-05-22 18:53:02,992] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 318])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])
[2026-05-22 18:53:03,524] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 327])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 44%|████▍     | 2225/5000 [26:48:03<29:12:57, 37.90s/it]
                                                         
{'loss': '0.1305', 'grad_norm': '6.238', 'learning_rate': '1.133e-05', 'epoch': '0.445'}

 44%|████▍     | 2225/5000 [26:48:03<29:12:57, 37.90s/it][2026-05-22 18:53:50,580] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 338])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])
[2026-05-22 18:53:50,593] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])

 45%|████▍     | 2226/5000 [26:48:49<31:11:40, 40.48s/it]
                                                         
{'loss': '0.1382', 'grad_norm': '6.396', 'learning_rate': '1.132e-05', 'epoch': '0.4452'}

 45%|████▍     | 2226/5000 [26:48:49<31:11:40, 40.48s/it][2026-05-22 18:54:37,008] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 309])
[2026-05-22 18:54:37,249] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 359])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 45%|████▍     | 2227/5000 [26:49:34<32:18:24, 41.94s/it]
                                                         
{'loss': '0.06259', 'grad_norm': '5.02', 'learning_rate': '1.132e-05', 'epoch': '0.4454'}

 45%|████▍     | 2227/5000 [26:49:34<32:18:24, 41.94s/it][2026-05-22 18:55:21,977] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 18:55:22,636] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 418])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 45%|████▍     | 2228/5000 [26:50:21<33:15:41, 43.20s/it]
                                                         
{'loss': '0.1162', 'grad_norm': '4.644', 'learning_rate': '1.131e-05', 'epoch': '0.4456'}

 45%|████▍     | 2228/5000 [26:50:21<33:15:41, 43.20s/it][2026-05-22 18:56:07,658] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 227])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 18:56:08,973] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 498])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 45%|████▍     | 2229/5000 [26:50:49<29:53:56, 38.84s/it]
                                                         
{'loss': '0.05867', 'grad_norm': '3.511', 'learning_rate': '1.131e-05', 'epoch': '0.4458'}

 45%|████▍     | 2229/5000 [26:50:49<29:53:56, 38.84s/it][2026-05-22 18:56:37,197] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 316])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 525])
[2026-05-22 18:56:37,870] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 45%|████▍     | 2230/5000 [26:51:18<27:40:27, 35.97s/it]
                                                         
{'loss': '0.07288', 'grad_norm': '4.289', 'learning_rate': '1.131e-05', 'epoch': '0.446'}

 45%|████▍     | 2230/5000 [26:51:18<27:40:27, 35.97s/it][2026-05-22 18:57:06,321] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 637])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 167])
[2026-05-22 18:57:06,387] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 649])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 135])

 45%|████▍     | 2231/5000 [26:51:49<26:22:56, 34.30s/it]
                                                         
{'loss': '0.2784', 'grad_norm': '7.455', 'learning_rate': '1.13e-05', 'epoch': '0.4462'}

 45%|████▍     | 2231/5000 [26:51:49<26:22:56, 34.30s/it][2026-05-22 18:57:36,088] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 644])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 18:57:36,816] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 642])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 45%|████▍     | 2232/5000 [26:52:21<25:49:20, 33.58s/it]
                                                         
{'loss': '0.1202', 'grad_norm': '5.694', 'learning_rate': '1.13e-05', 'epoch': '0.4464'}

 45%|████▍     | 2232/5000 [26:52:21<25:49:20, 33.58s/it][2026-05-22 18:58:07,790] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 653])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 128])
[2026-05-22 18:58:07,811] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 644])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 121])

 45%|████▍     | 2233/5000 [26:52:51<25:04:29, 32.62s/it]
                                                         
{'loss': '1.691', 'grad_norm': '41.75', 'learning_rate': '1.129e-05', 'epoch': '0.4466'}

 45%|████▍     | 2233/5000 [26:52:51<25:04:29, 32.62s/it][2026-05-22 18:58:38,187] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 641])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 122])
[2026-05-22 18:58:38,262] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 663])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 119])

 45%|████▍     | 2234/5000 [26:53:22<24:38:04, 32.06s/it]
                                                         
{'loss': '2.527', 'grad_norm': '39.08', 'learning_rate': '1.129e-05', 'epoch': '0.4468'}

 45%|████▍     | 2234/5000 [26:53:22<24:38:04, 32.06s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 18:59:08,948] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 147])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 251])
[2026-05-22 18:59:08,994] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 169])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 187])

 45%|████▍     | 2235/5000 [26:53:55<24:49:40, 32.33s/it]
                                                         
{'loss': '0.1348', 'grad_norm': '5.779', 'learning_rate': '1.129e-05', 'epoch': '0.447'}

 45%|████▍     | 2235/5000 [26:53:55<24:49:40, 32.33s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.Warning: text-only sample contains visual placeholders, ignoring visual features.

Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 18:59:41,905] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 313])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 225])
[2026-05-22 18:59:41,930] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 68])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 231])

 45%|████▍     | 2236/5000 [26:54:29<25:08:41, 32.75s/it]
                                                         
{'loss': '0.03615', 'grad_norm': '4.588', 'learning_rate': '1.128e-05', 'epoch': '0.4472'}

 45%|████▍     | 2236/5000 [26:54:29<25:08:41, 32.75s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 19:00:15,696] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 214])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 271])
[2026-05-22 19:00:16,305] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 449])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 445])

 45%|████▍     | 2237/5000 [26:55:02<25:12:27, 32.84s/it]
                                                         
{'loss': '1.837', 'grad_norm': '18.62', 'learning_rate': '1.128e-05', 'epoch': '0.4474'}

 45%|████▍     | 2237/5000 [26:55:02<25:12:27, 32.84s/it]Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
Warning: text-only sample contains visual placeholders, ignoring visual features.
[2026-05-22 19:00:49,289] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 19:00:49,360] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 45%|████▍     | 2238/5000 [26:55:34<25:08:00, 32.76s/it]
                                                         
{'loss': '1.748', 'grad_norm': '17.54', 'learning_rate': '1.127e-05', 'epoch': '0.4476'}

 45%|████▍     | 2238/5000 [26:55:34<25:08:00, 32.76s/it][2026-05-22 19:01:21,564] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 237])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 19:01:21,988] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 250])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 45%|████▍     | 2239/5000 [26:56:06<25:00:00, 32.60s/it]
                                                         
{'loss': '1.379', 'grad_norm': '20.5', 'learning_rate': '1.127e-05', 'epoch': '0.4478'}

 45%|████▍     | 2239/5000 [26:56:06<25:00:00, 32.60s/it][2026-05-22 19:01:54,412] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 252])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])
[2026-05-22 19:01:54,444] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 224])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 485])

 45%|████▍     | 2240/5000 [26:56:39<25:03:03, 32.68s/it]
                                                         
{'loss': '1.313', 'grad_norm': '16.46', 'learning_rate': '1.127e-05', 'epoch': '0.448'}

 45%|████▍     | 2240/5000 [26:56:39<25:03:03, 32.68s/it][2026-05-22 19:02:26,315] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 453])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 205])
[2026-05-22 19:02:27,414] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 457])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 45%|████▍     | 2241/5000 [26:57:03<23:00:19, 30.02s/it]
                                                         
{'loss': '0.8944', 'grad_norm': '16.84', 'learning_rate': '1.126e-05', 'epoch': '0.4482'}

 45%|████▍     | 2241/5000 [26:57:03<23:00:19, 30.02s/it][2026-05-22 19:02:50,662] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 965])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 425])
[2026-05-22 19:02:50,732] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 458])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 405])

 45%|████▍     | 2242/5000 [26:57:28<21:50:17, 28.51s/it]
                                                         
{'loss': '0.9088', 'grad_norm': '19.86', 'learning_rate': '1.126e-05', 'epoch': '0.4484'}

 45%|████▍     | 2242/5000 [26:57:28<21:50:17, 28.51s/it][2026-05-22 19:03:15,419] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 500])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 19])
[2026-05-22 19:03:16,015] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 466])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 20])

 45%|████▍     | 2243/5000 [26:57:59<22:28:43, 29.35s/it]
                                                         
{'loss': '0.4666', 'grad_norm': '7.104', 'learning_rate': '1.125e-05', 'epoch': '0.4486'}

 45%|████▍     | 2243/5000 [26:57:59<22:28:43, 29.35s/it][2026-05-22 19:03:47,371] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 992])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 21])
[2026-05-22 19:03:47,449] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 959])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 27])

 45%|████▍     | 2244/5000 [26:58:30<22:44:32, 29.71s/it]
                                                         
{'loss': '0.02875', 'grad_norm': '2.776', 'learning_rate': '1.125e-05', 'epoch': '0.4488'}

 45%|████▍     | 2244/5000 [26:58:30<22:44:32, 29.71s/it][2026-05-22 19:04:17,011] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 213])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 291])
[2026-05-22 19:04:18,142] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 487])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 287])

 45%|████▍     | 2245/5000 [26:58:57<22:10:07, 28.97s/it]
                                                         
{'loss': '0.3873', 'grad_norm': '7.755', 'learning_rate': '1.124e-05', 'epoch': '0.449'}

 45%|████▍     | 2245/5000 [26:58:57<22:10:07, 28.97s/it][2026-05-22 19:04:44,748] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 273])
[2026-05-22 19:04:44,861] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 297])

 45%|████▍     | 2246/5000 [26:59:26<22:10:47, 28.99s/it]
                                                         
{'loss': '0.1599', 'grad_norm': '5.071', 'learning_rate': '1.124e-05', 'epoch': '0.4492'}

 45%|████▍     | 2246/5000 [26:59:26<22:10:47, 28.99s/it][2026-05-22 19:05:13,414] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 262])
[2026-05-22 19:05:13,784] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 287])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])

 45%|████▍     | 2247/5000 [26:59:57<22:37:48, 29.59s/it]
                                                         
{'loss': '0.2699', 'grad_norm': '7.21', 'learning_rate': '1.124e-05', 'epoch': '0.4494'}

 45%|████▍     | 2247/5000 [26:59:57<22:37:48, 29.59s/it][2026-05-22 19:05:45,146] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 414])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 377])
[2026-05-22 19:05:45,197] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 209])

 45%|████▍     | 2248/5000 [27:00:45<26:48:22, 35.07s/it]
                                                         
{'loss': '0.1132', 'grad_norm': '6.15', 'learning_rate': '1.123e-05', 'epoch': '0.4496'}

 45%|████▍     | 2248/5000 [27:00:45<26:48:22, 35.07s/it][2026-05-22 19:06:32,413] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 317])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 258])
[2026-05-22 19:06:34,454] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 502])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 384])

 45%|████▍     | 2249/5000 [27:01:14<25:18:39, 33.12s/it]
                                                         
{'loss': '0.7578', 'grad_norm': '13.2', 'learning_rate': '1.123e-05', 'epoch': '0.4498'}

 45%|████▍     | 2249/5000 [27:01:14<25:18:39, 33.12s/it][2026-05-22 19:07:01,127] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 507])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 217])
[2026-05-22 19:07:02,266] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 502])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 258])

 45%|████▌     | 2250/5000 [27:01:57<27:43:11, 36.29s/it]
                                                         
{'loss': '0.4506', 'grad_norm': '10.95', 'learning_rate': '1.122e-05', 'epoch': '0.45'}

 45%|████▌     | 2250/5000 [27:01:57<27:43:11, 36.29s/it][2026-05-22 19:07:42,481] INFO [src.utils.basic_utils:19] Saving model to /workspace/VLM2Vec/outputs/Qwen3.5-08b-continued-mixed-alltasks/checkpoint-2250
[2026-05-22 19:07:43,464] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
[2026-05-22 19:07:43,474] INFO [httpx:1025] HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/infinitylogesh/Qwen3-0.8b-image-embeddings-merged-4750/379c6a8d9d7ac04778c52a114dab30ec0f533b3a/config.json "HTTP/1.1 200 OK"
[2026-05-22 19:07:45,651] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 431])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 281])
[2026-05-22 19:07:46,541] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 419])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 277])

 45%|████▌     | 2251/5000 [27:02:28<26:26:37, 34.63s/it]
                                                         
{'loss': '0.4542', 'grad_norm': '9.041', 'learning_rate': '1.122e-05', 'epoch': '0.4502'}

 45%|████▌     | 2251/5000 [27:02:28<26:26:37, 34.63s/it][2026-05-22 19:08:16,673] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 469])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 243])
[2026-05-22 19:08:17,001] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 489])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 229])

 45%|████▌     | 2252/5000 [27:02:57<25:07:54, 32.92s/it]
                                                         
{'loss': '0.1393', 'grad_norm': '5.339', 'learning_rate': '1.122e-05', 'epoch': '0.4504'}

 45%|████▌     | 2252/5000 [27:02:57<25:07:54, 32.92s/it][2026-05-22 19:08:44,230] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 249])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 19:08:45,240] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 485])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])

 45%|████▌     | 2253/5000 [27:03:25<24:03:48, 31.54s/it]
                                                         
{'loss': '1.016', 'grad_norm': '13.69', 'learning_rate': '1.121e-05', 'epoch': '0.4506'}

 45%|████▌     | 2253/5000 [27:03:25<24:03:48, 31.54s/it][2026-05-22 19:09:12,543] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 205])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 406])
[2026-05-22 19:09:13,072] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 425])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 367])

 45%|████▌     | 2254/5000 [27:03:55<23:43:09, 31.10s/it]
                                                         
{'loss': '0.7854', 'grad_norm': '11.59', 'learning_rate': '1.121e-05', 'epoch': '0.4508'}

 45%|████▌     | 2254/5000 [27:03:55<23:43:09, 31.10s/it][2026-05-22 19:09:42,008] INFO [src.utils.basic_utils:10] rank1: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 483])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])
[2026-05-22 19:09:42,599] INFO [src.utils.basic_utils:10] rank0: 		Qry collator: processed_qry_inputs['input_ids'].shape=torch.Size([192, 980])		Pos collator: processed_pos_inputs['input_ids'].shape=torch.Size([192, 127])