Hugging Face's logo

will702
/
unlimited-ocr-mxfp8-mlx

Image-Text-to-Text
Safetensors
MLX
multilingual
mlx-vlm
unlimited-ocr
quantized
apple-silicon
ocr
vision-language
document-parsing
conversational
8-bit precision
Model card Files
xet
 Community
will702 commited on
Commit
3b9928a
·
verified ·
1 Parent(s): cae9070

Add Unlimited-OCR MXFP8 with model_type=unlimited-ocr for mlx-vlm>=0.6

Browse files
Files changed (9) hide show
  1. README.md +72 -0
  2. chat_template.jinja +1 -0
  3. config.json +120 -0
  4. model.safetensors +3 -0
  5. model.safetensors.index.json +758 -0
  6. processor_config.json +28 -0
  7. special_tokens_map.json +39 -0
  8. tokenizer.json +0 -0
  9. tokenizer_config.json +19 -0
README.md ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ base_model: baidu/Unlimited-OCR
4
+ base_model_relation: quantized
5
+ pipeline_tag: image-text-to-text
6
+ library_name: mlx-vlm
7
+ tags:
8
+ - mlx
9
+ - mlx-vlm
10
+ - quantized
11
+ - apple-silicon
12
+ - unlimited-ocr
13
+ - ocr
14
+ - vision-language
15
+ - document-parsing
16
+ language:
17
+ - multilingual
18
+ ---
19
+
20
+ # Unlimited-OCR MXFP8 (mlx-vlm ≥ 0.6 ready)
21
+
22
+ Drop-in MLX pack of [sahilchachra/unlimited-ocr-mxfp8-mlx](https://huggingface.co/sahilchachra/unlimited-ocr-mxfp8-mlx) with configs fixed for **native** Unlimited-OCR loading on **mlx-vlm 0.6+**.
23
+
24
+ ## What changed vs the upstream MXFP8 pack
25
+
26
+ | File | Upstream (old shim) | This repo |
27
+ |---|---|---|
28
+ | `config.json` `model_type` | `deepseekocr` | **`unlimited-ocr`** |
29
+ | `processor_config.json` `processor_class` | `DeepseekOCRProcessor` | **`UnlimitedOCRHFProcessor`** |
30
+ | `processor_config.json` `sft_format` | `deepseek` | **`unlimitedocr`** |
31
+ | Weights | MXFP8 | **unchanged** |
32
+
33
+ On mlx-vlm 0.6+, the upstream `deepseekocr` shim often generates repetitive garbage. Routing to `unlimited-ocr` restores correct OCR.
34
+
35
+ See [FINDINGS.md](https://github.com/will702/unlimited-ocr-mlx/blob/main/FINDINGS.md) for details and benchmarks.
36
+
37
+ ## Usage
38
+
39
+ ```bash
40
+ pip install -U "mlx-vlm>=0.6.0" pymupdf
41
+ ```
42
+
43
+ ```python
44
+ from mlx_vlm import load, generate
45
+ from mlx_vlm.prompt_utils import apply_chat_template
46
+ from mlx_vlm.utils import load_config
47
+
48
+ model_id = "will702/unlimited-ocr-mxfp8-mlx" # update if your namespace differs
49
+ model, processor = load(model_id)
50
+ config = load_config(model_id)
51
+
52
+ prompt = apply_chat_template(processor, config, "Free OCR.", num_images=1)
53
+ out = generate(
54
+ model, processor,
55
+ prompt=prompt,
56
+ image="page.png",
57
+ max_tokens=4096,
58
+ temperature=0.0,
59
+ repetition_penalty=1.05,
60
+ )
61
+ text = out.text.replace("Ġ", " ").replace("Ċ", "\n")
62
+ print(text)
63
+ ```
64
+
65
+ Prefer prompt **`Free OCR.`** — `document parsing.` often stops immediately on this MLX path.
66
+
67
+ ## Credits
68
+
69
+ - Base model: [baidu/Unlimited-OCR](https://huggingface.co/baidu/Unlimited-OCR) (MIT)
70
+ - MXFP8 quantization: [sahilchachra/unlimited-ocr-mxfp8-mlx](https://huggingface.co/sahilchachra/unlimited-ocr-mxfp8-mlx)
71
+ - Runtime: [mlx-vlm](https://github.com/Blaizzy/mlx-vlm)
72
+ - Config fix + Mac tooling: [will702/unlimited-ocr-mlx](https://github.com/will702/unlimited-ocr-mlx)
chat_template.jinja ADDED
@@ -0,0 +1 @@
 
 
1
+ {% for message in messages %}{% if message['role'] == 'user' %}{% elif message['role'] == 'assistant' %}{% endif %}{{message['content']}} {% endfor %}{% if add_generation_prompt %}{% endif %}
config.json ADDED
@@ -0,0 +1,120 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "UnlimitedOCRForCausalLM"
4
+ ],
5
+ "bos_token_id": 0,
6
+ "candidate_resolutions": [
7
+ [
8
+ 1024,
9
+ 1024
10
+ ]
11
+ ],
12
+ "eos_token_id": 1,
13
+ "first_k_dense_replace": 1,
14
+ "global_view_pos": "head",
15
+ "hidden_size": 1280,
16
+ "intermediate_size": 6848,
17
+ "kv_lora_rank": null,
18
+ "language_config": {
19
+ "architectures": [
20
+ "DeepseekOCRForCausalLM"
21
+ ],
22
+ "bos_token_id": 0,
23
+ "eos_token_id": 1,
24
+ "first_k_dense_replace": 1,
25
+ "hidden_size": 1280,
26
+ "intermediate_size": 6848,
27
+ "kv_lora_rank": null,
28
+ "lm_head": true,
29
+ "max_position_embeddings": 32768,
30
+ "moe_intermediate_size": 896,
31
+ "n_group": 1,
32
+ "n_routed_experts": 64,
33
+ "n_shared_experts": 2,
34
+ "num_attention_heads": 10,
35
+ "num_experts_per_tok": 6,
36
+ "num_hidden_layers": 12,
37
+ "num_key_value_heads": 10,
38
+ "q_lora_rank": null,
39
+ "qk_nope_head_dim": 0,
40
+ "qk_rope_head_dim": 0,
41
+ "rm_head": false,
42
+ "topk_group": 1,
43
+ "topk_method": "greedy",
44
+ "torch_dtype": "bfloat16",
45
+ "use_mla": false,
46
+ "v_head_dim": 128,
47
+ "vocab_size": 129280,
48
+ "sliding_window_size": 128
49
+ },
50
+ "lm_head": true,
51
+ "max_position_embeddings": 32768,
52
+ "model_type": "unlimited-ocr",
53
+ "moe_intermediate_size": 896,
54
+ "n_group": 1,
55
+ "n_routed_experts": 64,
56
+ "n_shared_experts": 2,
57
+ "num_attention_heads": 10,
58
+ "num_experts_per_tok": 6,
59
+ "num_hidden_layers": 12,
60
+ "num_key_value_heads": 10,
61
+ "projector_config": {
62
+ "input_dim": 2048,
63
+ "model_type": "mlp_projector",
64
+ "n_embed": 1280,
65
+ "projector_type": "linear"
66
+ },
67
+ "q_lora_rank": null,
68
+ "qk_nope_head_dim": 0,
69
+ "qk_rope_head_dim": 0,
70
+ "quantization": {
71
+ "group_size": 32,
72
+ "bits": 8,
73
+ "mode": "mxfp8"
74
+ },
75
+ "quantization_config": {
76
+ "group_size": 32,
77
+ "bits": 8,
78
+ "mode": "mxfp8"
79
+ },
80
+ "rm_head": false,
81
+ "sliding_window": 128,
82
+ "sliding_window_size": 128,
83
+ "tile_tag": "2D",
84
+ "topk_group": 1,
85
+ "topk_method": "greedy",
86
+ "transformers_version": "4.46.3",
87
+ "use_mla": false,
88
+ "v_head_dim": 128,
89
+ "vision_config": {
90
+ "image_size": 1024,
91
+ "mlp_ratio": 3.7362,
92
+ "model_name": "deeplip_b_l",
93
+ "model_type": "vision",
94
+ "width": {
95
+ "clip-l-14-224": {
96
+ "heads": 16,
97
+ "image_size": 224,
98
+ "layers": 24,
99
+ "patch_size": 14,
100
+ "width": 1024
101
+ },
102
+ "sam_vit_b": {
103
+ "downsample_channels": [
104
+ 512,
105
+ 1024
106
+ ],
107
+ "global_attn_indexes": [
108
+ 2,
109
+ 5,
110
+ 8,
111
+ 11
112
+ ],
113
+ "heads": 12,
114
+ "layers": 12,
115
+ "width": 768
116
+ }
117
+ }
118
+ },
119
+ "vocab_size": 129280
120
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eef3e920d4ef9429ef6353b4b98c34ed8935f678f91ff3bcf7de8ed73a7530db
3
+ size 3827646853
model.safetensors.index.json ADDED
@@ -0,0 +1,758 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 3827553280
4
+ },
5
+ "weight_map": {
6
+ "image_newline": "model.safetensors",
7
+ "language_model.lm_head.scales": "model.safetensors",
8
+ "language_model.lm_head.weight": "model.safetensors",
9
+ "language_model.model.embed_tokens.scales": "model.safetensors",
10
+ "language_model.model.embed_tokens.weight": "model.safetensors",
11
+ "language_model.model.layers.0.input_layernorm.weight": "model.safetensors",
12
+ "language_model.model.layers.0.mlp.down_proj.scales": "model.safetensors",
13
+ "language_model.model.layers.0.mlp.down_proj.weight": "model.safetensors",
14
+ "language_model.model.layers.0.mlp.gate_proj.scales": "model.safetensors",
15
+ "language_model.model.layers.0.mlp.gate_proj.weight": "model.safetensors",
16
+ "language_model.model.layers.0.mlp.up_proj.scales": "model.safetensors",
17
+ "language_model.model.layers.0.mlp.up_proj.weight": "model.safetensors",
18
+ "language_model.model.layers.0.post_attention_layernorm.weight": "model.safetensors",
19
+ "language_model.model.layers.0.self_attn.k_proj.scales": "model.safetensors",
20
+ "language_model.model.layers.0.self_attn.k_proj.weight": "model.safetensors",
21
+ "language_model.model.layers.0.self_attn.o_proj.scales": "model.safetensors",
22
+ "language_model.model.layers.0.self_attn.o_proj.weight": "model.safetensors",
23
+ "language_model.model.layers.0.self_attn.q_proj.scales": "model.safetensors",
24
+ "language_model.model.layers.0.self_attn.q_proj.weight": "model.safetensors",
25
+ "language_model.model.layers.0.self_attn.v_proj.scales": "model.safetensors",
26
+ "language_model.model.layers.0.self_attn.v_proj.weight": "model.safetensors",
27
+ "language_model.model.layers.1.input_layernorm.weight": "model.safetensors",
28
+ "language_model.model.layers.1.mlp.gate.weight": "model.safetensors",
29
+ "language_model.model.layers.1.mlp.shared_experts.down_proj.scales": "model.safetensors",
30
+ "language_model.model.layers.1.mlp.shared_experts.down_proj.weight": "model.safetensors",
31
+ "language_model.model.layers.1.mlp.shared_experts.gate_proj.scales": "model.safetensors",
32
+ "language_model.model.layers.1.mlp.shared_experts.gate_proj.weight": "model.safetensors",
33
+ "language_model.model.layers.1.mlp.shared_experts.up_proj.scales": "model.safetensors",
34
+ "language_model.model.layers.1.mlp.shared_experts.up_proj.weight": "model.safetensors",
35
+ "language_model.model.layers.1.mlp.switch_mlp.down_proj.scales": "model.safetensors",
36
+ "language_model.model.layers.1.mlp.switch_mlp.down_proj.weight": "model.safetensors",
37
+ "language_model.model.layers.1.mlp.switch_mlp.gate_proj.scales": "model.safetensors",
38
+ "language_model.model.layers.1.mlp.switch_mlp.gate_proj.weight": "model.safetensors",
39
+ "language_model.model.layers.1.mlp.switch_mlp.up_proj.scales": "model.safetensors",
40
+ "language_model.model.layers.1.mlp.switch_mlp.up_proj.weight": "model.safetensors",
41
+ "language_model.model.layers.1.post_attention_layernorm.weight": "model.safetensors",
42
+ "language_model.model.layers.1.self_attn.k_proj.scales": "model.safetensors",
43
+ "language_model.model.layers.1.self_attn.k_proj.weight": "model.safetensors",
44
+ "language_model.model.layers.1.self_attn.o_proj.scales": "model.safetensors",
45
+ "language_model.model.layers.1.self_attn.o_proj.weight": "model.safetensors",
46
+ "language_model.model.layers.1.self_attn.q_proj.scales": "model.safetensors",
47
+ "language_model.model.layers.1.self_attn.q_proj.weight": "model.safetensors",
48
+ "language_model.model.layers.1.self_attn.v_proj.scales": "model.safetensors",
49
+ "language_model.model.layers.1.self_attn.v_proj.weight": "model.safetensors",
50
+ "language_model.model.layers.10.input_layernorm.weight": "model.safetensors",
51
+ "language_model.model.layers.10.mlp.gate.weight": "model.safetensors",
52
+ "language_model.model.layers.10.mlp.shared_experts.down_proj.scales": "model.safetensors",
53
+ "language_model.model.layers.10.mlp.shared_experts.down_proj.weight": "model.safetensors",
54
+ "language_model.model.layers.10.mlp.shared_experts.gate_proj.scales": "model.safetensors",
55
+ "language_model.model.layers.10.mlp.shared_experts.gate_proj.weight": "model.safetensors",
56
+ "language_model.model.layers.10.mlp.shared_experts.up_proj.scales": "model.safetensors",
57
+ "language_model.model.layers.10.mlp.shared_experts.up_proj.weight": "model.safetensors",
58
+ "language_model.model.layers.10.mlp.switch_mlp.down_proj.scales": "model.safetensors",
59
+ "language_model.model.layers.10.mlp.switch_mlp.down_proj.weight": "model.safetensors",
60
+ "language_model.model.layers.10.mlp.switch_mlp.gate_proj.scales": "model.safetensors",
61
+ "language_model.model.layers.10.mlp.switch_mlp.gate_proj.weight": "model.safetensors",
62
+ "language_model.model.layers.10.mlp.switch_mlp.up_proj.scales": "model.safetensors",
63
+ "language_model.model.layers.10.mlp.switch_mlp.up_proj.weight": "model.safetensors",
64
+ "language_model.model.layers.10.post_attention_layernorm.weight": "model.safetensors",
65
+ "language_model.model.layers.10.self_attn.k_proj.scales": "model.safetensors",
66
+ "language_model.model.layers.10.self_attn.k_proj.weight": "model.safetensors",
67
+ "language_model.model.layers.10.self_attn.o_proj.scales": "model.safetensors",
68
+ "language_model.model.layers.10.self_attn.o_proj.weight": "model.safetensors",
69
+ "language_model.model.layers.10.self_attn.q_proj.scales": "model.safetensors",
70
+ "language_model.model.layers.10.self_attn.q_proj.weight": "model.safetensors",
71
+ "language_model.model.layers.10.self_attn.v_proj.scales": "model.safetensors",
72
+ "language_model.model.layers.10.self_attn.v_proj.weight": "model.safetensors",
73
+ "language_model.model.layers.11.input_layernorm.weight": "model.safetensors",
74
+ "language_model.model.layers.11.mlp.gate.weight": "model.safetensors",
75
+ "language_model.model.layers.11.mlp.shared_experts.down_proj.scales": "model.safetensors",
76
+ "language_model.model.layers.11.mlp.shared_experts.down_proj.weight": "model.safetensors",
77
+ "language_model.model.layers.11.mlp.shared_experts.gate_proj.scales": "model.safetensors",
78
+ "language_model.model.layers.11.mlp.shared_experts.gate_proj.weight": "model.safetensors",
79
+ "language_model.model.layers.11.mlp.shared_experts.up_proj.scales": "model.safetensors",
80
+ "language_model.model.layers.11.mlp.shared_experts.up_proj.weight": "model.safetensors",
81
+ "language_model.model.layers.11.mlp.switch_mlp.down_proj.scales": "model.safetensors",
82
+ "language_model.model.layers.11.mlp.switch_mlp.down_proj.weight": "model.safetensors",
83
+ "language_model.model.layers.11.mlp.switch_mlp.gate_proj.scales": "model.safetensors",
84
+ "language_model.model.layers.11.mlp.switch_mlp.gate_proj.weight": "model.safetensors",
85
+ "language_model.model.layers.11.mlp.switch_mlp.up_proj.scales": "model.safetensors",
86
+ "language_model.model.layers.11.mlp.switch_mlp.up_proj.weight": "model.safetensors",
87
+ "language_model.model.layers.11.post_attention_layernorm.weight": "model.safetensors",
88
+ "language_model.model.layers.11.self_attn.k_proj.scales": "model.safetensors",
89
+ "language_model.model.layers.11.self_attn.k_proj.weight": "model.safetensors",
90
+ "language_model.model.layers.11.self_attn.o_proj.scales": "model.safetensors",
91
+ "language_model.model.layers.11.self_attn.o_proj.weight": "model.safetensors",
92
+ "language_model.model.layers.11.self_attn.q_proj.scales": "model.safetensors",
93
+ "language_model.model.layers.11.self_attn.q_proj.weight": "model.safetensors",
94
+ "language_model.model.layers.11.self_attn.v_proj.scales": "model.safetensors",
95
+ "language_model.model.layers.11.self_attn.v_proj.weight": "model.safetensors",
96
+ "language_model.model.layers.2.input_layernorm.weight": "model.safetensors",
97
+ "language_model.model.layers.2.mlp.gate.weight": "model.safetensors",
98
+ "language_model.model.layers.2.mlp.shared_experts.down_proj.scales": "model.safetensors",
99
+ "language_model.model.layers.2.mlp.shared_experts.down_proj.weight": "model.safetensors",
100
+ "language_model.model.layers.2.mlp.shared_experts.gate_proj.scales": "model.safetensors",
101
+ "language_model.model.layers.2.mlp.shared_experts.gate_proj.weight": "model.safetensors",
102
+ "language_model.model.layers.2.mlp.shared_experts.up_proj.scales": "model.safetensors",
103
+ "language_model.model.layers.2.mlp.shared_experts.up_proj.weight": "model.safetensors",
104
+ "language_model.model.layers.2.mlp.switch_mlp.down_proj.scales": "model.safetensors",
105
+ "language_model.model.layers.2.mlp.switch_mlp.down_proj.weight": "model.safetensors",
106
+ "language_model.model.layers.2.mlp.switch_mlp.gate_proj.scales": "model.safetensors",
107
+ "language_model.model.layers.2.mlp.switch_mlp.gate_proj.weight": "model.safetensors",
108
+ "language_model.model.layers.2.mlp.switch_mlp.up_proj.scales": "model.safetensors",
109
+ "language_model.model.layers.2.mlp.switch_mlp.up_proj.weight": "model.safetensors",
110
+ "language_model.model.layers.2.post_attention_layernorm.weight": "model.safetensors",
111
+ "language_model.model.layers.2.self_attn.k_proj.scales": "model.safetensors",
112
+ "language_model.model.layers.2.self_attn.k_proj.weight": "model.safetensors",
113
+ "language_model.model.layers.2.self_attn.o_proj.scales": "model.safetensors",
114
+ "language_model.model.layers.2.self_attn.o_proj.weight": "model.safetensors",
115
+ "language_model.model.layers.2.self_attn.q_proj.scales": "model.safetensors",
116
+ "language_model.model.layers.2.self_attn.q_proj.weight": "model.safetensors",
117
+ "language_model.model.layers.2.self_attn.v_proj.scales": "model.safetensors",
118
+ "language_model.model.layers.2.self_attn.v_proj.weight": "model.safetensors",
119
+ "language_model.model.layers.3.input_layernorm.weight": "model.safetensors",
120
+ "language_model.model.layers.3.mlp.gate.weight": "model.safetensors",
121
+ "language_model.model.layers.3.mlp.shared_experts.down_proj.scales": "model.safetensors",
122
+ "language_model.model.layers.3.mlp.shared_experts.down_proj.weight": "model.safetensors",
123
+ "language_model.model.layers.3.mlp.shared_experts.gate_proj.scales": "model.safetensors",
124
+ "language_model.model.layers.3.mlp.shared_experts.gate_proj.weight": "model.safetensors",
125
+ "language_model.model.layers.3.mlp.shared_experts.up_proj.scales": "model.safetensors",
126
+ "language_model.model.layers.3.mlp.shared_experts.up_proj.weight": "model.safetensors",
127
+ "language_model.model.layers.3.mlp.switch_mlp.down_proj.scales": "model.safetensors",
128
+ "language_model.model.layers.3.mlp.switch_mlp.down_proj.weight": "model.safetensors",
129
+ "language_model.model.layers.3.mlp.switch_mlp.gate_proj.scales": "model.safetensors",
130
+ "language_model.model.layers.3.mlp.switch_mlp.gate_proj.weight": "model.safetensors",
131
+ "language_model.model.layers.3.mlp.switch_mlp.up_proj.scales": "model.safetensors",
132
+ "language_model.model.layers.3.mlp.switch_mlp.up_proj.weight": "model.safetensors",
133
+ "language_model.model.layers.3.post_attention_layernorm.weight": "model.safetensors",
134
+ "language_model.model.layers.3.self_attn.k_proj.scales": "model.safetensors",
135
+ "language_model.model.layers.3.self_attn.k_proj.weight": "model.safetensors",
136
+ "language_model.model.layers.3.self_attn.o_proj.scales": "model.safetensors",
137
+ "language_model.model.layers.3.self_attn.o_proj.weight": "model.safetensors",
138
+ "language_model.model.layers.3.self_attn.q_proj.scales": "model.safetensors",
139
+ "language_model.model.layers.3.self_attn.q_proj.weight": "model.safetensors",
140
+ "language_model.model.layers.3.self_attn.v_proj.scales": "model.safetensors",
141
+ "language_model.model.layers.3.self_attn.v_proj.weight": "model.safetensors",
142
+ "language_model.model.layers.4.input_layernorm.weight": "model.safetensors",
143
+ "language_model.model.layers.4.mlp.gate.weight": "model.safetensors",
144
+ "language_model.model.layers.4.mlp.shared_experts.down_proj.scales": "model.safetensors",
145
+ "language_model.model.layers.4.mlp.shared_experts.down_proj.weight": "model.safetensors",
146
+ "language_model.model.layers.4.mlp.shared_experts.gate_proj.scales": "model.safetensors",
147
+ "language_model.model.layers.4.mlp.shared_experts.gate_proj.weight": "model.safetensors",
148
+ "language_model.model.layers.4.mlp.shared_experts.up_proj.scales": "model.safetensors",
149
+ "language_model.model.layers.4.mlp.shared_experts.up_proj.weight": "model.safetensors",
150
+ "language_model.model.layers.4.mlp.switch_mlp.down_proj.scales": "model.safetensors",
151
+ "language_model.model.layers.4.mlp.switch_mlp.down_proj.weight": "model.safetensors",
152
+ "language_model.model.layers.4.mlp.switch_mlp.gate_proj.scales": "model.safetensors",
153
+ "language_model.model.layers.4.mlp.switch_mlp.gate_proj.weight": "model.safetensors",
154
+ "language_model.model.layers.4.mlp.switch_mlp.up_proj.scales": "model.safetensors",
155
+ "language_model.model.layers.4.mlp.switch_mlp.up_proj.weight": "model.safetensors",
156
+ "language_model.model.layers.4.post_attention_layernorm.weight": "model.safetensors",
157
+ "language_model.model.layers.4.self_attn.k_proj.scales": "model.safetensors",
158
+ "language_model.model.layers.4.self_attn.k_proj.weight": "model.safetensors",
159
+ "language_model.model.layers.4.self_attn.o_proj.scales": "model.safetensors",
160
+ "language_model.model.layers.4.self_attn.o_proj.weight": "model.safetensors",
161
+ "language_model.model.layers.4.self_attn.q_proj.scales": "model.safetensors",
162
+ "language_model.model.layers.4.self_attn.q_proj.weight": "model.safetensors",
163
+ "language_model.model.layers.4.self_attn.v_proj.scales": "model.safetensors",
164
+ "language_model.model.layers.4.self_attn.v_proj.weight": "model.safetensors",
165
+ "language_model.model.layers.5.input_layernorm.weight": "model.safetensors",
166
+ "language_model.model.layers.5.mlp.gate.weight": "model.safetensors",
167
+ "language_model.model.layers.5.mlp.shared_experts.down_proj.scales": "model.safetensors",
168
+ "language_model.model.layers.5.mlp.shared_experts.down_proj.weight": "model.safetensors",
169
+ "language_model.model.layers.5.mlp.shared_experts.gate_proj.scales": "model.safetensors",
170
+ "language_model.model.layers.5.mlp.shared_experts.gate_proj.weight": "model.safetensors",
171
+ "language_model.model.layers.5.mlp.shared_experts.up_proj.scales": "model.safetensors",
172
+ "language_model.model.layers.5.mlp.shared_experts.up_proj.weight": "model.safetensors",
173
+ "language_model.model.layers.5.mlp.switch_mlp.down_proj.scales": "model.safetensors",
174
+ "language_model.model.layers.5.mlp.switch_mlp.down_proj.weight": "model.safetensors",
175
+ "language_model.model.layers.5.mlp.switch_mlp.gate_proj.scales": "model.safetensors",
176
+ "language_model.model.layers.5.mlp.switch_mlp.gate_proj.weight": "model.safetensors",
177
+ "language_model.model.layers.5.mlp.switch_mlp.up_proj.scales": "model.safetensors",
178
+ "language_model.model.layers.5.mlp.switch_mlp.up_proj.weight": "model.safetensors",
179
+ "language_model.model.layers.5.post_attention_layernorm.weight": "model.safetensors",
180
+ "language_model.model.layers.5.self_attn.k_proj.scales": "model.safetensors",
181
+ "language_model.model.layers.5.self_attn.k_proj.weight": "model.safetensors",
182
+ "language_model.model.layers.5.self_attn.o_proj.scales": "model.safetensors",
183
+ "language_model.model.layers.5.self_attn.o_proj.weight": "model.safetensors",
184
+ "language_model.model.layers.5.self_attn.q_proj.scales": "model.safetensors",
185
+ "language_model.model.layers.5.self_attn.q_proj.weight": "model.safetensors",
186
+ "language_model.model.layers.5.self_attn.v_proj.scales": "model.safetensors",
187
+ "language_model.model.layers.5.self_attn.v_proj.weight": "model.safetensors",
188
+ "language_model.model.layers.6.input_layernorm.weight": "model.safetensors",
189
+ "language_model.model.layers.6.mlp.gate.weight": "model.safetensors",
190
+ "language_model.model.layers.6.mlp.shared_experts.down_proj.scales": "model.safetensors",
191
+ "language_model.model.layers.6.mlp.shared_experts.down_proj.weight": "model.safetensors",
192
+ "language_model.model.layers.6.mlp.shared_experts.gate_proj.scales": "model.safetensors",
193
+ "language_model.model.layers.6.mlp.shared_experts.gate_proj.weight": "model.safetensors",
194
+ "language_model.model.layers.6.mlp.shared_experts.up_proj.scales": "model.safetensors",
195
+ "language_model.model.layers.6.mlp.shared_experts.up_proj.weight": "model.safetensors",
196
+ "language_model.model.layers.6.mlp.switch_mlp.down_proj.scales": "model.safetensors",
197
+ "language_model.model.layers.6.mlp.switch_mlp.down_proj.weight": "model.safetensors",
198
+ "language_model.model.layers.6.mlp.switch_mlp.gate_proj.scales": "model.safetensors",
199
+ "language_model.model.layers.6.mlp.switch_mlp.gate_proj.weight": "model.safetensors",
200
+ "language_model.model.layers.6.mlp.switch_mlp.up_proj.scales": "model.safetensors",
201
+ "language_model.model.layers.6.mlp.switch_mlp.up_proj.weight": "model.safetensors",
202
+ "language_model.model.layers.6.post_attention_layernorm.weight": "model.safetensors",
203
+ "language_model.model.layers.6.self_attn.k_proj.scales": "model.safetensors",
204
+ "language_model.model.layers.6.self_attn.k_proj.weight": "model.safetensors",
205
+ "language_model.model.layers.6.self_attn.o_proj.scales": "model.safetensors",
206
+ "language_model.model.layers.6.self_attn.o_proj.weight": "model.safetensors",
207
+ "language_model.model.layers.6.self_attn.q_proj.scales": "model.safetensors",
208
+ "language_model.model.layers.6.self_attn.q_proj.weight": "model.safetensors",
209
+ "language_model.model.layers.6.self_attn.v_proj.scales": "model.safetensors",
210
+ "language_model.model.layers.6.self_attn.v_proj.weight": "model.safetensors",
211
+ "language_model.model.layers.7.input_layernorm.weight": "model.safetensors",
212
+ "language_model.model.layers.7.mlp.gate.weight": "model.safetensors",
213
+ "language_model.model.layers.7.mlp.shared_experts.down_proj.scales": "model.safetensors",
214
+ "language_model.model.layers.7.mlp.shared_experts.down_proj.weight": "model.safetensors",
215
+ "language_model.model.layers.7.mlp.shared_experts.gate_proj.scales": "model.safetensors",
216
+ "language_model.model.layers.7.mlp.shared_experts.gate_proj.weight": "model.safetensors",
217
+ "language_model.model.layers.7.mlp.shared_experts.up_proj.scales": "model.safetensors",
218
+ "language_model.model.layers.7.mlp.shared_experts.up_proj.weight": "model.safetensors",
219
+ "language_model.model.layers.7.mlp.switch_mlp.down_proj.scales": "model.safetensors",
220
+ "language_model.model.layers.7.mlp.switch_mlp.down_proj.weight": "model.safetensors",
221
+ "language_model.model.layers.7.mlp.switch_mlp.gate_proj.scales": "model.safetensors",
222
+ "language_model.model.layers.7.mlp.switch_mlp.gate_proj.weight": "model.safetensors",
223
+ "language_model.model.layers.7.mlp.switch_mlp.up_proj.scales": "model.safetensors",
224
+ "language_model.model.layers.7.mlp.switch_mlp.up_proj.weight": "model.safetensors",
225
+ "language_model.model.layers.7.post_attention_layernorm.weight": "model.safetensors",
226
+ "language_model.model.layers.7.self_attn.k_proj.scales": "model.safetensors",
227
+ "language_model.model.layers.7.self_attn.k_proj.weight": "model.safetensors",
228
+ "language_model.model.layers.7.self_attn.o_proj.scales": "model.safetensors",
229
+ "language_model.model.layers.7.self_attn.o_proj.weight": "model.safetensors",
230
+ "language_model.model.layers.7.self_attn.q_proj.scales": "model.safetensors",
231
+ "language_model.model.layers.7.self_attn.q_proj.weight": "model.safetensors",
232
+ "language_model.model.layers.7.self_attn.v_proj.scales": "model.safetensors",
233
+ "language_model.model.layers.7.self_attn.v_proj.weight": "model.safetensors",
234
+ "language_model.model.layers.8.input_layernorm.weight": "model.safetensors",
235
+ "language_model.model.layers.8.mlp.gate.weight": "model.safetensors",
236
+ "language_model.model.layers.8.mlp.shared_experts.down_proj.scales": "model.safetensors",
237
+ "language_model.model.layers.8.mlp.shared_experts.down_proj.weight": "model.safetensors",
238
+ "language_model.model.layers.8.mlp.shared_experts.gate_proj.scales": "model.safetensors",
239
+ "language_model.model.layers.8.mlp.shared_experts.gate_proj.weight": "model.safetensors",
240
+ "language_model.model.layers.8.mlp.shared_experts.up_proj.scales": "model.safetensors",
241
+ "language_model.model.layers.8.mlp.shared_experts.up_proj.weight": "model.safetensors",
242
+ "language_model.model.layers.8.mlp.switch_mlp.down_proj.scales": "model.safetensors",
243
+ "language_model.model.layers.8.mlp.switch_mlp.down_proj.weight": "model.safetensors",
244
+ "language_model.model.layers.8.mlp.switch_mlp.gate_proj.scales": "model.safetensors",
245
+ "language_model.model.layers.8.mlp.switch_mlp.gate_proj.weight": "model.safetensors",
246
+ "language_model.model.layers.8.mlp.switch_mlp.up_proj.scales": "model.safetensors",
247
+ "language_model.model.layers.8.mlp.switch_mlp.up_proj.weight": "model.safetensors",
248
+ "language_model.model.layers.8.post_attention_layernorm.weight": "model.safetensors",
249
+ "language_model.model.layers.8.self_attn.k_proj.scales": "model.safetensors",
250
+ "language_model.model.layers.8.self_attn.k_proj.weight": "model.safetensors",
251
+ "language_model.model.layers.8.self_attn.o_proj.scales": "model.safetensors",
252
+ "language_model.model.layers.8.self_attn.o_proj.weight": "model.safetensors",
253
+ "language_model.model.layers.8.self_attn.q_proj.scales": "model.safetensors",
254
+ "language_model.model.layers.8.self_attn.q_proj.weight": "model.safetensors",
255
+ "language_model.model.layers.8.self_attn.v_proj.scales": "model.safetensors",
256
+ "language_model.model.layers.8.self_attn.v_proj.weight": "model.safetensors",
257
+ "language_model.model.layers.9.input_layernorm.weight": "model.safetensors",
258
+ "language_model.model.layers.9.mlp.gate.weight": "model.safetensors",
259
+ "language_model.model.layers.9.mlp.shared_experts.down_proj.scales": "model.safetensors",
260
+ "language_model.model.layers.9.mlp.shared_experts.down_proj.weight": "model.safetensors",
261
+ "language_model.model.layers.9.mlp.shared_experts.gate_proj.scales": "model.safetensors",
262
+ "language_model.model.layers.9.mlp.shared_experts.gate_proj.weight": "model.safetensors",
263
+ "language_model.model.layers.9.mlp.shared_experts.up_proj.scales": "model.safetensors",
264
+ "language_model.model.layers.9.mlp.shared_experts.up_proj.weight": "model.safetensors",
265
+ "language_model.model.layers.9.mlp.switch_mlp.down_proj.scales": "model.safetensors",
266
+ "language_model.model.layers.9.mlp.switch_mlp.down_proj.weight": "model.safetensors",
267
+ "language_model.model.layers.9.mlp.switch_mlp.gate_proj.scales": "model.safetensors",
268
+ "language_model.model.layers.9.mlp.switch_mlp.gate_proj.weight": "model.safetensors",
269
+ "language_model.model.layers.9.mlp.switch_mlp.up_proj.scales": "model.safetensors",
270
+ "language_model.model.layers.9.mlp.switch_mlp.up_proj.weight": "model.safetensors",
271
+ "language_model.model.layers.9.post_attention_layernorm.weight": "model.safetensors",
272
+ "language_model.model.layers.9.self_attn.k_proj.scales": "model.safetensors",
273
+ "language_model.model.layers.9.self_attn.k_proj.weight": "model.safetensors",
274
+ "language_model.model.layers.9.self_attn.o_proj.scales": "model.safetensors",
275
+ "language_model.model.layers.9.self_attn.o_proj.weight": "model.safetensors",
276
+ "language_model.model.layers.9.self_attn.q_proj.scales": "model.safetensors",
277
+ "language_model.model.layers.9.self_attn.q_proj.weight": "model.safetensors",
278
+ "language_model.model.layers.9.self_attn.v_proj.scales": "model.safetensors",
279
+ "language_model.model.layers.9.self_attn.v_proj.weight": "model.safetensors",
280
+ "language_model.model.norm.weight": "model.safetensors",
281
+ "projector.layers.bias": "model.safetensors",
282
+ "projector.layers.scales": "model.safetensors",
283
+ "projector.layers.weight": "model.safetensors",
284
+ "sam_model.blocks.0.attn.proj.bias": "model.safetensors",
285
+ "sam_model.blocks.0.attn.proj.weight": "model.safetensors",
286
+ "sam_model.blocks.0.attn.qkv.bias": "model.safetensors",
287
+ "sam_model.blocks.0.attn.qkv.weight": "model.safetensors",
288
+ "sam_model.blocks.0.attn.rel_pos_h": "model.safetensors",
289
+ "sam_model.blocks.0.attn.rel_pos_w": "model.safetensors",
290
+ "sam_model.blocks.0.mlp.lin1.bias": "model.safetensors",
291
+ "sam_model.blocks.0.mlp.lin1.weight": "model.safetensors",
292
+ "sam_model.blocks.0.mlp.lin2.bias": "model.safetensors",
293
+ "sam_model.blocks.0.mlp.lin2.weight": "model.safetensors",
294
+ "sam_model.blocks.0.norm1.bias": "model.safetensors",
295
+ "sam_model.blocks.0.norm1.weight": "model.safetensors",
296
+ "sam_model.blocks.0.norm2.bias": "model.safetensors",
297
+ "sam_model.blocks.0.norm2.weight": "model.safetensors",
298
+ "sam_model.blocks.1.attn.proj.bias": "model.safetensors",
299
+ "sam_model.blocks.1.attn.proj.weight": "model.safetensors",
300
+ "sam_model.blocks.1.attn.qkv.bias": "model.safetensors",
301
+ "sam_model.blocks.1.attn.qkv.weight": "model.safetensors",
302
+ "sam_model.blocks.1.attn.rel_pos_h": "model.safetensors",
303
+ "sam_model.blocks.1.attn.rel_pos_w": "model.safetensors",
304
+ "sam_model.blocks.1.mlp.lin1.bias": "model.safetensors",
305
+ "sam_model.blocks.1.mlp.lin1.weight": "model.safetensors",
306
+ "sam_model.blocks.1.mlp.lin2.bias": "model.safetensors",
307
+ "sam_model.blocks.1.mlp.lin2.weight": "model.safetensors",
308
+ "sam_model.blocks.1.norm1.bias": "model.safetensors",
309
+ "sam_model.blocks.1.norm1.weight": "model.safetensors",
310
+ "sam_model.blocks.1.norm2.bias": "model.safetensors",
311
+ "sam_model.blocks.1.norm2.weight": "model.safetensors",
312
+ "sam_model.blocks.10.attn.proj.bias": "model.safetensors",
313
+ "sam_model.blocks.10.attn.proj.weight": "model.safetensors",
314
+ "sam_model.blocks.10.attn.qkv.bias": "model.safetensors",
315
+ "sam_model.blocks.10.attn.qkv.weight": "model.safetensors",
316
+ "sam_model.blocks.10.attn.rel_pos_h": "model.safetensors",
317
+ "sam_model.blocks.10.attn.rel_pos_w": "model.safetensors",
318
+ "sam_model.blocks.10.mlp.lin1.bias": "model.safetensors",
319
+ "sam_model.blocks.10.mlp.lin1.weight": "model.safetensors",
320
+ "sam_model.blocks.10.mlp.lin2.bias": "model.safetensors",
321
+ "sam_model.blocks.10.mlp.lin2.weight": "model.safetensors",
322
+ "sam_model.blocks.10.norm1.bias": "model.safetensors",
323
+ "sam_model.blocks.10.norm1.weight": "model.safetensors",
324
+ "sam_model.blocks.10.norm2.bias": "model.safetensors",
325
+ "sam_model.blocks.10.norm2.weight": "model.safetensors",
326
+ "sam_model.blocks.11.attn.proj.bias": "model.safetensors",
327
+ "sam_model.blocks.11.attn.proj.weight": "model.safetensors",
328
+ "sam_model.blocks.11.attn.qkv.bias": "model.safetensors",
329
+ "sam_model.blocks.11.attn.qkv.weight": "model.safetensors",
330
+ "sam_model.blocks.11.attn.rel_pos_h": "model.safetensors",
331
+ "sam_model.blocks.11.attn.rel_pos_w": "model.safetensors",
332
+ "sam_model.blocks.11.mlp.lin1.bias": "model.safetensors",
333
+ "sam_model.blocks.11.mlp.lin1.weight": "model.safetensors",
334
+ "sam_model.blocks.11.mlp.lin2.bias": "model.safetensors",
335
+ "sam_model.blocks.11.mlp.lin2.weight": "model.safetensors",
336
+ "sam_model.blocks.11.norm1.bias": "model.safetensors",
337
+ "sam_model.blocks.11.norm1.weight": "model.safetensors",
338
+ "sam_model.blocks.11.norm2.bias": "model.safetensors",
339
+ "sam_model.blocks.11.norm2.weight": "model.safetensors",
340
+ "sam_model.blocks.2.attn.proj.bias": "model.safetensors",
341
+ "sam_model.blocks.2.attn.proj.weight": "model.safetensors",
342
+ "sam_model.blocks.2.attn.qkv.bias": "model.safetensors",
343
+ "sam_model.blocks.2.attn.qkv.weight": "model.safetensors",
344
+ "sam_model.blocks.2.attn.rel_pos_h": "model.safetensors",
345
+ "sam_model.blocks.2.attn.rel_pos_w": "model.safetensors",
346
+ "sam_model.blocks.2.mlp.lin1.bias": "model.safetensors",
347
+ "sam_model.blocks.2.mlp.lin1.weight": "model.safetensors",
348
+ "sam_model.blocks.2.mlp.lin2.bias": "model.safetensors",
349
+ "sam_model.blocks.2.mlp.lin2.weight": "model.safetensors",
350
+ "sam_model.blocks.2.norm1.bias": "model.safetensors",
351
+ "sam_model.blocks.2.norm1.weight": "model.safetensors",
352
+ "sam_model.blocks.2.norm2.bias": "model.safetensors",
353
+ "sam_model.blocks.2.norm2.weight": "model.safetensors",
354
+ "sam_model.blocks.3.attn.proj.bias": "model.safetensors",
355
+ "sam_model.blocks.3.attn.proj.weight": "model.safetensors",
356
+ "sam_model.blocks.3.attn.qkv.bias": "model.safetensors",
357
+ "sam_model.blocks.3.attn.qkv.weight": "model.safetensors",
358
+ "sam_model.blocks.3.attn.rel_pos_h": "model.safetensors",
359
+ "sam_model.blocks.3.attn.rel_pos_w": "model.safetensors",
360
+ "sam_model.blocks.3.mlp.lin1.bias": "model.safetensors",
361
+ "sam_model.blocks.3.mlp.lin1.weight": "model.safetensors",
362
+ "sam_model.blocks.3.mlp.lin2.bias": "model.safetensors",
363
+ "sam_model.blocks.3.mlp.lin2.weight": "model.safetensors",
364
+ "sam_model.blocks.3.norm1.bias": "model.safetensors",
365
+ "sam_model.blocks.3.norm1.weight": "model.safetensors",
366
+ "sam_model.blocks.3.norm2.bias": "model.safetensors",
367
+ "sam_model.blocks.3.norm2.weight": "model.safetensors",
368
+ "sam_model.blocks.4.attn.proj.bias": "model.safetensors",
369
+ "sam_model.blocks.4.attn.proj.weight": "model.safetensors",
370
+ "sam_model.blocks.4.attn.qkv.bias": "model.safetensors",
371
+ "sam_model.blocks.4.attn.qkv.weight": "model.safetensors",
372
+ "sam_model.blocks.4.attn.rel_pos_h": "model.safetensors",
373
+ "sam_model.blocks.4.attn.rel_pos_w": "model.safetensors",
374
+ "sam_model.blocks.4.mlp.lin1.bias": "model.safetensors",
375
+ "sam_model.blocks.4.mlp.lin1.weight": "model.safetensors",
376
+ "sam_model.blocks.4.mlp.lin2.bias": "model.safetensors",
377
+ "sam_model.blocks.4.mlp.lin2.weight": "model.safetensors",
378
+ "sam_model.blocks.4.norm1.bias": "model.safetensors",
379
+ "sam_model.blocks.4.norm1.weight": "model.safetensors",
380
+ "sam_model.blocks.4.norm2.bias": "model.safetensors",
381
+ "sam_model.blocks.4.norm2.weight": "model.safetensors",
382
+ "sam_model.blocks.5.attn.proj.bias": "model.safetensors",
383
+ "sam_model.blocks.5.attn.proj.weight": "model.safetensors",
384
+ "sam_model.blocks.5.attn.qkv.bias": "model.safetensors",
385
+ "sam_model.blocks.5.attn.qkv.weight": "model.safetensors",
386
+ "sam_model.blocks.5.attn.rel_pos_h": "model.safetensors",
387
+ "sam_model.blocks.5.attn.rel_pos_w": "model.safetensors",
388
+ "sam_model.blocks.5.mlp.lin1.bias": "model.safetensors",
389
+ "sam_model.blocks.5.mlp.lin1.weight": "model.safetensors",
390
+ "sam_model.blocks.5.mlp.lin2.bias": "model.safetensors",
391
+ "sam_model.blocks.5.mlp.lin2.weight": "model.safetensors",
392
+ "sam_model.blocks.5.norm1.bias": "model.safetensors",
393
+ "sam_model.blocks.5.norm1.weight": "model.safetensors",
394
+ "sam_model.blocks.5.norm2.bias": "model.safetensors",
395
+ "sam_model.blocks.5.norm2.weight": "model.safetensors",
396
+ "sam_model.blocks.6.attn.proj.bias": "model.safetensors",
397
+ "sam_model.blocks.6.attn.proj.weight": "model.safetensors",
398
+ "sam_model.blocks.6.attn.qkv.bias": "model.safetensors",
399
+ "sam_model.blocks.6.attn.qkv.weight": "model.safetensors",
400
+ "sam_model.blocks.6.attn.rel_pos_h": "model.safetensors",
401
+ "sam_model.blocks.6.attn.rel_pos_w": "model.safetensors",
402
+ "sam_model.blocks.6.mlp.lin1.bias": "model.safetensors",
403
+ "sam_model.blocks.6.mlp.lin1.weight": "model.safetensors",
404
+ "sam_model.blocks.6.mlp.lin2.bias": "model.safetensors",
405
+ "sam_model.blocks.6.mlp.lin2.weight": "model.safetensors",
406
+ "sam_model.blocks.6.norm1.bias": "model.safetensors",
407
+ "sam_model.blocks.6.norm1.weight": "model.safetensors",
408
+ "sam_model.blocks.6.norm2.bias": "model.safetensors",
409
+ "sam_model.blocks.6.norm2.weight": "model.safetensors",
410
+ "sam_model.blocks.7.attn.proj.bias": "model.safetensors",
411
+ "sam_model.blocks.7.attn.proj.weight": "model.safetensors",
412
+ "sam_model.blocks.7.attn.qkv.bias": "model.safetensors",
413
+ "sam_model.blocks.7.attn.qkv.weight": "model.safetensors",
414
+ "sam_model.blocks.7.attn.rel_pos_h": "model.safetensors",
415
+ "sam_model.blocks.7.attn.rel_pos_w": "model.safetensors",
416
+ "sam_model.blocks.7.mlp.lin1.bias": "model.safetensors",
417
+ "sam_model.blocks.7.mlp.lin1.weight": "model.safetensors",
418
+ "sam_model.blocks.7.mlp.lin2.bias": "model.safetensors",
419
+ "sam_model.blocks.7.mlp.lin2.weight": "model.safetensors",
420
+ "sam_model.blocks.7.norm1.bias": "model.safetensors",
421
+ "sam_model.blocks.7.norm1.weight": "model.safetensors",
422
+ "sam_model.blocks.7.norm2.bias": "model.safetensors",
423
+ "sam_model.blocks.7.norm2.weight": "model.safetensors",
424
+ "sam_model.blocks.8.attn.proj.bias": "model.safetensors",
425
+ "sam_model.blocks.8.attn.proj.weight": "model.safetensors",
426
+ "sam_model.blocks.8.attn.qkv.bias": "model.safetensors",
427
+ "sam_model.blocks.8.attn.qkv.weight": "model.safetensors",
428
+ "sam_model.blocks.8.attn.rel_pos_h": "model.safetensors",
429
+ "sam_model.blocks.8.attn.rel_pos_w": "model.safetensors",
430
+ "sam_model.blocks.8.mlp.lin1.bias": "model.safetensors",
431
+ "sam_model.blocks.8.mlp.lin1.weight": "model.safetensors",
432
+ "sam_model.blocks.8.mlp.lin2.bias": "model.safetensors",
433
+ "sam_model.blocks.8.mlp.lin2.weight": "model.safetensors",
434
+ "sam_model.blocks.8.norm1.bias": "model.safetensors",
435
+ "sam_model.blocks.8.norm1.weight": "model.safetensors",
436
+ "sam_model.blocks.8.norm2.bias": "model.safetensors",
437
+ "sam_model.blocks.8.norm2.weight": "model.safetensors",
438
+ "sam_model.blocks.9.attn.proj.bias": "model.safetensors",
439
+ "sam_model.blocks.9.attn.proj.weight": "model.safetensors",
440
+ "sam_model.blocks.9.attn.qkv.bias": "model.safetensors",
441
+ "sam_model.blocks.9.attn.qkv.weight": "model.safetensors",
442
+ "sam_model.blocks.9.attn.rel_pos_h": "model.safetensors",
443
+ "sam_model.blocks.9.attn.rel_pos_w": "model.safetensors",
444
+ "sam_model.blocks.9.mlp.lin1.bias": "model.safetensors",
445
+ "sam_model.blocks.9.mlp.lin1.weight": "model.safetensors",
446
+ "sam_model.blocks.9.mlp.lin2.bias": "model.safetensors",
447
+ "sam_model.blocks.9.mlp.lin2.weight": "model.safetensors",
448
+ "sam_model.blocks.9.norm1.bias": "model.safetensors",
449
+ "sam_model.blocks.9.norm1.weight": "model.safetensors",
450
+ "sam_model.blocks.9.norm2.bias": "model.safetensors",
451
+ "sam_model.blocks.9.norm2.weight": "model.safetensors",
452
+ "sam_model.neck.0.weight": "model.safetensors",
453
+ "sam_model.neck.1.bias": "model.safetensors",
454
+ "sam_model.neck.1.weight": "model.safetensors",
455
+ "sam_model.neck.2.weight": "model.safetensors",
456
+ "sam_model.neck.3.bias": "model.safetensors",
457
+ "sam_model.neck.3.weight": "model.safetensors",
458
+ "sam_model.net_2.weight": "model.safetensors",
459
+ "sam_model.net_3.weight": "model.safetensors",
460
+ "sam_model.patch_embed.proj.bias": "model.safetensors",
461
+ "sam_model.patch_embed.proj.weight": "model.safetensors",
462
+ "sam_model.pos_embed": "model.safetensors",
463
+ "view_separator": "model.safetensors",
464
+ "vision_model.embeddings.class_embedding": "model.safetensors",
465
+ "vision_model.embeddings.patch_embedding.weight": "model.safetensors",
466
+ "vision_model.embeddings.position_embedding.weight": "model.safetensors",
467
+ "vision_model.pre_layrnorm.bias": "model.safetensors",
468
+ "vision_model.pre_layrnorm.weight": "model.safetensors",
469
+ "vision_model.transformer.layers.0.layer_norm1.bias": "model.safetensors",
470
+ "vision_model.transformer.layers.0.layer_norm1.weight": "model.safetensors",
471
+ "vision_model.transformer.layers.0.layer_norm2.bias": "model.safetensors",
472
+ "vision_model.transformer.layers.0.layer_norm2.weight": "model.safetensors",
473
+ "vision_model.transformer.layers.0.mlp.fc1.bias": "model.safetensors",
474
+ "vision_model.transformer.layers.0.mlp.fc1.weight": "model.safetensors",
475
+ "vision_model.transformer.layers.0.mlp.fc2.bias": "model.safetensors",
476
+ "vision_model.transformer.layers.0.mlp.fc2.weight": "model.safetensors",
477
+ "vision_model.transformer.layers.0.self_attn.out_proj.bias": "model.safetensors",
478
+ "vision_model.transformer.layers.0.self_attn.out_proj.weight": "model.safetensors",
479
+ "vision_model.transformer.layers.0.self_attn.qkv_proj.bias": "model.safetensors",
480
+ "vision_model.transformer.layers.0.self_attn.qkv_proj.weight": "model.safetensors",
481
+ "vision_model.transformer.layers.1.layer_norm1.bias": "model.safetensors",
482
+ "vision_model.transformer.layers.1.layer_norm1.weight": "model.safetensors",
483
+ "vision_model.transformer.layers.1.layer_norm2.bias": "model.safetensors",
484
+ "vision_model.transformer.layers.1.layer_norm2.weight": "model.safetensors",
485
+ "vision_model.transformer.layers.1.mlp.fc1.bias": "model.safetensors",
486
+ "vision_model.transformer.layers.1.mlp.fc1.weight": "model.safetensors",
487
+ "vision_model.transformer.layers.1.mlp.fc2.bias": "model.safetensors",
488
+ "vision_model.transformer.layers.1.mlp.fc2.weight": "model.safetensors",
489
+ "vision_model.transformer.layers.1.self_attn.out_proj.bias": "model.safetensors",
490
+ "vision_model.transformer.layers.1.self_attn.out_proj.weight": "model.safetensors",
491
+ "vision_model.transformer.layers.1.self_attn.qkv_proj.bias": "model.safetensors",
492
+ "vision_model.transformer.layers.1.self_attn.qkv_proj.weight": "model.safetensors",
493
+ "vision_model.transformer.layers.10.layer_norm1.bias": "model.safetensors",
494
+ "vision_model.transformer.layers.10.layer_norm1.weight": "model.safetensors",
495
+ "vision_model.transformer.layers.10.layer_norm2.bias": "model.safetensors",
496
+ "vision_model.transformer.layers.10.layer_norm2.weight": "model.safetensors",
497
+ "vision_model.transformer.layers.10.mlp.fc1.bias": "model.safetensors",
498
+ "vision_model.transformer.layers.10.mlp.fc1.weight": "model.safetensors",
499
+ "vision_model.transformer.layers.10.mlp.fc2.bias": "model.safetensors",
500
+ "vision_model.transformer.layers.10.mlp.fc2.weight": "model.safetensors",
501
+ "vision_model.transformer.layers.10.self_attn.out_proj.bias": "model.safetensors",
502
+ "vision_model.transformer.layers.10.self_attn.out_proj.weight": "model.safetensors",
503
+ "vision_model.transformer.layers.10.self_attn.qkv_proj.bias": "model.safetensors",
504
+ "vision_model.transformer.layers.10.self_attn.qkv_proj.weight": "model.safetensors",
505
+ "vision_model.transformer.layers.11.layer_norm1.bias": "model.safetensors",
506
+ "vision_model.transformer.layers.11.layer_norm1.weight": "model.safetensors",
507
+ "vision_model.transformer.layers.11.layer_norm2.bias": "model.safetensors",
508
+ "vision_model.transformer.layers.11.layer_norm2.weight": "model.safetensors",
509
+ "vision_model.transformer.layers.11.mlp.fc1.bias": "model.safetensors",
510
+ "vision_model.transformer.layers.11.mlp.fc1.weight": "model.safetensors",
511
+ "vision_model.transformer.layers.11.mlp.fc2.bias": "model.safetensors",
512
+ "vision_model.transformer.layers.11.mlp.fc2.weight": "model.safetensors",
513
+ "vision_model.transformer.layers.11.self_attn.out_proj.bias": "model.safetensors",
514
+ "vision_model.transformer.layers.11.self_attn.out_proj.weight": "model.safetensors",
515
+ "vision_model.transformer.layers.11.self_attn.qkv_proj.bias": "model.safetensors",
516
+ "vision_model.transformer.layers.11.self_attn.qkv_proj.weight": "model.safetensors",
517
+ "vision_model.transformer.layers.12.layer_norm1.bias": "model.safetensors",
518
+ "vision_model.transformer.layers.12.layer_norm1.weight": "model.safetensors",
519
+ "vision_model.transformer.layers.12.layer_norm2.bias": "model.safetensors",
520
+ "vision_model.transformer.layers.12.layer_norm2.weight": "model.safetensors",
521
+ "vision_model.transformer.layers.12.mlp.fc1.bias": "model.safetensors",
522
+ "vision_model.transformer.layers.12.mlp.fc1.weight": "model.safetensors",
523
+ "vision_model.transformer.layers.12.mlp.fc2.bias": "model.safetensors",
524
+ "vision_model.transformer.layers.12.mlp.fc2.weight": "model.safetensors",
525
+ "vision_model.transformer.layers.12.self_attn.out_proj.bias": "model.safetensors",
526
+ "vision_model.transformer.layers.12.self_attn.out_proj.weight": "model.safetensors",
527
+ "vision_model.transformer.layers.12.self_attn.qkv_proj.bias": "model.safetensors",
528
+ "vision_model.transformer.layers.12.self_attn.qkv_proj.weight": "model.safetensors",
529
+ "vision_model.transformer.layers.13.layer_norm1.bias": "model.safetensors",
530
+ "vision_model.transformer.layers.13.layer_norm1.weight": "model.safetensors",
531
+ "vision_model.transformer.layers.13.layer_norm2.bias": "model.safetensors",
532
+ "vision_model.transformer.layers.13.layer_norm2.weight": "model.safetensors",
533
+ "vision_model.transformer.layers.13.mlp.fc1.bias": "model.safetensors",
534
+ "vision_model.transformer.layers.13.mlp.fc1.weight": "model.safetensors",
535
+ "vision_model.transformer.layers.13.mlp.fc2.bias": "model.safetensors",
536
+ "vision_model.transformer.layers.13.mlp.fc2.weight": "model.safetensors",
537
+ "vision_model.transformer.layers.13.self_attn.out_proj.bias": "model.safetensors",
538
+ "vision_model.transformer.layers.13.self_attn.out_proj.weight": "model.safetensors",
539
+ "vision_model.transformer.layers.13.self_attn.qkv_proj.bias": "model.safetensors",
540
+ "vision_model.transformer.layers.13.self_attn.qkv_proj.weight": "model.safetensors",
541
+ "vision_model.transformer.layers.14.layer_norm1.bias": "model.safetensors",
542
+ "vision_model.transformer.layers.14.layer_norm1.weight": "model.safetensors",
543
+ "vision_model.transformer.layers.14.layer_norm2.bias": "model.safetensors",
544
+ "vision_model.transformer.layers.14.layer_norm2.weight": "model.safetensors",
545
+ "vision_model.transformer.layers.14.mlp.fc1.bias": "model.safetensors",
546
+ "vision_model.transformer.layers.14.mlp.fc1.weight": "model.safetensors",
547
+ "vision_model.transformer.layers.14.mlp.fc2.bias": "model.safetensors",
548
+ "vision_model.transformer.layers.14.mlp.fc2.weight": "model.safetensors",
549
+ "vision_model.transformer.layers.14.self_attn.out_proj.bias": "model.safetensors",
550
+ "vision_model.transformer.layers.14.self_attn.out_proj.weight": "model.safetensors",
551
+ "vision_model.transformer.layers.14.self_attn.qkv_proj.bias": "model.safetensors",
552
+ "vision_model.transformer.layers.14.self_attn.qkv_proj.weight": "model.safetensors",
553
+ "vision_model.transformer.layers.15.layer_norm1.bias": "model.safetensors",
554
+ "vision_model.transformer.layers.15.layer_norm1.weight": "model.safetensors",
555
+ "vision_model.transformer.layers.15.layer_norm2.bias": "model.safetensors",
556
+ "vision_model.transformer.layers.15.layer_norm2.weight": "model.safetensors",
557
+ "vision_model.transformer.layers.15.mlp.fc1.bias": "model.safetensors",
558
+ "vision_model.transformer.layers.15.mlp.fc1.weight": "model.safetensors",
559
+ "vision_model.transformer.layers.15.mlp.fc2.bias": "model.safetensors",
560
+ "vision_model.transformer.layers.15.mlp.fc2.weight": "model.safetensors",
561
+ "vision_model.transformer.layers.15.self_attn.out_proj.bias": "model.safetensors",
562
+ "vision_model.transformer.layers.15.self_attn.out_proj.weight": "model.safetensors",
563
+ "vision_model.transformer.layers.15.self_attn.qkv_proj.bias": "model.safetensors",
564
+ "vision_model.transformer.layers.15.self_attn.qkv_proj.weight": "model.safetensors",
565
+ "vision_model.transformer.layers.16.layer_norm1.bias": "model.safetensors",
566
+ "vision_model.transformer.layers.16.layer_norm1.weight": "model.safetensors",
567
+ "vision_model.transformer.layers.16.layer_norm2.bias": "model.safetensors",
568
+ "vision_model.transformer.layers.16.layer_norm2.weight": "model.safetensors",
569
+ "vision_model.transformer.layers.16.mlp.fc1.bias": "model.safetensors",
570
+ "vision_model.transformer.layers.16.mlp.fc1.weight": "model.safetensors",
571
+ "vision_model.transformer.layers.16.mlp.fc2.bias": "model.safetensors",
572
+ "vision_model.transformer.layers.16.mlp.fc2.weight": "model.safetensors",
573
+ "vision_model.transformer.layers.16.self_attn.out_proj.bias": "model.safetensors",
574
+ "vision_model.transformer.layers.16.self_attn.out_proj.weight": "model.safetensors",
575
+ "vision_model.transformer.layers.16.self_attn.qkv_proj.bias": "model.safetensors",
576
+ "vision_model.transformer.layers.16.self_attn.qkv_proj.weight": "model.safetensors",
577
+ "vision_model.transformer.layers.17.layer_norm1.bias": "model.safetensors",
578
+ "vision_model.transformer.layers.17.layer_norm1.weight": "model.safetensors",
579
+ "vision_model.transformer.layers.17.layer_norm2.bias": "model.safetensors",
580
+ "vision_model.transformer.layers.17.layer_norm2.weight": "model.safetensors",
581
+ "vision_model.transformer.layers.17.mlp.fc1.bias": "model.safetensors",
582
+ "vision_model.transformer.layers.17.mlp.fc1.weight": "model.safetensors",
583
+ "vision_model.transformer.layers.17.mlp.fc2.bias": "model.safetensors",
584
+ "vision_model.transformer.layers.17.mlp.fc2.weight": "model.safetensors",
585
+ "vision_model.transformer.layers.17.self_attn.out_proj.bias": "model.safetensors",
586
+ "vision_model.transformer.layers.17.self_attn.out_proj.weight": "model.safetensors",
587
+ "vision_model.transformer.layers.17.self_attn.qkv_proj.bias": "model.safetensors",
588
+ "vision_model.transformer.layers.17.self_attn.qkv_proj.weight": "model.safetensors",
589
+ "vision_model.transformer.layers.18.layer_norm1.bias": "model.safetensors",
590
+ "vision_model.transformer.layers.18.layer_norm1.weight": "model.safetensors",
591
+ "vision_model.transformer.layers.18.layer_norm2.bias": "model.safetensors",
592
+ "vision_model.transformer.layers.18.layer_norm2.weight": "model.safetensors",
593
+ "vision_model.transformer.layers.18.mlp.fc1.bias": "model.safetensors",
594
+ "vision_model.transformer.layers.18.mlp.fc1.weight": "model.safetensors",
595
+ "vision_model.transformer.layers.18.mlp.fc2.bias": "model.safetensors",
596
+ "vision_model.transformer.layers.18.mlp.fc2.weight": "model.safetensors",
597
+ "vision_model.transformer.layers.18.self_attn.out_proj.bias": "model.safetensors",
598
+ "vision_model.transformer.layers.18.self_attn.out_proj.weight": "model.safetensors",
599
+ "vision_model.transformer.layers.18.self_attn.qkv_proj.bias": "model.safetensors",
600
+ "vision_model.transformer.layers.18.self_attn.qkv_proj.weight": "model.safetensors",
601
+ "vision_model.transformer.layers.19.layer_norm1.bias": "model.safetensors",
602
+ "vision_model.transformer.layers.19.layer_norm1.weight": "model.safetensors",
603
+ "vision_model.transformer.layers.19.layer_norm2.bias": "model.safetensors",
604
+ "vision_model.transformer.layers.19.layer_norm2.weight": "model.safetensors",
605
+ "vision_model.transformer.layers.19.mlp.fc1.bias": "model.safetensors",
606
+ "vision_model.transformer.layers.19.mlp.fc1.weight": "model.safetensors",
607
+ "vision_model.transformer.layers.19.mlp.fc2.bias": "model.safetensors",
608
+ "vision_model.transformer.layers.19.mlp.fc2.weight": "model.safetensors",
609
+ "vision_model.transformer.layers.19.self_attn.out_proj.bias": "model.safetensors",
610
+ "vision_model.transformer.layers.19.self_attn.out_proj.weight": "model.safetensors",
611
+ "vision_model.transformer.layers.19.self_attn.qkv_proj.bias": "model.safetensors",
612
+ "vision_model.transformer.layers.19.self_attn.qkv_proj.weight": "model.safetensors",
613
+ "vision_model.transformer.layers.2.layer_norm1.bias": "model.safetensors",
614
+ "vision_model.transformer.layers.2.layer_norm1.weight": "model.safetensors",
615
+ "vision_model.transformer.layers.2.layer_norm2.bias": "model.safetensors",
616
+ "vision_model.transformer.layers.2.layer_norm2.weight": "model.safetensors",
617
+ "vision_model.transformer.layers.2.mlp.fc1.bias": "model.safetensors",
618
+ "vision_model.transformer.layers.2.mlp.fc1.weight": "model.safetensors",
619
+ "vision_model.transformer.layers.2.mlp.fc2.bias": "model.safetensors",
620
+ "vision_model.transformer.layers.2.mlp.fc2.weight": "model.safetensors",
621
+ "vision_model.transformer.layers.2.self_attn.out_proj.bias": "model.safetensors",
622
+ "vision_model.transformer.layers.2.self_attn.out_proj.weight": "model.safetensors",
623
+ "vision_model.transformer.layers.2.self_attn.qkv_proj.bias": "model.safetensors",
624
+ "vision_model.transformer.layers.2.self_attn.qkv_proj.weight": "model.safetensors",
625
+ "vision_model.transformer.layers.20.layer_norm1.bias": "model.safetensors",
626
+ "vision_model.transformer.layers.20.layer_norm1.weight": "model.safetensors",
627
+ "vision_model.transformer.layers.20.layer_norm2.bias": "model.safetensors",
628
+ "vision_model.transformer.layers.20.layer_norm2.weight": "model.safetensors",
629
+ "vision_model.transformer.layers.20.mlp.fc1.bias": "model.safetensors",
630
+ "vision_model.transformer.layers.20.mlp.fc1.weight": "model.safetensors",
631
+ "vision_model.transformer.layers.20.mlp.fc2.bias": "model.safetensors",
632
+ "vision_model.transformer.layers.20.mlp.fc2.weight": "model.safetensors",
633
+ "vision_model.transformer.layers.20.self_attn.out_proj.bias": "model.safetensors",
634
+ "vision_model.transformer.layers.20.self_attn.out_proj.weight": "model.safetensors",
635
+ "vision_model.transformer.layers.20.self_attn.qkv_proj.bias": "model.safetensors",
636
+ "vision_model.transformer.layers.20.self_attn.qkv_proj.weight": "model.safetensors",
637
+ "vision_model.transformer.layers.21.layer_norm1.bias": "model.safetensors",
638
+ "vision_model.transformer.layers.21.layer_norm1.weight": "model.safetensors",
639
+ "vision_model.transformer.layers.21.layer_norm2.bias": "model.safetensors",
640
+ "vision_model.transformer.layers.21.layer_norm2.weight": "model.safetensors",
641
+ "vision_model.transformer.layers.21.mlp.fc1.bias": "model.safetensors",
642
+ "vision_model.transformer.layers.21.mlp.fc1.weight": "model.safetensors",
643
+ "vision_model.transformer.layers.21.mlp.fc2.bias": "model.safetensors",
644
+ "vision_model.transformer.layers.21.mlp.fc2.weight": "model.safetensors",
645
+ "vision_model.transformer.layers.21.self_attn.out_proj.bias": "model.safetensors",
646
+ "vision_model.transformer.layers.21.self_attn.out_proj.weight": "model.safetensors",
647
+ "vision_model.transformer.layers.21.self_attn.qkv_proj.bias": "model.safetensors",
648
+ "vision_model.transformer.layers.21.self_attn.qkv_proj.weight": "model.safetensors",
649
+ "vision_model.transformer.layers.22.layer_norm1.bias": "model.safetensors",
650
+ "vision_model.transformer.layers.22.layer_norm1.weight": "model.safetensors",
651
+ "vision_model.transformer.layers.22.layer_norm2.bias": "model.safetensors",
652
+ "vision_model.transformer.layers.22.layer_norm2.weight": "model.safetensors",
653
+ "vision_model.transformer.layers.22.mlp.fc1.bias": "model.safetensors",
654
+ "vision_model.transformer.layers.22.mlp.fc1.weight": "model.safetensors",
655
+ "vision_model.transformer.layers.22.mlp.fc2.bias": "model.safetensors",
656
+ "vision_model.transformer.layers.22.mlp.fc2.weight": "model.safetensors",
657
+ "vision_model.transformer.layers.22.self_attn.out_proj.bias": "model.safetensors",
658
+ "vision_model.transformer.layers.22.self_attn.out_proj.weight": "model.safetensors",
659
+ "vision_model.transformer.layers.22.self_attn.qkv_proj.bias": "model.safetensors",
660
+ "vision_model.transformer.layers.22.self_attn.qkv_proj.weight": "model.safetensors",
661
+ "vision_model.transformer.layers.23.layer_norm1.bias": "model.safetensors",
662
+ "vision_model.transformer.layers.23.layer_norm1.weight": "model.safetensors",
663
+ "vision_model.transformer.layers.23.layer_norm2.bias": "model.safetensors",
664
+ "vision_model.transformer.layers.23.layer_norm2.weight": "model.safetensors",
665
+ "vision_model.transformer.layers.23.mlp.fc1.bias": "model.safetensors",
666
+ "vision_model.transformer.layers.23.mlp.fc1.weight": "model.safetensors",
667
+ "vision_model.transformer.layers.23.mlp.fc2.bias": "model.safetensors",
668
+ "vision_model.transformer.layers.23.mlp.fc2.weight": "model.safetensors",
669
+ "vision_model.transformer.layers.23.self_attn.out_proj.bias": "model.safetensors",
670
+ "vision_model.transformer.layers.23.self_attn.out_proj.weight": "model.safetensors",
671
+ "vision_model.transformer.layers.23.self_attn.qkv_proj.bias": "model.safetensors",
672
+ "vision_model.transformer.layers.23.self_attn.qkv_proj.weight": "model.safetensors",
673
+ "vision_model.transformer.layers.3.layer_norm1.bias": "model.safetensors",
674
+ "vision_model.transformer.layers.3.layer_norm1.weight": "model.safetensors",
675
+ "vision_model.transformer.layers.3.layer_norm2.bias": "model.safetensors",
676
+ "vision_model.transformer.layers.3.layer_norm2.weight": "model.safetensors",
677
+ "vision_model.transformer.layers.3.mlp.fc1.bias": "model.safetensors",
678
+ "vision_model.transformer.layers.3.mlp.fc1.weight": "model.safetensors",
679
+ "vision_model.transformer.layers.3.mlp.fc2.bias": "model.safetensors",
680
+ "vision_model.transformer.layers.3.mlp.fc2.weight": "model.safetensors",
681
+ "vision_model.transformer.layers.3.self_attn.out_proj.bias": "model.safetensors",
682
+ "vision_model.transformer.layers.3.self_attn.out_proj.weight": "model.safetensors",
683
+ "vision_model.transformer.layers.3.self_attn.qkv_proj.bias": "model.safetensors",
684
+ "vision_model.transformer.layers.3.self_attn.qkv_proj.weight": "model.safetensors",
685
+ "vision_model.transformer.layers.4.layer_norm1.bias": "model.safetensors",
686
+ "vision_model.transformer.layers.4.layer_norm1.weight": "model.safetensors",
687
+ "vision_model.transformer.layers.4.layer_norm2.bias": "model.safetensors",
688
+ "vision_model.transformer.layers.4.layer_norm2.weight": "model.safetensors",
689
+ "vision_model.transformer.layers.4.mlp.fc1.bias": "model.safetensors",
690
+ "vision_model.transformer.layers.4.mlp.fc1.weight": "model.safetensors",
691
+ "vision_model.transformer.layers.4.mlp.fc2.bias": "model.safetensors",
692
+ "vision_model.transformer.layers.4.mlp.fc2.weight": "model.safetensors",
693
+ "vision_model.transformer.layers.4.self_attn.out_proj.bias": "model.safetensors",
694
+ "vision_model.transformer.layers.4.self_attn.out_proj.weight": "model.safetensors",
695
+ "vision_model.transformer.layers.4.self_attn.qkv_proj.bias": "model.safetensors",
696
+ "vision_model.transformer.layers.4.self_attn.qkv_proj.weight": "model.safetensors",
697
+ "vision_model.transformer.layers.5.layer_norm1.bias": "model.safetensors",
698
+ "vision_model.transformer.layers.5.layer_norm1.weight": "model.safetensors",
699
+ "vision_model.transformer.layers.5.layer_norm2.bias": "model.safetensors",
700
+ "vision_model.transformer.layers.5.layer_norm2.weight": "model.safetensors",
701
+ "vision_model.transformer.layers.5.mlp.fc1.bias": "model.safetensors",
702
+ "vision_model.transformer.layers.5.mlp.fc1.weight": "model.safetensors",
703
+ "vision_model.transformer.layers.5.mlp.fc2.bias": "model.safetensors",
704
+ "vision_model.transformer.layers.5.mlp.fc2.weight": "model.safetensors",
705
+ "vision_model.transformer.layers.5.self_attn.out_proj.bias": "model.safetensors",
706
+ "vision_model.transformer.layers.5.self_attn.out_proj.weight": "model.safetensors",
707
+ "vision_model.transformer.layers.5.self_attn.qkv_proj.bias": "model.safetensors",
708
+ "vision_model.transformer.layers.5.self_attn.qkv_proj.weight": "model.safetensors",
709
+ "vision_model.transformer.layers.6.layer_norm1.bias": "model.safetensors",
710
+ "vision_model.transformer.layers.6.layer_norm1.weight": "model.safetensors",
711
+ "vision_model.transformer.layers.6.layer_norm2.bias": "model.safetensors",
712
+ "vision_model.transformer.layers.6.layer_norm2.weight": "model.safetensors",
713
+ "vision_model.transformer.layers.6.mlp.fc1.bias": "model.safetensors",
714
+ "vision_model.transformer.layers.6.mlp.fc1.weight": "model.safetensors",
715
+ "vision_model.transformer.layers.6.mlp.fc2.bias": "model.safetensors",
716
+ "vision_model.transformer.layers.6.mlp.fc2.weight": "model.safetensors",
717
+ "vision_model.transformer.layers.6.self_attn.out_proj.bias": "model.safetensors",
718
+ "vision_model.transformer.layers.6.self_attn.out_proj.weight": "model.safetensors",
719
+ "vision_model.transformer.layers.6.self_attn.qkv_proj.bias": "model.safetensors",
720
+ "vision_model.transformer.layers.6.self_attn.qkv_proj.weight": "model.safetensors",
721
+ "vision_model.transformer.layers.7.layer_norm1.bias": "model.safetensors",
722
+ "vision_model.transformer.layers.7.layer_norm1.weight": "model.safetensors",
723
+ "vision_model.transformer.layers.7.layer_norm2.bias": "model.safetensors",
724
+ "vision_model.transformer.layers.7.layer_norm2.weight": "model.safetensors",
725
+ "vision_model.transformer.layers.7.mlp.fc1.bias": "model.safetensors",
726
+ "vision_model.transformer.layers.7.mlp.fc1.weight": "model.safetensors",
727
+ "vision_model.transformer.layers.7.mlp.fc2.bias": "model.safetensors",
728
+ "vision_model.transformer.layers.7.mlp.fc2.weight": "model.safetensors",
729
+ "vision_model.transformer.layers.7.self_attn.out_proj.bias": "model.safetensors",
730
+ "vision_model.transformer.layers.7.self_attn.out_proj.weight": "model.safetensors",
731
+ "vision_model.transformer.layers.7.self_attn.qkv_proj.bias": "model.safetensors",
732
+ "vision_model.transformer.layers.7.self_attn.qkv_proj.weight": "model.safetensors",
733
+ "vision_model.transformer.layers.8.layer_norm1.bias": "model.safetensors",
734
+ "vision_model.transformer.layers.8.layer_norm1.weight": "model.safetensors",
735
+ "vision_model.transformer.layers.8.layer_norm2.bias": "model.safetensors",
736
+ "vision_model.transformer.layers.8.layer_norm2.weight": "model.safetensors",
737
+ "vision_model.transformer.layers.8.mlp.fc1.bias": "model.safetensors",
738
+ "vision_model.transformer.layers.8.mlp.fc1.weight": "model.safetensors",
739
+ "vision_model.transformer.layers.8.mlp.fc2.bias": "model.safetensors",
740
+ "vision_model.transformer.layers.8.mlp.fc2.weight": "model.safetensors",
741
+ "vision_model.transformer.layers.8.self_attn.out_proj.bias": "model.safetensors",
742
+ "vision_model.transformer.layers.8.self_attn.out_proj.weight": "model.safetensors",
743
+ "vision_model.transformer.layers.8.self_attn.qkv_proj.bias": "model.safetensors",
744
+ "vision_model.transformer.layers.8.self_attn.qkv_proj.weight": "model.safetensors",
745
+ "vision_model.transformer.layers.9.layer_norm1.bias": "model.safetensors",
746
+ "vision_model.transformer.layers.9.layer_norm1.weight": "model.safetensors",
747
+ "vision_model.transformer.layers.9.layer_norm2.bias": "model.safetensors",
748
+ "vision_model.transformer.layers.9.layer_norm2.weight": "model.safetensors",
749
+ "vision_model.transformer.layers.9.mlp.fc1.bias": "model.safetensors",
750
+ "vision_model.transformer.layers.9.mlp.fc1.weight": "model.safetensors",
751
+ "vision_model.transformer.layers.9.mlp.fc2.bias": "model.safetensors",
752
+ "vision_model.transformer.layers.9.mlp.fc2.weight": "model.safetensors",
753
+ "vision_model.transformer.layers.9.self_attn.out_proj.bias": "model.safetensors",
754
+ "vision_model.transformer.layers.9.self_attn.out_proj.weight": "model.safetensors",
755
+ "vision_model.transformer.layers.9.self_attn.qkv_proj.bias": "model.safetensors",
756
+ "vision_model.transformer.layers.9.self_attn.qkv_proj.weight": "model.safetensors"
757
+ }
758
+ }
processor_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_special_token": false,
3
+ "candidate_resolutions": [
4
+ [
5
+ 1024,
6
+ 1024
7
+ ]
8
+ ],
9
+ "downsample_ratio": 4,
10
+ "ignore_id": -100,
11
+ "image_mean": [
12
+ 0.5,
13
+ 0.5,
14
+ 0.5
15
+ ],
16
+ "image_std": [
17
+ 0.5,
18
+ 0.5,
19
+ 0.5
20
+ ],
21
+ "image_token": "<image>",
22
+ "mask_prompt": false,
23
+ "normalize": true,
24
+ "pad_token": "<\uff5c\u2581pad\u2581\uff5c>",
25
+ "patch_size": 16,
26
+ "processor_class": "UnlimitedOCRHFProcessor",
27
+ "sft_format": "unlimitedocr"
28
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ {
4
+ "content": "<|User|>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ {
11
+ "content": "<|Assistant|>",
12
+ "lstrip": false,
13
+ "normalized": false,
14
+ "rstrip": false,
15
+ "single_word": false
16
+ }
17
+ ],
18
+ "bos_token": {
19
+ "content": "<｜begin▁of▁sentence｜>",
20
+ "lstrip": false,
21
+ "normalized": false,
22
+ "rstrip": false,
23
+ "single_word": false
24
+ },
25
+ "eos_token": {
26
+ "content": "<｜end▁of▁sentence｜>",
27
+ "lstrip": false,
28
+ "normalized": false,
29
+ "rstrip": false,
30
+ "single_word": false
31
+ },
32
+ "pad_token": {
33
+ "content": "<｜▁pad▁｜>",
34
+ "lstrip": false,
35
+ "normalized": false,
36
+ "rstrip": false,
37
+ "single_word": false
38
+ }
39
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": null,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<｜begin▁of▁sentence｜>",
5
+ "clean_up_tokenization_spaces": false,
6
+ "eos_token": "<｜end▁of▁sentence｜>",
7
+ "extra_special_tokens": [
8
+ "<|User|>",
9
+ "<|Assistant|>"
10
+ ],
11
+ "is_local": true,
12
+ "local_files_only": false,
13
+ "model_max_length": 1000000000000000019884624838656,
14
+ "pad_token": "<｜▁pad▁｜>",
15
+ "processor_class": "DeepseekOCRProcessor",
16
+ "tokenizer_class": "LlamaTokenizer",
17
+ "unk_token": null,
18
+ "use_default_system_prompt": false
19
+ }