Text Generation
Transformers
Safetensors
qwen3_next
qwen3
Mixture of Experts
int2
quantized
autoround
2-bit precision
conversational
2-bit
auto-round
Instructions to use YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound") model = AutoModelForCausalLM.from_pretrained("YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound
- SGLang
How to use YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound with Docker Model Runner:
docker model run hf.co/YCWTG/Qwen3-Coder-Next-int2-mixed-AutoRound
| { | |
| "bits": 2, | |
| "data_type": "int", | |
| "group_size": 128, | |
| "sym": true, | |
| "iters": 1000, | |
| "lr": 0.002, | |
| "low_gpu_mem_usage": true, | |
| "minmax_lr": 0.002, | |
| "nsamples": 512, | |
| "autoround_version": "0.10.1", | |
| "quant_method": "auto-round", | |
| "packing_format": "auto_round:auto_gptq", | |
| "extra_config": { | |
| "model.layers.0.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.0.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.0.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.0.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.0.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.1.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.1.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.1.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.1.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.1.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.2.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.2.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.2.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.2.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.2.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.3.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.3.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.3.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.3.self_attn.o_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.3.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.3.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.4.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.4.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.4.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.4.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.4.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.5.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.5.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.5.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.5.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.5.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.6.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.6.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.6.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.6.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.6.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.7.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.7.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.7.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.7.self_attn.o_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.7.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.7.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.8.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.8.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.8.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.8.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.8.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.9.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.9.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.9.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.9.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.9.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.10.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.10.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.10.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.10.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.10.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.11.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.11.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.11.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.11.self_attn.o_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.11.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.11.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.12.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.12.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.12.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.12.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.12.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.13.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.13.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.13.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.13.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.13.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.14.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.14.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.14.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.14.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.14.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.15.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.15.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.15.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.15.self_attn.o_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.15.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.15.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.16.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.16.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.16.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.16.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.16.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.17.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.17.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.17.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.17.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.17.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.18.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.18.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.18.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.18.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.18.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.19.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.19.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.19.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.19.self_attn.o_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.19.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.19.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.20.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.20.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.20.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.20.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.20.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.21.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.21.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.21.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.21.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.21.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.22.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.22.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.22.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.22.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.22.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.23.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.23.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.23.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.23.self_attn.o_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.23.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.23.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.24.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.24.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.24.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.24.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.24.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.25.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.25.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.25.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.25.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.25.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.26.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.26.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.26.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.26.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.26.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.27.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.27.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.27.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.27.self_attn.o_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.27.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.27.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.28.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.28.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.28.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.28.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.28.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.29.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.29.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.29.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.29.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.29.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.30.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.30.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.30.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.30.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.30.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.31.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.31.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.31.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.31.self_attn.o_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.31.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.31.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.32.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.32.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.32.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.32.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.32.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.33.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.33.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.33.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.33.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.33.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.34.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.34.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.34.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.34.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.34.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.35.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.35.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.35.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.35.self_attn.o_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.35.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.35.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.36.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.36.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.36.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.36.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.36.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.37.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.37.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.37.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.37.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.37.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.38.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.38.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.38.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.38.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.38.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.39.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.39.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.39.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.39.self_attn.o_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.39.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.39.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.40.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.40.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.40.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.40.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.40.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.41.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.41.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.41.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.41.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.41.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.42.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.42.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.42.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.42.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.42.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.43.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.43.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.43.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.43.self_attn.o_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.43.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.43.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.44.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.44.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.44.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.44.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.44.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.45.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.45.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.45.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.45.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.45.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.46.linear_attn.in_proj_qkvz": { | |
| "bits": 16 | |
| }, | |
| "model.layers.46.linear_attn.in_proj_ba": { | |
| "bits": 16 | |
| }, | |
| "model.layers.46.linear_attn.out_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.46.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.46.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| }, | |
| "model.layers.47.self_attn.q_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.47.self_attn.k_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.47.self_attn.v_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.47.self_attn.o_proj": { | |
| "bits": 16 | |
| }, | |
| "model.layers.47.mlp.gate": { | |
| "bits": 16 | |
| }, | |
| "model.layers.47.mlp.shared_expert_gate": { | |
| "bits": 16, | |
| "data_type": "fp" | |
| } | |
| } | |
| } |