Hugging Face's logo

hadeseus
/
HunyuanOCR-mlx

Image-Text-to-Text
Transformers
Safetensors
MLX
multilingual
hunyuan_vl
text-generation
ocr
hunyuan
vision-language
image-to-text
1B
end-to-end
conversational
Model card Files
xet
 Community
hadeseus commited on
Commit
dd3aded
·
verified ·
1 Parent(s): b17b936

Upload folder using huggingface_hub

Browse files
Files changed (11) hide show
  1. README.md +30 -0
  2. chat_template.jinja +1 -0
  3. config.json +83 -0
  4. generation_config.json +13 -0
  5. model.safetensors +3 -0
  6. model.safetensors.index.json +720 -0
  7. preprocessor_config.json +20 -0
  8. processor_config.json +26 -0
  9. special_tokens_map.json +5 -0
  10. tokenizer.json +0 -0
  11. tokenizer_config.json +23 -0
README.md ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ language:
4
+ - multilingual
5
+ pipeline_tag: image-text-to-text
6
+ library_name: transformers
7
+ base_model:
8
+ - tencent/HunyuanOCR
9
+ tags:
10
+ - ocr
11
+ - hunyuan
12
+ - vision-language
13
+ - image-to-text
14
+ - 1B
15
+ - end-to-end
16
+ - mlx
17
+ ---
18
+
19
+ # hadeseus/HunyuanOCR-mlx
20
+ This model was converted to MLX format from [`tencent/HunyuanOCR`]() using mlx-vlm version **0.3.12**.
21
+ Refer to the [original model card](https://huggingface.co/tencent/HunyuanOCR) for more details on the model.
22
+ ## Use with mlx
23
+
24
+ ```bash
25
+ pip install -U mlx-vlm
26
+ ```
27
+
28
+ ```bash
29
+ python -m mlx_vlm.generate --model hadeseus/HunyuanOCR-mlx --max-tokens 100 --temperature 0.0 --prompt "Describe this image." --image <path_to_image>
30
+ ```
chat_template.jinja ADDED
@@ -0,0 +1 @@
 
 
1
+ {% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% if messages[0]['content'] is string %}{% set system_message = messages[0]['content'] %}{% else %}{% set system_message = messages[0]['content']['text'] %}{% endif %}<｜hy_begin▁of▁sentence｜>{{ system_message }}<｜hy_place▁holder▁no▁3｜>{% else %}{% set loop_messages = messages %}<｜hy_begin▁of▁sentence｜>{% endif %}{% for message in loop_messages %}{% if message['role'] == 'user' %}{% if message['content'] is string %}{{ message['content'] }}{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}<｜hy_place▁holder▁no▁100｜><｜hy_place▁holder▁no▁102｜><｜hy_place▁holder▁no▁101｜>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}{% endif %}<｜hy_User｜>{% elif message['role'] == 'assistant' %}{{ message['content'] }}<｜hy_Assistant｜>{% endif %}{% endfor %}
config.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "HunYuanVLForConditionalGeneration"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attention_head_dim": 128,
8
+ "bos_token_id": 120000,
9
+ "dtype": "bfloat16",
10
+ "eod_token_id": 120020,
11
+ "eos_token_id": [
12
+ 120007,
13
+ 120020
14
+ ],
15
+ "head_dim": 128,
16
+ "hidden_act": "silu",
17
+ "hidden_size": 1024,
18
+ "image_end_token_id": 120119,
19
+ "image_newline_token_id": 120121,
20
+ "image_start_token_id": 120118,
21
+ "image_token_id": 120120,
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 3584,
24
+ "max_position_embeddings": 32768,
25
+ "mlp_bias": false,
26
+ "model_type": "hunyuan_vl",
27
+ "norm_type": "rms",
28
+ "num_attention_heads": 16,
29
+ "num_experts": 1,
30
+ "num_hidden_layers": 24,
31
+ "num_key_value_heads": 8,
32
+ "org_vocab_size": 120818,
33
+ "pad_id": 120002,
34
+ "pad_token_id": -1,
35
+ "pretraining_tp": 1,
36
+ "rms_norm_eps": 1e-05,
37
+ "rope_scaling": {
38
+ "alpha": 1000.0,
39
+ "beta_fast": 32,
40
+ "beta_slow": 1,
41
+ "factor": 1.0,
42
+ "mscale": 1.0,
43
+ "mscale_all_dim": 1.0,
44
+ "type": "xdrope",
45
+ "xdrope_section": [
46
+ 16,
47
+ 16,
48
+ 16,
49
+ 16
50
+ ]
51
+ },
52
+ "rope_theta": 10000.0,
53
+ "routed_scaling_factor": 1.0,
54
+ "sep_token_id": 0,
55
+ "text_end_id": 8,
56
+ "text_start_id": 7,
57
+ "tie_word_embeddings": true,
58
+ "transformers_version": "4.49.0",
59
+ "use_cache": true,
60
+ "use_cla": false,
61
+ "use_qk_norm": true,
62
+ "vision_config": {
63
+ "add_patchemb_bias": true,
64
+ "attention_dropout": 0.0,
65
+ "cat_extra_token": 1,
66
+ "hidden_act": "gelu",
67
+ "hidden_dropout": 0.0,
68
+ "hidden_size": 1152,
69
+ "img_max_token_num": 4096,
70
+ "intermediate_size": 4304,
71
+ "interpolate_mode": "bilinear",
72
+ "max_image_size": 2048,
73
+ "max_vit_seq_len": 16384,
74
+ "num_attention_heads": 16,
75
+ "num_channels": 3,
76
+ "num_hidden_layers": 27,
77
+ "out_hidden_size": 1024,
78
+ "patch_size": 16,
79
+ "rms_norm_eps": 1e-05,
80
+ "spatial_merge_size": 2
81
+ },
82
+ "vocab_size": 120818
83
+ }
generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 120000,
3
+ "pad_token_id": 120002,
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 120007,
7
+ 120020
8
+ ],
9
+ "repetition_penalty": 1.03,
10
+ "top_k": 1,
11
+ "top_p": 1.0,
12
+ "temperature":0.0
13
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2072905c83dcf18e491713deebad615e4d71398efa282f83daec8044839f8041
3
+ size 1992504820
model.safetensors.index.json ADDED
@@ -0,0 +1,720 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 1992416224
4
+ },
5
+ "weight_map": {
6
+ "language_model.model.embed_tokens.weight": "model.safetensors",
7
+ "language_model.model.layers.0.input_layernorm.weight": "model.safetensors",
8
+ "language_model.model.layers.0.mlp.down_proj.weight": "model.safetensors",
9
+ "language_model.model.layers.0.mlp.gate_proj.weight": "model.safetensors",
10
+ "language_model.model.layers.0.mlp.up_proj.weight": "model.safetensors",
11
+ "language_model.model.layers.0.post_attention_layernorm.weight": "model.safetensors",
12
+ "language_model.model.layers.0.self_attn.k_proj.weight": "model.safetensors",
13
+ "language_model.model.layers.0.self_attn.key_layernorm.weight": "model.safetensors",
14
+ "language_model.model.layers.0.self_attn.o_proj.weight": "model.safetensors",
15
+ "language_model.model.layers.0.self_attn.q_proj.weight": "model.safetensors",
16
+ "language_model.model.layers.0.self_attn.query_layernorm.weight": "model.safetensors",
17
+ "language_model.model.layers.0.self_attn.v_proj.weight": "model.safetensors",
18
+ "language_model.model.layers.1.input_layernorm.weight": "model.safetensors",
19
+ "language_model.model.layers.1.mlp.down_proj.weight": "model.safetensors",
20
+ "language_model.model.layers.1.mlp.gate_proj.weight": "model.safetensors",
21
+ "language_model.model.layers.1.mlp.up_proj.weight": "model.safetensors",
22
+ "language_model.model.layers.1.post_attention_layernorm.weight": "model.safetensors",
23
+ "language_model.model.layers.1.self_attn.k_proj.weight": "model.safetensors",
24
+ "language_model.model.layers.1.self_attn.key_layernorm.weight": "model.safetensors",
25
+ "language_model.model.layers.1.self_attn.o_proj.weight": "model.safetensors",
26
+ "language_model.model.layers.1.self_attn.q_proj.weight": "model.safetensors",
27
+ "language_model.model.layers.1.self_attn.query_layernorm.weight": "model.safetensors",
28
+ "language_model.model.layers.1.self_attn.v_proj.weight": "model.safetensors",
29
+ "language_model.model.layers.10.input_layernorm.weight": "model.safetensors",
30
+ "language_model.model.layers.10.mlp.down_proj.weight": "model.safetensors",
31
+ "language_model.model.layers.10.mlp.gate_proj.weight": "model.safetensors",
32
+ "language_model.model.layers.10.mlp.up_proj.weight": "model.safetensors",
33
+ "language_model.model.layers.10.post_attention_layernorm.weight": "model.safetensors",
34
+ "language_model.model.layers.10.self_attn.k_proj.weight": "model.safetensors",
35
+ "language_model.model.layers.10.self_attn.key_layernorm.weight": "model.safetensors",
36
+ "language_model.model.layers.10.self_attn.o_proj.weight": "model.safetensors",
37
+ "language_model.model.layers.10.self_attn.q_proj.weight": "model.safetensors",
38
+ "language_model.model.layers.10.self_attn.query_layernorm.weight": "model.safetensors",
39
+ "language_model.model.layers.10.self_attn.v_proj.weight": "model.safetensors",
40
+ "language_model.model.layers.11.input_layernorm.weight": "model.safetensors",
41
+ "language_model.model.layers.11.mlp.down_proj.weight": "model.safetensors",
42
+ "language_model.model.layers.11.mlp.gate_proj.weight": "model.safetensors",
43
+ "language_model.model.layers.11.mlp.up_proj.weight": "model.safetensors",
44
+ "language_model.model.layers.11.post_attention_layernorm.weight": "model.safetensors",
45
+ "language_model.model.layers.11.self_attn.k_proj.weight": "model.safetensors",
46
+ "language_model.model.layers.11.self_attn.key_layernorm.weight": "model.safetensors",
47
+ "language_model.model.layers.11.self_attn.o_proj.weight": "model.safetensors",
48
+ "language_model.model.layers.11.self_attn.q_proj.weight": "model.safetensors",
49
+ "language_model.model.layers.11.self_attn.query_layernorm.weight": "model.safetensors",
50
+ "language_model.model.layers.11.self_attn.v_proj.weight": "model.safetensors",
51
+ "language_model.model.layers.12.input_layernorm.weight": "model.safetensors",
52
+ "language_model.model.layers.12.mlp.down_proj.weight": "model.safetensors",
53
+ "language_model.model.layers.12.mlp.gate_proj.weight": "model.safetensors",
54
+ "language_model.model.layers.12.mlp.up_proj.weight": "model.safetensors",
55
+ "language_model.model.layers.12.post_attention_layernorm.weight": "model.safetensors",
56
+ "language_model.model.layers.12.self_attn.k_proj.weight": "model.safetensors",
57
+ "language_model.model.layers.12.self_attn.key_layernorm.weight": "model.safetensors",
58
+ "language_model.model.layers.12.self_attn.o_proj.weight": "model.safetensors",
59
+ "language_model.model.layers.12.self_attn.q_proj.weight": "model.safetensors",
60
+ "language_model.model.layers.12.self_attn.query_layernorm.weight": "model.safetensors",
61
+ "language_model.model.layers.12.self_attn.v_proj.weight": "model.safetensors",
62
+ "language_model.model.layers.13.input_layernorm.weight": "model.safetensors",
63
+ "language_model.model.layers.13.mlp.down_proj.weight": "model.safetensors",
64
+ "language_model.model.layers.13.mlp.gate_proj.weight": "model.safetensors",
65
+ "language_model.model.layers.13.mlp.up_proj.weight": "model.safetensors",
66
+ "language_model.model.layers.13.post_attention_layernorm.weight": "model.safetensors",
67
+ "language_model.model.layers.13.self_attn.k_proj.weight": "model.safetensors",
68
+ "language_model.model.layers.13.self_attn.key_layernorm.weight": "model.safetensors",
69
+ "language_model.model.layers.13.self_attn.o_proj.weight": "model.safetensors",
70
+ "language_model.model.layers.13.self_attn.q_proj.weight": "model.safetensors",
71
+ "language_model.model.layers.13.self_attn.query_layernorm.weight": "model.safetensors",
72
+ "language_model.model.layers.13.self_attn.v_proj.weight": "model.safetensors",
73
+ "language_model.model.layers.14.input_layernorm.weight": "model.safetensors",
74
+ "language_model.model.layers.14.mlp.down_proj.weight": "model.safetensors",
75
+ "language_model.model.layers.14.mlp.gate_proj.weight": "model.safetensors",
76
+ "language_model.model.layers.14.mlp.up_proj.weight": "model.safetensors",
77
+ "language_model.model.layers.14.post_attention_layernorm.weight": "model.safetensors",
78
+ "language_model.model.layers.14.self_attn.k_proj.weight": "model.safetensors",
79
+ "language_model.model.layers.14.self_attn.key_layernorm.weight": "model.safetensors",
80
+ "language_model.model.layers.14.self_attn.o_proj.weight": "model.safetensors",
81
+ "language_model.model.layers.14.self_attn.q_proj.weight": "model.safetensors",
82
+ "language_model.model.layers.14.self_attn.query_layernorm.weight": "model.safetensors",
83
+ "language_model.model.layers.14.self_attn.v_proj.weight": "model.safetensors",
84
+ "language_model.model.layers.15.input_layernorm.weight": "model.safetensors",
85
+ "language_model.model.layers.15.mlp.down_proj.weight": "model.safetensors",
86
+ "language_model.model.layers.15.mlp.gate_proj.weight": "model.safetensors",
87
+ "language_model.model.layers.15.mlp.up_proj.weight": "model.safetensors",
88
+ "language_model.model.layers.15.post_attention_layernorm.weight": "model.safetensors",
89
+ "language_model.model.layers.15.self_attn.k_proj.weight": "model.safetensors",
90
+ "language_model.model.layers.15.self_attn.key_layernorm.weight": "model.safetensors",
91
+ "language_model.model.layers.15.self_attn.o_proj.weight": "model.safetensors",
92
+ "language_model.model.layers.15.self_attn.q_proj.weight": "model.safetensors",
93
+ "language_model.model.layers.15.self_attn.query_layernorm.weight": "model.safetensors",
94
+ "language_model.model.layers.15.self_attn.v_proj.weight": "model.safetensors",
95
+ "language_model.model.layers.16.input_layernorm.weight": "model.safetensors",
96
+ "language_model.model.layers.16.mlp.down_proj.weight": "model.safetensors",
97
+ "language_model.model.layers.16.mlp.gate_proj.weight": "model.safetensors",
98
+ "language_model.model.layers.16.mlp.up_proj.weight": "model.safetensors",
99
+ "language_model.model.layers.16.post_attention_layernorm.weight": "model.safetensors",
100
+ "language_model.model.layers.16.self_attn.k_proj.weight": "model.safetensors",
101
+ "language_model.model.layers.16.self_attn.key_layernorm.weight": "model.safetensors",
102
+ "language_model.model.layers.16.self_attn.o_proj.weight": "model.safetensors",
103
+ "language_model.model.layers.16.self_attn.q_proj.weight": "model.safetensors",
104
+ "language_model.model.layers.16.self_attn.query_layernorm.weight": "model.safetensors",
105
+ "language_model.model.layers.16.self_attn.v_proj.weight": "model.safetensors",
106
+ "language_model.model.layers.17.input_layernorm.weight": "model.safetensors",
107
+ "language_model.model.layers.17.mlp.down_proj.weight": "model.safetensors",
108
+ "language_model.model.layers.17.mlp.gate_proj.weight": "model.safetensors",
109
+ "language_model.model.layers.17.mlp.up_proj.weight": "model.safetensors",
110
+ "language_model.model.layers.17.post_attention_layernorm.weight": "model.safetensors",
111
+ "language_model.model.layers.17.self_attn.k_proj.weight": "model.safetensors",
112
+ "language_model.model.layers.17.self_attn.key_layernorm.weight": "model.safetensors",
113
+ "language_model.model.layers.17.self_attn.o_proj.weight": "model.safetensors",
114
+ "language_model.model.layers.17.self_attn.q_proj.weight": "model.safetensors",
115
+ "language_model.model.layers.17.self_attn.query_layernorm.weight": "model.safetensors",
116
+ "language_model.model.layers.17.self_attn.v_proj.weight": "model.safetensors",
117
+ "language_model.model.layers.18.input_layernorm.weight": "model.safetensors",
118
+ "language_model.model.layers.18.mlp.down_proj.weight": "model.safetensors",
119
+ "language_model.model.layers.18.mlp.gate_proj.weight": "model.safetensors",
120
+ "language_model.model.layers.18.mlp.up_proj.weight": "model.safetensors",
121
+ "language_model.model.layers.18.post_attention_layernorm.weight": "model.safetensors",
122
+ "language_model.model.layers.18.self_attn.k_proj.weight": "model.safetensors",
123
+ "language_model.model.layers.18.self_attn.key_layernorm.weight": "model.safetensors",
124
+ "language_model.model.layers.18.self_attn.o_proj.weight": "model.safetensors",
125
+ "language_model.model.layers.18.self_attn.q_proj.weight": "model.safetensors",
126
+ "language_model.model.layers.18.self_attn.query_layernorm.weight": "model.safetensors",
127
+ "language_model.model.layers.18.self_attn.v_proj.weight": "model.safetensors",
128
+ "language_model.model.layers.19.input_layernorm.weight": "model.safetensors",
129
+ "language_model.model.layers.19.mlp.down_proj.weight": "model.safetensors",
130
+ "language_model.model.layers.19.mlp.gate_proj.weight": "model.safetensors",
131
+ "language_model.model.layers.19.mlp.up_proj.weight": "model.safetensors",
132
+ "language_model.model.layers.19.post_attention_layernorm.weight": "model.safetensors",
133
+ "language_model.model.layers.19.self_attn.k_proj.weight": "model.safetensors",
134
+ "language_model.model.layers.19.self_attn.key_layernorm.weight": "model.safetensors",
135
+ "language_model.model.layers.19.self_attn.o_proj.weight": "model.safetensors",
136
+ "language_model.model.layers.19.self_attn.q_proj.weight": "model.safetensors",
137
+ "language_model.model.layers.19.self_attn.query_layernorm.weight": "model.safetensors",
138
+ "language_model.model.layers.19.self_attn.v_proj.weight": "model.safetensors",
139
+ "language_model.model.layers.2.input_layernorm.weight": "model.safetensors",
140
+ "language_model.model.layers.2.mlp.down_proj.weight": "model.safetensors",
141
+ "language_model.model.layers.2.mlp.gate_proj.weight": "model.safetensors",
142
+ "language_model.model.layers.2.mlp.up_proj.weight": "model.safetensors",
143
+ "language_model.model.layers.2.post_attention_layernorm.weight": "model.safetensors",
144
+ "language_model.model.layers.2.self_attn.k_proj.weight": "model.safetensors",
145
+ "language_model.model.layers.2.self_attn.key_layernorm.weight": "model.safetensors",
146
+ "language_model.model.layers.2.self_attn.o_proj.weight": "model.safetensors",
147
+ "language_model.model.layers.2.self_attn.q_proj.weight": "model.safetensors",
148
+ "language_model.model.layers.2.self_attn.query_layernorm.weight": "model.safetensors",
149
+ "language_model.model.layers.2.self_attn.v_proj.weight": "model.safetensors",
150
+ "language_model.model.layers.20.input_layernorm.weight": "model.safetensors",
151
+ "language_model.model.layers.20.mlp.down_proj.weight": "model.safetensors",
152
+ "language_model.model.layers.20.mlp.gate_proj.weight": "model.safetensors",
153
+ "language_model.model.layers.20.mlp.up_proj.weight": "model.safetensors",
154
+ "language_model.model.layers.20.post_attention_layernorm.weight": "model.safetensors",
155
+ "language_model.model.layers.20.self_attn.k_proj.weight": "model.safetensors",
156
+ "language_model.model.layers.20.self_attn.key_layernorm.weight": "model.safetensors",
157
+ "language_model.model.layers.20.self_attn.o_proj.weight": "model.safetensors",
158
+ "language_model.model.layers.20.self_attn.q_proj.weight": "model.safetensors",
159
+ "language_model.model.layers.20.self_attn.query_layernorm.weight": "model.safetensors",
160
+ "language_model.model.layers.20.self_attn.v_proj.weight": "model.safetensors",
161
+ "language_model.model.layers.21.input_layernorm.weight": "model.safetensors",
162
+ "language_model.model.layers.21.mlp.down_proj.weight": "model.safetensors",
163
+ "language_model.model.layers.21.mlp.gate_proj.weight": "model.safetensors",
164
+ "language_model.model.layers.21.mlp.up_proj.weight": "model.safetensors",
165
+ "language_model.model.layers.21.post_attention_layernorm.weight": "model.safetensors",
166
+ "language_model.model.layers.21.self_attn.k_proj.weight": "model.safetensors",
167
+ "language_model.model.layers.21.self_attn.key_layernorm.weight": "model.safetensors",
168
+ "language_model.model.layers.21.self_attn.o_proj.weight": "model.safetensors",
169
+ "language_model.model.layers.21.self_attn.q_proj.weight": "model.safetensors",
170
+ "language_model.model.layers.21.self_attn.query_layernorm.weight": "model.safetensors",
171
+ "language_model.model.layers.21.self_attn.v_proj.weight": "model.safetensors",
172
+ "language_model.model.layers.22.input_layernorm.weight": "model.safetensors",
173
+ "language_model.model.layers.22.mlp.down_proj.weight": "model.safetensors",
174
+ "language_model.model.layers.22.mlp.gate_proj.weight": "model.safetensors",
175
+ "language_model.model.layers.22.mlp.up_proj.weight": "model.safetensors",
176
+ "language_model.model.layers.22.post_attention_layernorm.weight": "model.safetensors",
177
+ "language_model.model.layers.22.self_attn.k_proj.weight": "model.safetensors",
178
+ "language_model.model.layers.22.self_attn.key_layernorm.weight": "model.safetensors",
179
+ "language_model.model.layers.22.self_attn.o_proj.weight": "model.safetensors",
180
+ "language_model.model.layers.22.self_attn.q_proj.weight": "model.safetensors",
181
+ "language_model.model.layers.22.self_attn.query_layernorm.weight": "model.safetensors",
182
+ "language_model.model.layers.22.self_attn.v_proj.weight": "model.safetensors",
183
+ "language_model.model.layers.23.input_layernorm.weight": "model.safetensors",
184
+ "language_model.model.layers.23.mlp.down_proj.weight": "model.safetensors",
185
+ "language_model.model.layers.23.mlp.gate_proj.weight": "model.safetensors",
186
+ "language_model.model.layers.23.mlp.up_proj.weight": "model.safetensors",
187
+ "language_model.model.layers.23.post_attention_layernorm.weight": "model.safetensors",
188
+ "language_model.model.layers.23.self_attn.k_proj.weight": "model.safetensors",
189
+ "language_model.model.layers.23.self_attn.key_layernorm.weight": "model.safetensors",
190
+ "language_model.model.layers.23.self_attn.o_proj.weight": "model.safetensors",
191
+ "language_model.model.layers.23.self_attn.q_proj.weight": "model.safetensors",
192
+ "language_model.model.layers.23.self_attn.query_layernorm.weight": "model.safetensors",
193
+ "language_model.model.layers.23.self_attn.v_proj.weight": "model.safetensors",
194
+ "language_model.model.layers.3.input_layernorm.weight": "model.safetensors",
195
+ "language_model.model.layers.3.mlp.down_proj.weight": "model.safetensors",
196
+ "language_model.model.layers.3.mlp.gate_proj.weight": "model.safetensors",
197
+ "language_model.model.layers.3.mlp.up_proj.weight": "model.safetensors",
198
+ "language_model.model.layers.3.post_attention_layernorm.weight": "model.safetensors",
199
+ "language_model.model.layers.3.self_attn.k_proj.weight": "model.safetensors",
200
+ "language_model.model.layers.3.self_attn.key_layernorm.weight": "model.safetensors",
201
+ "language_model.model.layers.3.self_attn.o_proj.weight": "model.safetensors",
202
+ "language_model.model.layers.3.self_attn.q_proj.weight": "model.safetensors",
203
+ "language_model.model.layers.3.self_attn.query_layernorm.weight": "model.safetensors",
204
+ "language_model.model.layers.3.self_attn.v_proj.weight": "model.safetensors",
205
+ "language_model.model.layers.4.input_layernorm.weight": "model.safetensors",
206
+ "language_model.model.layers.4.mlp.down_proj.weight": "model.safetensors",
207
+ "language_model.model.layers.4.mlp.gate_proj.weight": "model.safetensors",
208
+ "language_model.model.layers.4.mlp.up_proj.weight": "model.safetensors",
209
+ "language_model.model.layers.4.post_attention_layernorm.weight": "model.safetensors",
210
+ "language_model.model.layers.4.self_attn.k_proj.weight": "model.safetensors",
211
+ "language_model.model.layers.4.self_attn.key_layernorm.weight": "model.safetensors",
212
+ "language_model.model.layers.4.self_attn.o_proj.weight": "model.safetensors",
213
+ "language_model.model.layers.4.self_attn.q_proj.weight": "model.safetensors",
214
+ "language_model.model.layers.4.self_attn.query_layernorm.weight": "model.safetensors",
215
+ "language_model.model.layers.4.self_attn.v_proj.weight": "model.safetensors",
216
+ "language_model.model.layers.5.input_layernorm.weight": "model.safetensors",
217
+ "language_model.model.layers.5.mlp.down_proj.weight": "model.safetensors",
218
+ "language_model.model.layers.5.mlp.gate_proj.weight": "model.safetensors",
219
+ "language_model.model.layers.5.mlp.up_proj.weight": "model.safetensors",
220
+ "language_model.model.layers.5.post_attention_layernorm.weight": "model.safetensors",
221
+ "language_model.model.layers.5.self_attn.k_proj.weight": "model.safetensors",
222
+ "language_model.model.layers.5.self_attn.key_layernorm.weight": "model.safetensors",
223
+ "language_model.model.layers.5.self_attn.o_proj.weight": "model.safetensors",
224
+ "language_model.model.layers.5.self_attn.q_proj.weight": "model.safetensors",
225
+ "language_model.model.layers.5.self_attn.query_layernorm.weight": "model.safetensors",
226
+ "language_model.model.layers.5.self_attn.v_proj.weight": "model.safetensors",
227
+ "language_model.model.layers.6.input_layernorm.weight": "model.safetensors",
228
+ "language_model.model.layers.6.mlp.down_proj.weight": "model.safetensors",
229
+ "language_model.model.layers.6.mlp.gate_proj.weight": "model.safetensors",
230
+ "language_model.model.layers.6.mlp.up_proj.weight": "model.safetensors",
231
+ "language_model.model.layers.6.post_attention_layernorm.weight": "model.safetensors",
232
+ "language_model.model.layers.6.self_attn.k_proj.weight": "model.safetensors",
233
+ "language_model.model.layers.6.self_attn.key_layernorm.weight": "model.safetensors",
234
+ "language_model.model.layers.6.self_attn.o_proj.weight": "model.safetensors",
235
+ "language_model.model.layers.6.self_attn.q_proj.weight": "model.safetensors",
236
+ "language_model.model.layers.6.self_attn.query_layernorm.weight": "model.safetensors",
237
+ "language_model.model.layers.6.self_attn.v_proj.weight": "model.safetensors",
238
+ "language_model.model.layers.7.input_layernorm.weight": "model.safetensors",
239
+ "language_model.model.layers.7.mlp.down_proj.weight": "model.safetensors",
240
+ "language_model.model.layers.7.mlp.gate_proj.weight": "model.safetensors",
241
+ "language_model.model.layers.7.mlp.up_proj.weight": "model.safetensors",
242
+ "language_model.model.layers.7.post_attention_layernorm.weight": "model.safetensors",
243
+ "language_model.model.layers.7.self_attn.k_proj.weight": "model.safetensors",
244
+ "language_model.model.layers.7.self_attn.key_layernorm.weight": "model.safetensors",
245
+ "language_model.model.layers.7.self_attn.o_proj.weight": "model.safetensors",
246
+ "language_model.model.layers.7.self_attn.q_proj.weight": "model.safetensors",
247
+ "language_model.model.layers.7.self_attn.query_layernorm.weight": "model.safetensors",
248
+ "language_model.model.layers.7.self_attn.v_proj.weight": "model.safetensors",
249
+ "language_model.model.layers.8.input_layernorm.weight": "model.safetensors",
250
+ "language_model.model.layers.8.mlp.down_proj.weight": "model.safetensors",
251
+ "language_model.model.layers.8.mlp.gate_proj.weight": "model.safetensors",
252
+ "language_model.model.layers.8.mlp.up_proj.weight": "model.safetensors",
253
+ "language_model.model.layers.8.post_attention_layernorm.weight": "model.safetensors",
254
+ "language_model.model.layers.8.self_attn.k_proj.weight": "model.safetensors",
255
+ "language_model.model.layers.8.self_attn.key_layernorm.weight": "model.safetensors",
256
+ "language_model.model.layers.8.self_attn.o_proj.weight": "model.safetensors",
257
+ "language_model.model.layers.8.self_attn.q_proj.weight": "model.safetensors",
258
+ "language_model.model.layers.8.self_attn.query_layernorm.weight": "model.safetensors",
259
+ "language_model.model.layers.8.self_attn.v_proj.weight": "model.safetensors",
260
+ "language_model.model.layers.9.input_layernorm.weight": "model.safetensors",
261
+ "language_model.model.layers.9.mlp.down_proj.weight": "model.safetensors",
262
+ "language_model.model.layers.9.mlp.gate_proj.weight": "model.safetensors",
263
+ "language_model.model.layers.9.mlp.up_proj.weight": "model.safetensors",
264
+ "language_model.model.layers.9.post_attention_layernorm.weight": "model.safetensors",
265
+ "language_model.model.layers.9.self_attn.k_proj.weight": "model.safetensors",
266
+ "language_model.model.layers.9.self_attn.key_layernorm.weight": "model.safetensors",
267
+ "language_model.model.layers.9.self_attn.o_proj.weight": "model.safetensors",
268
+ "language_model.model.layers.9.self_attn.q_proj.weight": "model.safetensors",
269
+ "language_model.model.layers.9.self_attn.query_layernorm.weight": "model.safetensors",
270
+ "language_model.model.layers.9.self_attn.v_proj.weight": "model.safetensors",
271
+ "language_model.model.norm.weight": "model.safetensors",
272
+ "vision_tower.embeddings.patch_embedding.bias": "model.safetensors",
273
+ "vision_tower.embeddings.patch_embedding.weight": "model.safetensors",
274
+ "vision_tower.embeddings.position_embedding.weight": "model.safetensors",
275
+ "vision_tower.layers.0.input_layernorm.bias": "model.safetensors",
276
+ "vision_tower.layers.0.input_layernorm.weight": "model.safetensors",
277
+ "vision_tower.layers.0.mlp.dense_4h_to_h.bias": "model.safetensors",
278
+ "vision_tower.layers.0.mlp.dense_4h_to_h.weight": "model.safetensors",
279
+ "vision_tower.layers.0.mlp.dense_h_to_4h.bias": "model.safetensors",
280
+ "vision_tower.layers.0.mlp.dense_h_to_4h.weight": "model.safetensors",
281
+ "vision_tower.layers.0.post_attention_layernorm.bias": "model.safetensors",
282
+ "vision_tower.layers.0.post_attention_layernorm.weight": "model.safetensors",
283
+ "vision_tower.layers.0.self_attn.k_proj.bias": "model.safetensors",
284
+ "vision_tower.layers.0.self_attn.k_proj.weight": "model.safetensors",
285
+ "vision_tower.layers.0.self_attn.o_proj.bias": "model.safetensors",
286
+ "vision_tower.layers.0.self_attn.o_proj.weight": "model.safetensors",
287
+ "vision_tower.layers.0.self_attn.q_proj.bias": "model.safetensors",
288
+ "vision_tower.layers.0.self_attn.q_proj.weight": "model.safetensors",
289
+ "vision_tower.layers.0.self_attn.v_proj.bias": "model.safetensors",
290
+ "vision_tower.layers.0.self_attn.v_proj.weight": "model.safetensors",
291
+ "vision_tower.layers.1.input_layernorm.bias": "model.safetensors",
292
+ "vision_tower.layers.1.input_layernorm.weight": "model.safetensors",
293
+ "vision_tower.layers.1.mlp.dense_4h_to_h.bias": "model.safetensors",
294
+ "vision_tower.layers.1.mlp.dense_4h_to_h.weight": "model.safetensors",
295
+ "vision_tower.layers.1.mlp.dense_h_to_4h.bias": "model.safetensors",
296
+ "vision_tower.layers.1.mlp.dense_h_to_4h.weight": "model.safetensors",
297
+ "vision_tower.layers.1.post_attention_layernorm.bias": "model.safetensors",
298
+ "vision_tower.layers.1.post_attention_layernorm.weight": "model.safetensors",
299
+ "vision_tower.layers.1.self_attn.k_proj.bias": "model.safetensors",
300
+ "vision_tower.layers.1.self_attn.k_proj.weight": "model.safetensors",
301
+ "vision_tower.layers.1.self_attn.o_proj.bias": "model.safetensors",
302
+ "vision_tower.layers.1.self_attn.o_proj.weight": "model.safetensors",
303
+ "vision_tower.layers.1.self_attn.q_proj.bias": "model.safetensors",
304
+ "vision_tower.layers.1.self_attn.q_proj.weight": "model.safetensors",
305
+ "vision_tower.layers.1.self_attn.v_proj.bias": "model.safetensors",
306
+ "vision_tower.layers.1.self_attn.v_proj.weight": "model.safetensors",
307
+ "vision_tower.layers.10.input_layernorm.bias": "model.safetensors",
308
+ "vision_tower.layers.10.input_layernorm.weight": "model.safetensors",
309
+ "vision_tower.layers.10.mlp.dense_4h_to_h.bias": "model.safetensors",
310
+ "vision_tower.layers.10.mlp.dense_4h_to_h.weight": "model.safetensors",
311
+ "vision_tower.layers.10.mlp.dense_h_to_4h.bias": "model.safetensors",
312
+ "vision_tower.layers.10.mlp.dense_h_to_4h.weight": "model.safetensors",
313
+ "vision_tower.layers.10.post_attention_layernorm.bias": "model.safetensors",
314
+ "vision_tower.layers.10.post_attention_layernorm.weight": "model.safetensors",
315
+ "vision_tower.layers.10.self_attn.k_proj.bias": "model.safetensors",
316
+ "vision_tower.layers.10.self_attn.k_proj.weight": "model.safetensors",
317
+ "vision_tower.layers.10.self_attn.o_proj.bias": "model.safetensors",
318
+ "vision_tower.layers.10.self_attn.o_proj.weight": "model.safetensors",
319
+ "vision_tower.layers.10.self_attn.q_proj.bias": "model.safetensors",
320
+ "vision_tower.layers.10.self_attn.q_proj.weight": "model.safetensors",
321
+ "vision_tower.layers.10.self_attn.v_proj.bias": "model.safetensors",
322
+ "vision_tower.layers.10.self_attn.v_proj.weight": "model.safetensors",
323
+ "vision_tower.layers.11.input_layernorm.bias": "model.safetensors",
324
+ "vision_tower.layers.11.input_layernorm.weight": "model.safetensors",
325
+ "vision_tower.layers.11.mlp.dense_4h_to_h.bias": "model.safetensors",
326
+ "vision_tower.layers.11.mlp.dense_4h_to_h.weight": "model.safetensors",
327
+ "vision_tower.layers.11.mlp.dense_h_to_4h.bias": "model.safetensors",
328
+ "vision_tower.layers.11.mlp.dense_h_to_4h.weight": "model.safetensors",
329
+ "vision_tower.layers.11.post_attention_layernorm.bias": "model.safetensors",
330
+ "vision_tower.layers.11.post_attention_layernorm.weight": "model.safetensors",
331
+ "vision_tower.layers.11.self_attn.k_proj.bias": "model.safetensors",
332
+ "vision_tower.layers.11.self_attn.k_proj.weight": "model.safetensors",
333
+ "vision_tower.layers.11.self_attn.o_proj.bias": "model.safetensors",
334
+ "vision_tower.layers.11.self_attn.o_proj.weight": "model.safetensors",
335
+ "vision_tower.layers.11.self_attn.q_proj.bias": "model.safetensors",
336
+ "vision_tower.layers.11.self_attn.q_proj.weight": "model.safetensors",
337
+ "vision_tower.layers.11.self_attn.v_proj.bias": "model.safetensors",
338
+ "vision_tower.layers.11.self_attn.v_proj.weight": "model.safetensors",
339
+ "vision_tower.layers.12.input_layernorm.bias": "model.safetensors",
340
+ "vision_tower.layers.12.input_layernorm.weight": "model.safetensors",
341
+ "vision_tower.layers.12.mlp.dense_4h_to_h.bias": "model.safetensors",
342
+ "vision_tower.layers.12.mlp.dense_4h_to_h.weight": "model.safetensors",
343
+ "vision_tower.layers.12.mlp.dense_h_to_4h.bias": "model.safetensors",
344
+ "vision_tower.layers.12.mlp.dense_h_to_4h.weight": "model.safetensors",
345
+ "vision_tower.layers.12.post_attention_layernorm.bias": "model.safetensors",
346
+ "vision_tower.layers.12.post_attention_layernorm.weight": "model.safetensors",
347
+ "vision_tower.layers.12.self_attn.k_proj.bias": "model.safetensors",
348
+ "vision_tower.layers.12.self_attn.k_proj.weight": "model.safetensors",
349
+ "vision_tower.layers.12.self_attn.o_proj.bias": "model.safetensors",
350
+ "vision_tower.layers.12.self_attn.o_proj.weight": "model.safetensors",
351
+ "vision_tower.layers.12.self_attn.q_proj.bias": "model.safetensors",
352
+ "vision_tower.layers.12.self_attn.q_proj.weight": "model.safetensors",
353
+ "vision_tower.layers.12.self_attn.v_proj.bias": "model.safetensors",
354
+ "vision_tower.layers.12.self_attn.v_proj.weight": "model.safetensors",
355
+ "vision_tower.layers.13.input_layernorm.bias": "model.safetensors",
356
+ "vision_tower.layers.13.input_layernorm.weight": "model.safetensors",
357
+ "vision_tower.layers.13.mlp.dense_4h_to_h.bias": "model.safetensors",
358
+ "vision_tower.layers.13.mlp.dense_4h_to_h.weight": "model.safetensors",
359
+ "vision_tower.layers.13.mlp.dense_h_to_4h.bias": "model.safetensors",
360
+ "vision_tower.layers.13.mlp.dense_h_to_4h.weight": "model.safetensors",
361
+ "vision_tower.layers.13.post_attention_layernorm.bias": "model.safetensors",
362
+ "vision_tower.layers.13.post_attention_layernorm.weight": "model.safetensors",
363
+ "vision_tower.layers.13.self_attn.k_proj.bias": "model.safetensors",
364
+ "vision_tower.layers.13.self_attn.k_proj.weight": "model.safetensors",
365
+ "vision_tower.layers.13.self_attn.o_proj.bias": "model.safetensors",
366
+ "vision_tower.layers.13.self_attn.o_proj.weight": "model.safetensors",
367
+ "vision_tower.layers.13.self_attn.q_proj.bias": "model.safetensors",
368
+ "vision_tower.layers.13.self_attn.q_proj.weight": "model.safetensors",
369
+ "vision_tower.layers.13.self_attn.v_proj.bias": "model.safetensors",
370
+ "vision_tower.layers.13.self_attn.v_proj.weight": "model.safetensors",
371
+ "vision_tower.layers.14.input_layernorm.bias": "model.safetensors",
372
+ "vision_tower.layers.14.input_layernorm.weight": "model.safetensors",
373
+ "vision_tower.layers.14.mlp.dense_4h_to_h.bias": "model.safetensors",
374
+ "vision_tower.layers.14.mlp.dense_4h_to_h.weight": "model.safetensors",
375
+ "vision_tower.layers.14.mlp.dense_h_to_4h.bias": "model.safetensors",
376
+ "vision_tower.layers.14.mlp.dense_h_to_4h.weight": "model.safetensors",
377
+ "vision_tower.layers.14.post_attention_layernorm.bias": "model.safetensors",
378
+ "vision_tower.layers.14.post_attention_layernorm.weight": "model.safetensors",
379
+ "vision_tower.layers.14.self_attn.k_proj.bias": "model.safetensors",
380
+ "vision_tower.layers.14.self_attn.k_proj.weight": "model.safetensors",
381
+ "vision_tower.layers.14.self_attn.o_proj.bias": "model.safetensors",
382
+ "vision_tower.layers.14.self_attn.o_proj.weight": "model.safetensors",
383
+ "vision_tower.layers.14.self_attn.q_proj.bias": "model.safetensors",
384
+ "vision_tower.layers.14.self_attn.q_proj.weight": "model.safetensors",
385
+ "vision_tower.layers.14.self_attn.v_proj.bias": "model.safetensors",
386
+ "vision_tower.layers.14.self_attn.v_proj.weight": "model.safetensors",
387
+ "vision_tower.layers.15.input_layernorm.bias": "model.safetensors",
388
+ "vision_tower.layers.15.input_layernorm.weight": "model.safetensors",
389
+ "vision_tower.layers.15.mlp.dense_4h_to_h.bias": "model.safetensors",
390
+ "vision_tower.layers.15.mlp.dense_4h_to_h.weight": "model.safetensors",
391
+ "vision_tower.layers.15.mlp.dense_h_to_4h.bias": "model.safetensors",
392
+ "vision_tower.layers.15.mlp.dense_h_to_4h.weight": "model.safetensors",
393
+ "vision_tower.layers.15.post_attention_layernorm.bias": "model.safetensors",
394
+ "vision_tower.layers.15.post_attention_layernorm.weight": "model.safetensors",
395
+ "vision_tower.layers.15.self_attn.k_proj.bias": "model.safetensors",
396
+ "vision_tower.layers.15.self_attn.k_proj.weight": "model.safetensors",
397
+ "vision_tower.layers.15.self_attn.o_proj.bias": "model.safetensors",
398
+ "vision_tower.layers.15.self_attn.o_proj.weight": "model.safetensors",
399
+ "vision_tower.layers.15.self_attn.q_proj.bias": "model.safetensors",
400
+ "vision_tower.layers.15.self_attn.q_proj.weight": "model.safetensors",
401
+ "vision_tower.layers.15.self_attn.v_proj.bias": "model.safetensors",
402
+ "vision_tower.layers.15.self_attn.v_proj.weight": "model.safetensors",
403
+ "vision_tower.layers.16.input_layernorm.bias": "model.safetensors",
404
+ "vision_tower.layers.16.input_layernorm.weight": "model.safetensors",
405
+ "vision_tower.layers.16.mlp.dense_4h_to_h.bias": "model.safetensors",
406
+ "vision_tower.layers.16.mlp.dense_4h_to_h.weight": "model.safetensors",
407
+ "vision_tower.layers.16.mlp.dense_h_to_4h.bias": "model.safetensors",
408
+ "vision_tower.layers.16.mlp.dense_h_to_4h.weight": "model.safetensors",
409
+ "vision_tower.layers.16.post_attention_layernorm.bias": "model.safetensors",
410
+ "vision_tower.layers.16.post_attention_layernorm.weight": "model.safetensors",
411
+ "vision_tower.layers.16.self_attn.k_proj.bias": "model.safetensors",
412
+ "vision_tower.layers.16.self_attn.k_proj.weight": "model.safetensors",
413
+ "vision_tower.layers.16.self_attn.o_proj.bias": "model.safetensors",
414
+ "vision_tower.layers.16.self_attn.o_proj.weight": "model.safetensors",
415
+ "vision_tower.layers.16.self_attn.q_proj.bias": "model.safetensors",
416
+ "vision_tower.layers.16.self_attn.q_proj.weight": "model.safetensors",
417
+ "vision_tower.layers.16.self_attn.v_proj.bias": "model.safetensors",
418
+ "vision_tower.layers.16.self_attn.v_proj.weight": "model.safetensors",
419
+ "vision_tower.layers.17.input_layernorm.bias": "model.safetensors",
420
+ "vision_tower.layers.17.input_layernorm.weight": "model.safetensors",
421
+ "vision_tower.layers.17.mlp.dense_4h_to_h.bias": "model.safetensors",
422
+ "vision_tower.layers.17.mlp.dense_4h_to_h.weight": "model.safetensors",
423
+ "vision_tower.layers.17.mlp.dense_h_to_4h.bias": "model.safetensors",
424
+ "vision_tower.layers.17.mlp.dense_h_to_4h.weight": "model.safetensors",
425
+ "vision_tower.layers.17.post_attention_layernorm.bias": "model.safetensors",
426
+ "vision_tower.layers.17.post_attention_layernorm.weight": "model.safetensors",
427
+ "vision_tower.layers.17.self_attn.k_proj.bias": "model.safetensors",
428
+ "vision_tower.layers.17.self_attn.k_proj.weight": "model.safetensors",
429
+ "vision_tower.layers.17.self_attn.o_proj.bias": "model.safetensors",
430
+ "vision_tower.layers.17.self_attn.o_proj.weight": "model.safetensors",
431
+ "vision_tower.layers.17.self_attn.q_proj.bias": "model.safetensors",
432
+ "vision_tower.layers.17.self_attn.q_proj.weight": "model.safetensors",
433
+ "vision_tower.layers.17.self_attn.v_proj.bias": "model.safetensors",
434
+ "vision_tower.layers.17.self_attn.v_proj.weight": "model.safetensors",
435
+ "vision_tower.layers.18.input_layernorm.bias": "model.safetensors",
436
+ "vision_tower.layers.18.input_layernorm.weight": "model.safetensors",
437
+ "vision_tower.layers.18.mlp.dense_4h_to_h.bias": "model.safetensors",
438
+ "vision_tower.layers.18.mlp.dense_4h_to_h.weight": "model.safetensors",
439
+ "vision_tower.layers.18.mlp.dense_h_to_4h.bias": "model.safetensors",
440
+ "vision_tower.layers.18.mlp.dense_h_to_4h.weight": "model.safetensors",
441
+ "vision_tower.layers.18.post_attention_layernorm.bias": "model.safetensors",
442
+ "vision_tower.layers.18.post_attention_layernorm.weight": "model.safetensors",
443
+ "vision_tower.layers.18.self_attn.k_proj.bias": "model.safetensors",
444
+ "vision_tower.layers.18.self_attn.k_proj.weight": "model.safetensors",
445
+ "vision_tower.layers.18.self_attn.o_proj.bias": "model.safetensors",
446
+ "vision_tower.layers.18.self_attn.o_proj.weight": "model.safetensors",
447
+ "vision_tower.layers.18.self_attn.q_proj.bias": "model.safetensors",
448
+ "vision_tower.layers.18.self_attn.q_proj.weight": "model.safetensors",
449
+ "vision_tower.layers.18.self_attn.v_proj.bias": "model.safetensors",
450
+ "vision_tower.layers.18.self_attn.v_proj.weight": "model.safetensors",
451
+ "vision_tower.layers.19.input_layernorm.bias": "model.safetensors",
452
+ "vision_tower.layers.19.input_layernorm.weight": "model.safetensors",
453
+ "vision_tower.layers.19.mlp.dense_4h_to_h.bias": "model.safetensors",
454
+ "vision_tower.layers.19.mlp.dense_4h_to_h.weight": "model.safetensors",
455
+ "vision_tower.layers.19.mlp.dense_h_to_4h.bias": "model.safetensors",
456
+ "vision_tower.layers.19.mlp.dense_h_to_4h.weight": "model.safetensors",
457
+ "vision_tower.layers.19.post_attention_layernorm.bias": "model.safetensors",
458
+ "vision_tower.layers.19.post_attention_layernorm.weight": "model.safetensors",
459
+ "vision_tower.layers.19.self_attn.k_proj.bias": "model.safetensors",
460
+ "vision_tower.layers.19.self_attn.k_proj.weight": "model.safetensors",
461
+ "vision_tower.layers.19.self_attn.o_proj.bias": "model.safetensors",
462
+ "vision_tower.layers.19.self_attn.o_proj.weight": "model.safetensors",
463
+ "vision_tower.layers.19.self_attn.q_proj.bias": "model.safetensors",
464
+ "vision_tower.layers.19.self_attn.q_proj.weight": "model.safetensors",
465
+ "vision_tower.layers.19.self_attn.v_proj.bias": "model.safetensors",
466
+ "vision_tower.layers.19.self_attn.v_proj.weight": "model.safetensors",
467
+ "vision_tower.layers.2.input_layernorm.bias": "model.safetensors",
468
+ "vision_tower.layers.2.input_layernorm.weight": "model.safetensors",
469
+ "vision_tower.layers.2.mlp.dense_4h_to_h.bias": "model.safetensors",
470
+ "vision_tower.layers.2.mlp.dense_4h_to_h.weight": "model.safetensors",
471
+ "vision_tower.layers.2.mlp.dense_h_to_4h.bias": "model.safetensors",
472
+ "vision_tower.layers.2.mlp.dense_h_to_4h.weight": "model.safetensors",
473
+ "vision_tower.layers.2.post_attention_layernorm.bias": "model.safetensors",
474
+ "vision_tower.layers.2.post_attention_layernorm.weight": "model.safetensors",
475
+ "vision_tower.layers.2.self_attn.k_proj.bias": "model.safetensors",
476
+ "vision_tower.layers.2.self_attn.k_proj.weight": "model.safetensors",
477
+ "vision_tower.layers.2.self_attn.o_proj.bias": "model.safetensors",
478
+ "vision_tower.layers.2.self_attn.o_proj.weight": "model.safetensors",
479
+ "vision_tower.layers.2.self_attn.q_proj.bias": "model.safetensors",
480
+ "vision_tower.layers.2.self_attn.q_proj.weight": "model.safetensors",
481
+ "vision_tower.layers.2.self_attn.v_proj.bias": "model.safetensors",
482
+ "vision_tower.layers.2.self_attn.v_proj.weight": "model.safetensors",
483
+ "vision_tower.layers.20.input_layernorm.bias": "model.safetensors",
484
+ "vision_tower.layers.20.input_layernorm.weight": "model.safetensors",
485
+ "vision_tower.layers.20.mlp.dense_4h_to_h.bias": "model.safetensors",
486
+ "vision_tower.layers.20.mlp.dense_4h_to_h.weight": "model.safetensors",
487
+ "vision_tower.layers.20.mlp.dense_h_to_4h.bias": "model.safetensors",
488
+ "vision_tower.layers.20.mlp.dense_h_to_4h.weight": "model.safetensors",
489
+ "vision_tower.layers.20.post_attention_layernorm.bias": "model.safetensors",
490
+ "vision_tower.layers.20.post_attention_layernorm.weight": "model.safetensors",
491
+ "vision_tower.layers.20.self_attn.k_proj.bias": "model.safetensors",
492
+ "vision_tower.layers.20.self_attn.k_proj.weight": "model.safetensors",
493
+ "vision_tower.layers.20.self_attn.o_proj.bias": "model.safetensors",
494
+ "vision_tower.layers.20.self_attn.o_proj.weight": "model.safetensors",
495
+ "vision_tower.layers.20.self_attn.q_proj.bias": "model.safetensors",
496
+ "vision_tower.layers.20.self_attn.q_proj.weight": "model.safetensors",
497
+ "vision_tower.layers.20.self_attn.v_proj.bias": "model.safetensors",
498
+ "vision_tower.layers.20.self_attn.v_proj.weight": "model.safetensors",
499
+ "vision_tower.layers.21.input_layernorm.bias": "model.safetensors",
500
+ "vision_tower.layers.21.input_layernorm.weight": "model.safetensors",
501
+ "vision_tower.layers.21.mlp.dense_4h_to_h.bias": "model.safetensors",
502
+ "vision_tower.layers.21.mlp.dense_4h_to_h.weight": "model.safetensors",
503
+ "vision_tower.layers.21.mlp.dense_h_to_4h.bias": "model.safetensors",
504
+ "vision_tower.layers.21.mlp.dense_h_to_4h.weight": "model.safetensors",
505
+ "vision_tower.layers.21.post_attention_layernorm.bias": "model.safetensors",
506
+ "vision_tower.layers.21.post_attention_layernorm.weight": "model.safetensors",
507
+ "vision_tower.layers.21.self_attn.k_proj.bias": "model.safetensors",
508
+ "vision_tower.layers.21.self_attn.k_proj.weight": "model.safetensors",
509
+ "vision_tower.layers.21.self_attn.o_proj.bias": "model.safetensors",
510
+ "vision_tower.layers.21.self_attn.o_proj.weight": "model.safetensors",
511
+ "vision_tower.layers.21.self_attn.q_proj.bias": "model.safetensors",
512
+ "vision_tower.layers.21.self_attn.q_proj.weight": "model.safetensors",
513
+ "vision_tower.layers.21.self_attn.v_proj.bias": "model.safetensors",
514
+ "vision_tower.layers.21.self_attn.v_proj.weight": "model.safetensors",
515
+ "vision_tower.layers.22.input_layernorm.bias": "model.safetensors",
516
+ "vision_tower.layers.22.input_layernorm.weight": "model.safetensors",
517
+ "vision_tower.layers.22.mlp.dense_4h_to_h.bias": "model.safetensors",
518
+ "vision_tower.layers.22.mlp.dense_4h_to_h.weight": "model.safetensors",
519
+ "vision_tower.layers.22.mlp.dense_h_to_4h.bias": "model.safetensors",
520
+ "vision_tower.layers.22.mlp.dense_h_to_4h.weight": "model.safetensors",
521
+ "vision_tower.layers.22.post_attention_layernorm.bias": "model.safetensors",
522
+ "vision_tower.layers.22.post_attention_layernorm.weight": "model.safetensors",
523
+ "vision_tower.layers.22.self_attn.k_proj.bias": "model.safetensors",
524
+ "vision_tower.layers.22.self_attn.k_proj.weight": "model.safetensors",
525
+ "vision_tower.layers.22.self_attn.o_proj.bias": "model.safetensors",
526
+ "vision_tower.layers.22.self_attn.o_proj.weight": "model.safetensors",
527
+ "vision_tower.layers.22.self_attn.q_proj.bias": "model.safetensors",
528
+ "vision_tower.layers.22.self_attn.q_proj.weight": "model.safetensors",
529
+ "vision_tower.layers.22.self_attn.v_proj.bias": "model.safetensors",
530
+ "vision_tower.layers.22.self_attn.v_proj.weight": "model.safetensors",
531
+ "vision_tower.layers.23.input_layernorm.bias": "model.safetensors",
532
+ "vision_tower.layers.23.input_layernorm.weight": "model.safetensors",
533
+ "vision_tower.layers.23.mlp.dense_4h_to_h.bias": "model.safetensors",
534
+ "vision_tower.layers.23.mlp.dense_4h_to_h.weight": "model.safetensors",
535
+ "vision_tower.layers.23.mlp.dense_h_to_4h.bias": "model.safetensors",
536
+ "vision_tower.layers.23.mlp.dense_h_to_4h.weight": "model.safetensors",
537
+ "vision_tower.layers.23.post_attention_layernorm.bias": "model.safetensors",
538
+ "vision_tower.layers.23.post_attention_layernorm.weight": "model.safetensors",
539
+ "vision_tower.layers.23.self_attn.k_proj.bias": "model.safetensors",
540
+ "vision_tower.layers.23.self_attn.k_proj.weight": "model.safetensors",
541
+ "vision_tower.layers.23.self_attn.o_proj.bias": "model.safetensors",
542
+ "vision_tower.layers.23.self_attn.o_proj.weight": "model.safetensors",
543
+ "vision_tower.layers.23.self_attn.q_proj.bias": "model.safetensors",
544
+ "vision_tower.layers.23.self_attn.q_proj.weight": "model.safetensors",
545
+ "vision_tower.layers.23.self_attn.v_proj.bias": "model.safetensors",
546
+ "vision_tower.layers.23.self_attn.v_proj.weight": "model.safetensors",
547
+ "vision_tower.layers.24.input_layernorm.bias": "model.safetensors",
548
+ "vision_tower.layers.24.input_layernorm.weight": "model.safetensors",
549
+ "vision_tower.layers.24.mlp.dense_4h_to_h.bias": "model.safetensors",
550
+ "vision_tower.layers.24.mlp.dense_4h_to_h.weight": "model.safetensors",
551
+ "vision_tower.layers.24.mlp.dense_h_to_4h.bias": "model.safetensors",
552
+ "vision_tower.layers.24.mlp.dense_h_to_4h.weight": "model.safetensors",
553
+ "vision_tower.layers.24.post_attention_layernorm.bias": "model.safetensors",
554
+ "vision_tower.layers.24.post_attention_layernorm.weight": "model.safetensors",
555
+ "vision_tower.layers.24.self_attn.k_proj.bias": "model.safetensors",
556
+ "vision_tower.layers.24.self_attn.k_proj.weight": "model.safetensors",
557
+ "vision_tower.layers.24.self_attn.o_proj.bias": "model.safetensors",
558
+ "vision_tower.layers.24.self_attn.o_proj.weight": "model.safetensors",
559
+ "vision_tower.layers.24.self_attn.q_proj.bias": "model.safetensors",
560
+ "vision_tower.layers.24.self_attn.q_proj.weight": "model.safetensors",
561
+ "vision_tower.layers.24.self_attn.v_proj.bias": "model.safetensors",
562
+ "vision_tower.layers.24.self_attn.v_proj.weight": "model.safetensors",
563
+ "vision_tower.layers.25.input_layernorm.bias": "model.safetensors",
564
+ "vision_tower.layers.25.input_layernorm.weight": "model.safetensors",
565
+ "vision_tower.layers.25.mlp.dense_4h_to_h.bias": "model.safetensors",
566
+ "vision_tower.layers.25.mlp.dense_4h_to_h.weight": "model.safetensors",
567
+ "vision_tower.layers.25.mlp.dense_h_to_4h.bias": "model.safetensors",
568
+ "vision_tower.layers.25.mlp.dense_h_to_4h.weight": "model.safetensors",
569
+ "vision_tower.layers.25.post_attention_layernorm.bias": "model.safetensors",
570
+ "vision_tower.layers.25.post_attention_layernorm.weight": "model.safetensors",
571
+ "vision_tower.layers.25.self_attn.k_proj.bias": "model.safetensors",
572
+ "vision_tower.layers.25.self_attn.k_proj.weight": "model.safetensors",
573
+ "vision_tower.layers.25.self_attn.o_proj.bias": "model.safetensors",
574
+ "vision_tower.layers.25.self_attn.o_proj.weight": "model.safetensors",
575
+ "vision_tower.layers.25.self_attn.q_proj.bias": "model.safetensors",
576
+ "vision_tower.layers.25.self_attn.q_proj.weight": "model.safetensors",
577
+ "vision_tower.layers.25.self_attn.v_proj.bias": "model.safetensors",
578
+ "vision_tower.layers.25.self_attn.v_proj.weight": "model.safetensors",
579
+ "vision_tower.layers.26.input_layernorm.bias": "model.safetensors",
580
+ "vision_tower.layers.26.input_layernorm.weight": "model.safetensors",
581
+ "vision_tower.layers.26.mlp.dense_4h_to_h.bias": "model.safetensors",
582
+ "vision_tower.layers.26.mlp.dense_4h_to_h.weight": "model.safetensors",
583
+ "vision_tower.layers.26.mlp.dense_h_to_4h.bias": "model.safetensors",
584
+ "vision_tower.layers.26.mlp.dense_h_to_4h.weight": "model.safetensors",
585
+ "vision_tower.layers.26.post_attention_layernorm.bias": "model.safetensors",
586
+ "vision_tower.layers.26.post_attention_layernorm.weight": "model.safetensors",
587
+ "vision_tower.layers.26.self_attn.k_proj.bias": "model.safetensors",
588
+ "vision_tower.layers.26.self_attn.k_proj.weight": "model.safetensors",
589
+ "vision_tower.layers.26.self_attn.o_proj.bias": "model.safetensors",
590
+ "vision_tower.layers.26.self_attn.o_proj.weight": "model.safetensors",
591
+ "vision_tower.layers.26.self_attn.q_proj.bias": "model.safetensors",
592
+ "vision_tower.layers.26.self_attn.q_proj.weight": "model.safetensors",
593
+ "vision_tower.layers.26.self_attn.v_proj.bias": "model.safetensors",
594
+ "vision_tower.layers.26.self_attn.v_proj.weight": "model.safetensors",
595
+ "vision_tower.layers.3.input_layernorm.bias": "model.safetensors",
596
+ "vision_tower.layers.3.input_layernorm.weight": "model.safetensors",
597
+ "vision_tower.layers.3.mlp.dense_4h_to_h.bias": "model.safetensors",
598
+ "vision_tower.layers.3.mlp.dense_4h_to_h.weight": "model.safetensors",
599
+ "vision_tower.layers.3.mlp.dense_h_to_4h.bias": "model.safetensors",
600
+ "vision_tower.layers.3.mlp.dense_h_to_4h.weight": "model.safetensors",
601
+ "vision_tower.layers.3.post_attention_layernorm.bias": "model.safetensors",
602
+ "vision_tower.layers.3.post_attention_layernorm.weight": "model.safetensors",
603
+ "vision_tower.layers.3.self_attn.k_proj.bias": "model.safetensors",
604
+ "vision_tower.layers.3.self_attn.k_proj.weight": "model.safetensors",
605
+ "vision_tower.layers.3.self_attn.o_proj.bias": "model.safetensors",
606
+ "vision_tower.layers.3.self_attn.o_proj.weight": "model.safetensors",
607
+ "vision_tower.layers.3.self_attn.q_proj.bias": "model.safetensors",
608
+ "vision_tower.layers.3.self_attn.q_proj.weight": "model.safetensors",
609
+ "vision_tower.layers.3.self_attn.v_proj.bias": "model.safetensors",
610
+ "vision_tower.layers.3.self_attn.v_proj.weight": "model.safetensors",
611
+ "vision_tower.layers.4.input_layernorm.bias": "model.safetensors",
612
+ "vision_tower.layers.4.input_layernorm.weight": "model.safetensors",
613
+ "vision_tower.layers.4.mlp.dense_4h_to_h.bias": "model.safetensors",
614
+ "vision_tower.layers.4.mlp.dense_4h_to_h.weight": "model.safetensors",
615
+ "vision_tower.layers.4.mlp.dense_h_to_4h.bias": "model.safetensors",
616
+ "vision_tower.layers.4.mlp.dense_h_to_4h.weight": "model.safetensors",
617
+ "vision_tower.layers.4.post_attention_layernorm.bias": "model.safetensors",
618
+ "vision_tower.layers.4.post_attention_layernorm.weight": "model.safetensors",
619
+ "vision_tower.layers.4.self_attn.k_proj.bias": "model.safetensors",
620
+ "vision_tower.layers.4.self_attn.k_proj.weight": "model.safetensors",
621
+ "vision_tower.layers.4.self_attn.o_proj.bias": "model.safetensors",
622
+ "vision_tower.layers.4.self_attn.o_proj.weight": "model.safetensors",
623
+ "vision_tower.layers.4.self_attn.q_proj.bias": "model.safetensors",
624
+ "vision_tower.layers.4.self_attn.q_proj.weight": "model.safetensors",
625
+ "vision_tower.layers.4.self_attn.v_proj.bias": "model.safetensors",
626
+ "vision_tower.layers.4.self_attn.v_proj.weight": "model.safetensors",
627
+ "vision_tower.layers.5.input_layernorm.bias": "model.safetensors",
628
+ "vision_tower.layers.5.input_layernorm.weight": "model.safetensors",
629
+ "vision_tower.layers.5.mlp.dense_4h_to_h.bias": "model.safetensors",
630
+ "vision_tower.layers.5.mlp.dense_4h_to_h.weight": "model.safetensors",
631
+ "vision_tower.layers.5.mlp.dense_h_to_4h.bias": "model.safetensors",
632
+ "vision_tower.layers.5.mlp.dense_h_to_4h.weight": "model.safetensors",
633
+ "vision_tower.layers.5.post_attention_layernorm.bias": "model.safetensors",
634
+ "vision_tower.layers.5.post_attention_layernorm.weight": "model.safetensors",
635
+ "vision_tower.layers.5.self_attn.k_proj.bias": "model.safetensors",
636
+ "vision_tower.layers.5.self_attn.k_proj.weight": "model.safetensors",
637
+ "vision_tower.layers.5.self_attn.o_proj.bias": "model.safetensors",
638
+ "vision_tower.layers.5.self_attn.o_proj.weight": "model.safetensors",
639
+ "vision_tower.layers.5.self_attn.q_proj.bias": "model.safetensors",
640
+ "vision_tower.layers.5.self_attn.q_proj.weight": "model.safetensors",
641
+ "vision_tower.layers.5.self_attn.v_proj.bias": "model.safetensors",
642
+ "vision_tower.layers.5.self_attn.v_proj.weight": "model.safetensors",
643
+ "vision_tower.layers.6.input_layernorm.bias": "model.safetensors",
644
+ "vision_tower.layers.6.input_layernorm.weight": "model.safetensors",
645
+ "vision_tower.layers.6.mlp.dense_4h_to_h.bias": "model.safetensors",
646
+ "vision_tower.layers.6.mlp.dense_4h_to_h.weight": "model.safetensors",
647
+ "vision_tower.layers.6.mlp.dense_h_to_4h.bias": "model.safetensors",
648
+ "vision_tower.layers.6.mlp.dense_h_to_4h.weight": "model.safetensors",
649
+ "vision_tower.layers.6.post_attention_layernorm.bias": "model.safetensors",
650
+ "vision_tower.layers.6.post_attention_layernorm.weight": "model.safetensors",
651
+ "vision_tower.layers.6.self_attn.k_proj.bias": "model.safetensors",
652
+ "vision_tower.layers.6.self_attn.k_proj.weight": "model.safetensors",
653
+ "vision_tower.layers.6.self_attn.o_proj.bias": "model.safetensors",
654
+ "vision_tower.layers.6.self_attn.o_proj.weight": "model.safetensors",
655
+ "vision_tower.layers.6.self_attn.q_proj.bias": "model.safetensors",
656
+ "vision_tower.layers.6.self_attn.q_proj.weight": "model.safetensors",
657
+ "vision_tower.layers.6.self_attn.v_proj.bias": "model.safetensors",
658
+ "vision_tower.layers.6.self_attn.v_proj.weight": "model.safetensors",
659
+ "vision_tower.layers.7.input_layernorm.bias": "model.safetensors",
660
+ "vision_tower.layers.7.input_layernorm.weight": "model.safetensors",
661
+ "vision_tower.layers.7.mlp.dense_4h_to_h.bias": "model.safetensors",
662
+ "vision_tower.layers.7.mlp.dense_4h_to_h.weight": "model.safetensors",
663
+ "vision_tower.layers.7.mlp.dense_h_to_4h.bias": "model.safetensors",
664
+ "vision_tower.layers.7.mlp.dense_h_to_4h.weight": "model.safetensors",
665
+ "vision_tower.layers.7.post_attention_layernorm.bias": "model.safetensors",
666
+ "vision_tower.layers.7.post_attention_layernorm.weight": "model.safetensors",
667
+ "vision_tower.layers.7.self_attn.k_proj.bias": "model.safetensors",
668
+ "vision_tower.layers.7.self_attn.k_proj.weight": "model.safetensors",
669
+ "vision_tower.layers.7.self_attn.o_proj.bias": "model.safetensors",
670
+ "vision_tower.layers.7.self_attn.o_proj.weight": "model.safetensors",
671
+ "vision_tower.layers.7.self_attn.q_proj.bias": "model.safetensors",
672
+ "vision_tower.layers.7.self_attn.q_proj.weight": "model.safetensors",
673
+ "vision_tower.layers.7.self_attn.v_proj.bias": "model.safetensors",
674
+ "vision_tower.layers.7.self_attn.v_proj.weight": "model.safetensors",
675
+ "vision_tower.layers.8.input_layernorm.bias": "model.safetensors",
676
+ "vision_tower.layers.8.input_layernorm.weight": "model.safetensors",
677
+ "vision_tower.layers.8.mlp.dense_4h_to_h.bias": "model.safetensors",
678
+ "vision_tower.layers.8.mlp.dense_4h_to_h.weight": "model.safetensors",
679
+ "vision_tower.layers.8.mlp.dense_h_to_4h.bias": "model.safetensors",
680
+ "vision_tower.layers.8.mlp.dense_h_to_4h.weight": "model.safetensors",
681
+ "vision_tower.layers.8.post_attention_layernorm.bias": "model.safetensors",
682
+ "vision_tower.layers.8.post_attention_layernorm.weight": "model.safetensors",
683
+ "vision_tower.layers.8.self_attn.k_proj.bias": "model.safetensors",
684
+ "vision_tower.layers.8.self_attn.k_proj.weight": "model.safetensors",
685
+ "vision_tower.layers.8.self_attn.o_proj.bias": "model.safetensors",
686
+ "vision_tower.layers.8.self_attn.o_proj.weight": "model.safetensors",
687
+ "vision_tower.layers.8.self_attn.q_proj.bias": "model.safetensors",
688
+ "vision_tower.layers.8.self_attn.q_proj.weight": "model.safetensors",
689
+ "vision_tower.layers.8.self_attn.v_proj.bias": "model.safetensors",
690
+ "vision_tower.layers.8.self_attn.v_proj.weight": "model.safetensors",
691
+ "vision_tower.layers.9.input_layernorm.bias": "model.safetensors",
692
+ "vision_tower.layers.9.input_layernorm.weight": "model.safetensors",
693
+ "vision_tower.layers.9.mlp.dense_4h_to_h.bias": "model.safetensors",
694
+ "vision_tower.layers.9.mlp.dense_4h_to_h.weight": "model.safetensors",
695
+ "vision_tower.layers.9.mlp.dense_h_to_4h.bias": "model.safetensors",
696
+ "vision_tower.layers.9.mlp.dense_h_to_4h.weight": "model.safetensors",
697
+ "vision_tower.layers.9.post_attention_layernorm.bias": "model.safetensors",
698
+ "vision_tower.layers.9.post_attention_layernorm.weight": "model.safetensors",
699
+ "vision_tower.layers.9.self_attn.k_proj.bias": "model.safetensors",
700
+ "vision_tower.layers.9.self_attn.k_proj.weight": "model.safetensors",
701
+ "vision_tower.layers.9.self_attn.o_proj.bias": "model.safetensors",
702
+ "vision_tower.layers.9.self_attn.o_proj.weight": "model.safetensors",
703
+ "vision_tower.layers.9.self_attn.q_proj.bias": "model.safetensors",
704
+ "vision_tower.layers.9.self_attn.q_proj.weight": "model.safetensors",
705
+ "vision_tower.layers.9.self_attn.v_proj.bias": "model.safetensors",
706
+ "vision_tower.layers.9.self_attn.v_proj.weight": "model.safetensors",
707
+ "vision_tower.perceive.after_rms.weight": "model.safetensors",
708
+ "vision_tower.perceive.before_rms.weight": "model.safetensors",
709
+ "vision_tower.perceive.image_begin": "model.safetensors",
710
+ "vision_tower.perceive.image_end": "model.safetensors",
711
+ "vision_tower.perceive.image_newline": "model.safetensors",
712
+ "vision_tower.perceive.image_sep": "model.safetensors",
713
+ "vision_tower.perceive.mlp.bias": "model.safetensors",
714
+ "vision_tower.perceive.mlp.weight": "model.safetensors",
715
+ "vision_tower.perceive.proj.0.bias": "model.safetensors",
716
+ "vision_tower.perceive.proj.0.weight": "model.safetensors",
717
+ "vision_tower.perceive.proj.2.bias": "model.safetensors",
718
+ "vision_tower.perceive.proj.2.weight": "model.safetensors"
719
+ }
720
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "min_pixels": 262144,
3
+ "max_pixels": 4194304,
4
+ "patch_size": 16,
5
+ "resample": 1,
6
+ "temporal_patch_size": 1,
7
+ "merge_size": 2,
8
+ "image_mean": [
9
+ 0.48145466,
10
+ 0.4578275,
11
+ 0.40821073
12
+ ],
13
+ "image_std": [
14
+ 0.26862954,
15
+ 0.26130258,
16
+ 0.27577711
17
+ ],
18
+ "image_processor_type": "HunYuanVLImageProcessor",
19
+ "processor_class": "HunYuanVLProcessor"
20
+ }
processor_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_processor": {
3
+ "_from_auto": true,
4
+ "do_convert_rgb": true,
5
+ "do_normalize": true,
6
+ "do_resize": true,
7
+ "image_mean": [
8
+ 0.48145466,
9
+ 0.4578275,
10
+ 0.40821073
11
+ ],
12
+ "image_processor_type": "HunYuanVLImageProcessor",
13
+ "image_std": [
14
+ 0.26862954,
15
+ 0.26130258,
16
+ 0.27577711
17
+ ],
18
+ "max_pixels": 4194304,
19
+ "merge_size": 2,
20
+ "min_pixels": 262144,
21
+ "patch_size": 16,
22
+ "temporal_patch_size": 1,
23
+ "use_fast": true
24
+ },
25
+ "processor_class": "HunYuanVLProcessor"
26
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<｜hy_begin▁of▁sentence｜>",
3
+ "eos_token": "<｜hy_place▁holder▁no▁2｜>",
4
+ "pad_token": "<｜hy_▁pad▁｜>"
5
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "bos_token": "<｜hy_begin▁of▁sentence｜>",
4
+ "clean_up_tokenization_spaces": true,
5
+ "eos_token": "<｜hy_Assistant｜>",
6
+ "image_end_token": "<｜hy_place▁holder▁no▁101｜>",
7
+ "image_start_token": "<｜hy_place▁holder▁no▁100｜>",
8
+ "image_token": "<｜hy_place▁holder▁no▁102｜>",
9
+ "is_local": true,
10
+ "model_max_length": 1000000000000000019884624838656,
11
+ "model_specific_special_tokens": {
12
+ "image_end_token": "<｜hy_place▁holder▁no▁101｜>",
13
+ "image_start_token": "<｜hy_place▁holder▁no▁100｜>",
14
+ "image_token": "<｜hy_place▁holder▁no▁102｜>",
15
+ "video_end_token": "<｜hy_place▁holder▁no▁105｜>",
16
+ "video_start_token": "<｜hy_place▁holder▁no▁104｜>"
17
+ },
18
+ "pad_token": "<｜hy_▁pad▁｜>",
19
+ "processor_class": "HunYuanVLProcessor",
20
+ "tokenizer_class": "TokenizersBackend",
21
+ "video_end_token": "<｜hy_place▁holder▁no▁105｜>",
22
+ "video_start_token": "<｜hy_place▁holder▁no▁104｜>"
23
+ }