hermeschen-ezcon commited on
Commit
ce33218
·
verified ·
1 Parent(s): 6b7e8d1

Upload folder using huggingface_hub

Browse files
README.md CHANGED
@@ -1,4 +1,6 @@
1
  ---
 
 
2
  license_name: qwen-research
3
  license_link: https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct/blob/main/LICENSE
4
  language:
@@ -6,13 +8,14 @@ language:
6
  pipeline_tag: image-text-to-text
7
  tags:
8
  - multimodal
 
9
  - mlx
10
  library_name: transformers
11
  ---
12
 
13
  # EZCon/Qwen2.5-VL-3B-Instruct-8bit-mlx
14
- This model was converted to MLX format from [`Qwen/Qwen2.5-VL-3B-Instruct`]() using mlx-vlm version **0.1.21**.
15
- Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct) for more details on the model.
16
  ## Use with mlx
17
 
18
  ```bash
 
1
  ---
2
+ base_model:
3
+ - Qwen/Qwen2.5-VL-3B-Instruct
4
  license_name: qwen-research
5
  license_link: https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct/blob/main/LICENSE
6
  language:
 
8
  pipeline_tag: image-text-to-text
9
  tags:
10
  - multimodal
11
+ - unsloth
12
  - mlx
13
  library_name: transformers
14
  ---
15
 
16
  # EZCon/Qwen2.5-VL-3B-Instruct-8bit-mlx
17
+ This model was converted to MLX format from [`unsloth/Qwen2.5-VL-3B-Instruct`]() using mlx-vlm version **0.3.2**.
18
+ Refer to the [original model card](https://huggingface.co/unsloth/Qwen2.5-VL-3B-Instruct) for more details on the model.
19
  ## Use with mlx
20
 
21
  ```bash
chat_template.json CHANGED
@@ -1,3 +1,3 @@
1
  {
2
- "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
3
- }
 
1
  {
2
+ "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
3
+ }
config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_attn_implementation_autoset": false,
3
  "add_cross_attention": false,
4
  "architectures": [
5
  "Qwen2_5_VLForConditionalGeneration"
@@ -7,7 +6,7 @@
7
  "attention_dropout": 0.0,
8
  "bad_words_ids": null,
9
  "begin_suppress_tokens": null,
10
- "bos_token_id": 151643,
11
  "chunk_size_feed_forward": 0,
12
  "cross_attention_hidden_size": null,
13
  "decoder_start_token_id": null,
@@ -51,13 +50,1129 @@
51
  "output_attentions": false,
52
  "output_hidden_states": false,
53
  "output_scores": false,
54
- "pad_token_id": null,
55
  "prefix": null,
56
  "problem_type": null,
57
  "pruned_heads": {},
58
  "quantization": {
59
  "group_size": 64,
60
- "bits": 8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
61
  },
62
  "remove_invalid_values": false,
63
  "repetition_penalty": 1.0,
@@ -65,13 +1180,13 @@
65
  "return_dict_in_generate": false,
66
  "rms_norm_eps": 1e-06,
67
  "rope_scaling": {
68
- "type": "default",
69
  "mrope_section": [
70
  16,
71
  24,
72
  24
73
  ],
74
- "rope_type": "default"
 
75
  },
76
  "rope_theta": 1000000.0,
77
  "sep_token_id": null,
@@ -87,7 +1202,7 @@
87
  "num_hidden_layers": 36,
88
  "num_attention_heads": 16,
89
  "use_sliding_window": false,
90
- "sliding_window": 32768,
91
  "max_window_layers": 70,
92
  "num_key_value_heads": 2,
93
  "hidden_act": "silu",
@@ -97,21 +1212,58 @@
97
  "rope_theta": 1000000.0,
98
  "attention_dropout": 0.0,
99
  "rope_scaling": {
100
- "type": "default",
101
  "mrope_section": [
102
  16,
103
  24,
104
  24
105
  ],
106
- "rope_type": "default"
 
107
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
108
  "return_dict": true,
109
  "output_hidden_states": false,
110
- "output_attentions": false,
111
  "torchscript": false,
112
  "torch_dtype": "bfloat16",
113
- "use_bfloat16": false,
114
- "tf_legacy_loss": false,
115
  "pruned_heads": {},
116
  "tie_word_embeddings": true,
117
  "chunk_size_feed_forward": 0,
@@ -120,6 +1272,27 @@
120
  "cross_attention_hidden_size": null,
121
  "add_cross_attention": false,
122
  "tie_encoder_decoder": false,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
123
  "max_length": 20,
124
  "min_length": 0,
125
  "do_sample": false,
@@ -145,33 +1318,15 @@
145
  "exponential_decay_length_penalty": null,
146
  "suppress_tokens": null,
147
  "begin_suppress_tokens": null,
148
- "architectures": [
149
- "Qwen2_5_VLForConditionalGeneration"
150
- ],
151
- "finetuning_task": null,
152
- "id2label": {
153
- "0": "LABEL_0",
154
- "1": "LABEL_1"
155
- },
156
- "label2id": {
157
- "LABEL_0": 0,
158
- "LABEL_1": 1
159
- },
160
- "tokenizer_class": null,
161
- "prefix": null,
162
- "bos_token_id": 151643,
163
- "pad_token_id": null,
164
- "eos_token_id": 151645,
165
- "sep_token_id": null,
166
- "decoder_start_token_id": null,
167
- "task_specific_params": null,
168
- "problem_type": null,
169
  "_name_or_path": "",
170
- "_attn_implementation_autoset": false,
171
- "vision_start_token_id": 151652,
172
  "vision_end_token_id": 151653,
 
173
  "vision_token_id": 151654,
174
- "model_type": "qwen2_5_vl_text"
 
 
175
  },
176
  "tf_legacy_loss": false,
177
  "tie_encoder_decoder": false,
@@ -179,10 +1334,10 @@
179
  "tokenizer_class": null,
180
  "top_k": 50,
181
  "top_p": 1.0,
182
- "torch_dtype": "bfloat16",
183
  "torchscript": false,
184
- "transformers_version": "4.52.0.dev0",
185
  "typical_p": 1.0,
 
186
  "use_bfloat16": false,
187
  "use_cache": true,
188
  "use_sliding_window": false,
@@ -190,11 +1345,8 @@
190
  "vision_config": {
191
  "return_dict": true,
192
  "output_hidden_states": false,
193
- "output_attentions": false,
194
  "torchscript": false,
195
- "torch_dtype": null,
196
- "use_bfloat16": false,
197
- "tf_legacy_loss": false,
198
  "pruned_heads": {},
199
  "tie_word_embeddings": true,
200
  "chunk_size_feed_forward": 0,
@@ -203,6 +1355,25 @@
203
  "cross_attention_hidden_size": null,
204
  "add_cross_attention": false,
205
  "tie_encoder_decoder": false,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
206
  "max_length": 20,
207
  "min_length": 0,
208
  "do_sample": false,
@@ -228,29 +1399,12 @@
228
  "exponential_decay_length_penalty": null,
229
  "suppress_tokens": null,
230
  "begin_suppress_tokens": null,
231
- "architectures": null,
232
- "finetuning_task": null,
233
- "id2label": {
234
- "0": "LABEL_0",
235
- "1": "LABEL_1"
236
- },
237
- "label2id": {
238
- "LABEL_0": 0,
239
- "LABEL_1": 1
240
- },
241
- "tokenizer_class": null,
242
- "prefix": null,
243
- "bos_token_id": null,
244
- "pad_token_id": null,
245
- "eos_token_id": null,
246
- "sep_token_id": null,
247
- "decoder_start_token_id": null,
248
- "task_specific_params": null,
249
- "problem_type": null,
250
  "_name_or_path": "",
251
- "_attn_implementation_autoset": false,
252
  "in_chans": 3,
 
253
  "spatial_patch_size": 14,
 
 
254
  "depth": 32,
255
  "hidden_size": 1280,
256
  "hidden_act": "silu",
@@ -270,7 +1424,7 @@
270
  ],
271
  "out_hidden_size": 2048,
272
  "initializer_range": 0.02,
273
- "model_type": "qwen2_5_vl"
274
  },
275
  "vision_end_token_id": 151653,
276
  "vision_start_token_id": 151652,
 
1
  {
 
2
  "add_cross_attention": false,
3
  "architectures": [
4
  "Qwen2_5_VLForConditionalGeneration"
 
6
  "attention_dropout": 0.0,
7
  "bad_words_ids": null,
8
  "begin_suppress_tokens": null,
9
+ "bos_token_id": null,
10
  "chunk_size_feed_forward": 0,
11
  "cross_attention_hidden_size": null,
12
  "decoder_start_token_id": null,
 
50
  "output_attentions": false,
51
  "output_hidden_states": false,
52
  "output_scores": false,
53
+ "pad_token_id": 151654,
54
  "prefix": null,
55
  "problem_type": null,
56
  "pruned_heads": {},
57
  "quantization": {
58
  "group_size": 64,
59
+ "bits": 8,
60
+ "vision_tower.patch_embed.proj": false,
61
+ "vision_tower.rotary_pos_emb": false,
62
+ "vision_tower.blocks.0.norm1": false,
63
+ "vision_tower.blocks.0.norm2": false,
64
+ "vision_tower.blocks.0.attn.qkv": false,
65
+ "vision_tower.blocks.0.attn.proj": false,
66
+ "vision_tower.blocks.0.mlp.gate_proj": false,
67
+ "vision_tower.blocks.0.mlp.up_proj": false,
68
+ "vision_tower.blocks.0.mlp.down_proj": false,
69
+ "vision_tower.blocks.1.norm1": false,
70
+ "vision_tower.blocks.1.norm2": false,
71
+ "vision_tower.blocks.1.attn.qkv": false,
72
+ "vision_tower.blocks.1.attn.proj": false,
73
+ "vision_tower.blocks.1.mlp.gate_proj": false,
74
+ "vision_tower.blocks.1.mlp.up_proj": false,
75
+ "vision_tower.blocks.1.mlp.down_proj": false,
76
+ "vision_tower.blocks.2.norm1": false,
77
+ "vision_tower.blocks.2.norm2": false,
78
+ "vision_tower.blocks.2.attn.qkv": false,
79
+ "vision_tower.blocks.2.attn.proj": false,
80
+ "vision_tower.blocks.2.mlp.gate_proj": false,
81
+ "vision_tower.blocks.2.mlp.up_proj": false,
82
+ "vision_tower.blocks.2.mlp.down_proj": false,
83
+ "vision_tower.blocks.3.norm1": false,
84
+ "vision_tower.blocks.3.norm2": false,
85
+ "vision_tower.blocks.3.attn.qkv": false,
86
+ "vision_tower.blocks.3.attn.proj": false,
87
+ "vision_tower.blocks.3.mlp.gate_proj": false,
88
+ "vision_tower.blocks.3.mlp.up_proj": false,
89
+ "vision_tower.blocks.3.mlp.down_proj": false,
90
+ "vision_tower.blocks.4.norm1": false,
91
+ "vision_tower.blocks.4.norm2": false,
92
+ "vision_tower.blocks.4.attn.qkv": false,
93
+ "vision_tower.blocks.4.attn.proj": false,
94
+ "vision_tower.blocks.4.mlp.gate_proj": false,
95
+ "vision_tower.blocks.4.mlp.up_proj": false,
96
+ "vision_tower.blocks.4.mlp.down_proj": false,
97
+ "vision_tower.blocks.5.norm1": false,
98
+ "vision_tower.blocks.5.norm2": false,
99
+ "vision_tower.blocks.5.attn.qkv": false,
100
+ "vision_tower.blocks.5.attn.proj": false,
101
+ "vision_tower.blocks.5.mlp.gate_proj": false,
102
+ "vision_tower.blocks.5.mlp.up_proj": false,
103
+ "vision_tower.blocks.5.mlp.down_proj": false,
104
+ "vision_tower.blocks.6.norm1": false,
105
+ "vision_tower.blocks.6.norm2": false,
106
+ "vision_tower.blocks.6.attn.qkv": false,
107
+ "vision_tower.blocks.6.attn.proj": false,
108
+ "vision_tower.blocks.6.mlp.gate_proj": false,
109
+ "vision_tower.blocks.6.mlp.up_proj": false,
110
+ "vision_tower.blocks.6.mlp.down_proj": false,
111
+ "vision_tower.blocks.7.norm1": false,
112
+ "vision_tower.blocks.7.norm2": false,
113
+ "vision_tower.blocks.7.attn.qkv": false,
114
+ "vision_tower.blocks.7.attn.proj": false,
115
+ "vision_tower.blocks.7.mlp.gate_proj": false,
116
+ "vision_tower.blocks.7.mlp.up_proj": false,
117
+ "vision_tower.blocks.7.mlp.down_proj": false,
118
+ "vision_tower.blocks.8.norm1": false,
119
+ "vision_tower.blocks.8.norm2": false,
120
+ "vision_tower.blocks.8.attn.qkv": false,
121
+ "vision_tower.blocks.8.attn.proj": false,
122
+ "vision_tower.blocks.8.mlp.gate_proj": false,
123
+ "vision_tower.blocks.8.mlp.up_proj": false,
124
+ "vision_tower.blocks.8.mlp.down_proj": false,
125
+ "vision_tower.blocks.9.norm1": false,
126
+ "vision_tower.blocks.9.norm2": false,
127
+ "vision_tower.blocks.9.attn.qkv": false,
128
+ "vision_tower.blocks.9.attn.proj": false,
129
+ "vision_tower.blocks.9.mlp.gate_proj": false,
130
+ "vision_tower.blocks.9.mlp.up_proj": false,
131
+ "vision_tower.blocks.9.mlp.down_proj": false,
132
+ "vision_tower.blocks.10.norm1": false,
133
+ "vision_tower.blocks.10.norm2": false,
134
+ "vision_tower.blocks.10.attn.qkv": false,
135
+ "vision_tower.blocks.10.attn.proj": false,
136
+ "vision_tower.blocks.10.mlp.gate_proj": false,
137
+ "vision_tower.blocks.10.mlp.up_proj": false,
138
+ "vision_tower.blocks.10.mlp.down_proj": false,
139
+ "vision_tower.blocks.11.norm1": false,
140
+ "vision_tower.blocks.11.norm2": false,
141
+ "vision_tower.blocks.11.attn.qkv": false,
142
+ "vision_tower.blocks.11.attn.proj": false,
143
+ "vision_tower.blocks.11.mlp.gate_proj": false,
144
+ "vision_tower.blocks.11.mlp.up_proj": false,
145
+ "vision_tower.blocks.11.mlp.down_proj": false,
146
+ "vision_tower.blocks.12.norm1": false,
147
+ "vision_tower.blocks.12.norm2": false,
148
+ "vision_tower.blocks.12.attn.qkv": false,
149
+ "vision_tower.blocks.12.attn.proj": false,
150
+ "vision_tower.blocks.12.mlp.gate_proj": false,
151
+ "vision_tower.blocks.12.mlp.up_proj": false,
152
+ "vision_tower.blocks.12.mlp.down_proj": false,
153
+ "vision_tower.blocks.13.norm1": false,
154
+ "vision_tower.blocks.13.norm2": false,
155
+ "vision_tower.blocks.13.attn.qkv": false,
156
+ "vision_tower.blocks.13.attn.proj": false,
157
+ "vision_tower.blocks.13.mlp.gate_proj": false,
158
+ "vision_tower.blocks.13.mlp.up_proj": false,
159
+ "vision_tower.blocks.13.mlp.down_proj": false,
160
+ "vision_tower.blocks.14.norm1": false,
161
+ "vision_tower.blocks.14.norm2": false,
162
+ "vision_tower.blocks.14.attn.qkv": false,
163
+ "vision_tower.blocks.14.attn.proj": false,
164
+ "vision_tower.blocks.14.mlp.gate_proj": false,
165
+ "vision_tower.blocks.14.mlp.up_proj": false,
166
+ "vision_tower.blocks.14.mlp.down_proj": false,
167
+ "vision_tower.blocks.15.norm1": false,
168
+ "vision_tower.blocks.15.norm2": false,
169
+ "vision_tower.blocks.15.attn.qkv": false,
170
+ "vision_tower.blocks.15.attn.proj": false,
171
+ "vision_tower.blocks.15.mlp.gate_proj": false,
172
+ "vision_tower.blocks.15.mlp.up_proj": false,
173
+ "vision_tower.blocks.15.mlp.down_proj": false,
174
+ "vision_tower.blocks.16.norm1": false,
175
+ "vision_tower.blocks.16.norm2": false,
176
+ "vision_tower.blocks.16.attn.qkv": false,
177
+ "vision_tower.blocks.16.attn.proj": false,
178
+ "vision_tower.blocks.16.mlp.gate_proj": false,
179
+ "vision_tower.blocks.16.mlp.up_proj": false,
180
+ "vision_tower.blocks.16.mlp.down_proj": false,
181
+ "vision_tower.blocks.17.norm1": false,
182
+ "vision_tower.blocks.17.norm2": false,
183
+ "vision_tower.blocks.17.attn.qkv": false,
184
+ "vision_tower.blocks.17.attn.proj": false,
185
+ "vision_tower.blocks.17.mlp.gate_proj": false,
186
+ "vision_tower.blocks.17.mlp.up_proj": false,
187
+ "vision_tower.blocks.17.mlp.down_proj": false,
188
+ "vision_tower.blocks.18.norm1": false,
189
+ "vision_tower.blocks.18.norm2": false,
190
+ "vision_tower.blocks.18.attn.qkv": false,
191
+ "vision_tower.blocks.18.attn.proj": false,
192
+ "vision_tower.blocks.18.mlp.gate_proj": false,
193
+ "vision_tower.blocks.18.mlp.up_proj": false,
194
+ "vision_tower.blocks.18.mlp.down_proj": false,
195
+ "vision_tower.blocks.19.norm1": false,
196
+ "vision_tower.blocks.19.norm2": false,
197
+ "vision_tower.blocks.19.attn.qkv": false,
198
+ "vision_tower.blocks.19.attn.proj": false,
199
+ "vision_tower.blocks.19.mlp.gate_proj": false,
200
+ "vision_tower.blocks.19.mlp.up_proj": false,
201
+ "vision_tower.blocks.19.mlp.down_proj": false,
202
+ "vision_tower.blocks.20.norm1": false,
203
+ "vision_tower.blocks.20.norm2": false,
204
+ "vision_tower.blocks.20.attn.qkv": false,
205
+ "vision_tower.blocks.20.attn.proj": false,
206
+ "vision_tower.blocks.20.mlp.gate_proj": false,
207
+ "vision_tower.blocks.20.mlp.up_proj": false,
208
+ "vision_tower.blocks.20.mlp.down_proj": false,
209
+ "vision_tower.blocks.21.norm1": false,
210
+ "vision_tower.blocks.21.norm2": false,
211
+ "vision_tower.blocks.21.attn.qkv": false,
212
+ "vision_tower.blocks.21.attn.proj": false,
213
+ "vision_tower.blocks.21.mlp.gate_proj": false,
214
+ "vision_tower.blocks.21.mlp.up_proj": false,
215
+ "vision_tower.blocks.21.mlp.down_proj": false,
216
+ "vision_tower.blocks.22.norm1": false,
217
+ "vision_tower.blocks.22.norm2": false,
218
+ "vision_tower.blocks.22.attn.qkv": false,
219
+ "vision_tower.blocks.22.attn.proj": false,
220
+ "vision_tower.blocks.22.mlp.gate_proj": false,
221
+ "vision_tower.blocks.22.mlp.up_proj": false,
222
+ "vision_tower.blocks.22.mlp.down_proj": false,
223
+ "vision_tower.blocks.23.norm1": false,
224
+ "vision_tower.blocks.23.norm2": false,
225
+ "vision_tower.blocks.23.attn.qkv": false,
226
+ "vision_tower.blocks.23.attn.proj": false,
227
+ "vision_tower.blocks.23.mlp.gate_proj": false,
228
+ "vision_tower.blocks.23.mlp.up_proj": false,
229
+ "vision_tower.blocks.23.mlp.down_proj": false,
230
+ "vision_tower.blocks.24.norm1": false,
231
+ "vision_tower.blocks.24.norm2": false,
232
+ "vision_tower.blocks.24.attn.qkv": false,
233
+ "vision_tower.blocks.24.attn.proj": false,
234
+ "vision_tower.blocks.24.mlp.gate_proj": false,
235
+ "vision_tower.blocks.24.mlp.up_proj": false,
236
+ "vision_tower.blocks.24.mlp.down_proj": false,
237
+ "vision_tower.blocks.25.norm1": false,
238
+ "vision_tower.blocks.25.norm2": false,
239
+ "vision_tower.blocks.25.attn.qkv": false,
240
+ "vision_tower.blocks.25.attn.proj": false,
241
+ "vision_tower.blocks.25.mlp.gate_proj": false,
242
+ "vision_tower.blocks.25.mlp.up_proj": false,
243
+ "vision_tower.blocks.25.mlp.down_proj": false,
244
+ "vision_tower.blocks.26.norm1": false,
245
+ "vision_tower.blocks.26.norm2": false,
246
+ "vision_tower.blocks.26.attn.qkv": false,
247
+ "vision_tower.blocks.26.attn.proj": false,
248
+ "vision_tower.blocks.26.mlp.gate_proj": false,
249
+ "vision_tower.blocks.26.mlp.up_proj": false,
250
+ "vision_tower.blocks.26.mlp.down_proj": false,
251
+ "vision_tower.blocks.27.norm1": false,
252
+ "vision_tower.blocks.27.norm2": false,
253
+ "vision_tower.blocks.27.attn.qkv": false,
254
+ "vision_tower.blocks.27.attn.proj": false,
255
+ "vision_tower.blocks.27.mlp.gate_proj": false,
256
+ "vision_tower.blocks.27.mlp.up_proj": false,
257
+ "vision_tower.blocks.27.mlp.down_proj": false,
258
+ "vision_tower.blocks.28.norm1": false,
259
+ "vision_tower.blocks.28.norm2": false,
260
+ "vision_tower.blocks.28.attn.qkv": false,
261
+ "vision_tower.blocks.28.attn.proj": false,
262
+ "vision_tower.blocks.28.mlp.gate_proj": false,
263
+ "vision_tower.blocks.28.mlp.up_proj": false,
264
+ "vision_tower.blocks.28.mlp.down_proj": false,
265
+ "vision_tower.blocks.29.norm1": false,
266
+ "vision_tower.blocks.29.norm2": false,
267
+ "vision_tower.blocks.29.attn.qkv": false,
268
+ "vision_tower.blocks.29.attn.proj": false,
269
+ "vision_tower.blocks.29.mlp.gate_proj": false,
270
+ "vision_tower.blocks.29.mlp.up_proj": false,
271
+ "vision_tower.blocks.29.mlp.down_proj": false,
272
+ "vision_tower.blocks.30.norm1": false,
273
+ "vision_tower.blocks.30.norm2": false,
274
+ "vision_tower.blocks.30.attn.qkv": false,
275
+ "vision_tower.blocks.30.attn.proj": false,
276
+ "vision_tower.blocks.30.mlp.gate_proj": false,
277
+ "vision_tower.blocks.30.mlp.up_proj": false,
278
+ "vision_tower.blocks.30.mlp.down_proj": false,
279
+ "vision_tower.blocks.31.norm1": false,
280
+ "vision_tower.blocks.31.norm2": false,
281
+ "vision_tower.blocks.31.attn.qkv": false,
282
+ "vision_tower.blocks.31.attn.proj": false,
283
+ "vision_tower.blocks.31.mlp.gate_proj": false,
284
+ "vision_tower.blocks.31.mlp.up_proj": false,
285
+ "vision_tower.blocks.31.mlp.down_proj": false,
286
+ "vision_tower.merger.ln_q": false,
287
+ "vision_tower.merger.mlp.0": false,
288
+ "vision_tower.merger.mlp.1": false,
289
+ "vision_tower.merger.mlp.2": false,
290
+ "language_model.model.embed_tokens": true,
291
+ "language_model.model.layers.0.self_attn.q_proj": true,
292
+ "language_model.model.layers.0.self_attn.k_proj": true,
293
+ "language_model.model.layers.0.self_attn.v_proj": true,
294
+ "language_model.model.layers.0.self_attn.o_proj": true,
295
+ "language_model.model.layers.0.mlp.gate_proj": true,
296
+ "language_model.model.layers.0.mlp.down_proj": true,
297
+ "language_model.model.layers.0.mlp.up_proj": true,
298
+ "language_model.model.layers.0.input_layernorm": false,
299
+ "language_model.model.layers.0.post_attention_layernorm": false,
300
+ "language_model.model.layers.1.self_attn.q_proj": true,
301
+ "language_model.model.layers.1.self_attn.k_proj": true,
302
+ "language_model.model.layers.1.self_attn.v_proj": true,
303
+ "language_model.model.layers.1.self_attn.o_proj": true,
304
+ "language_model.model.layers.1.mlp.gate_proj": true,
305
+ "language_model.model.layers.1.mlp.down_proj": true,
306
+ "language_model.model.layers.1.mlp.up_proj": true,
307
+ "language_model.model.layers.1.input_layernorm": false,
308
+ "language_model.model.layers.1.post_attention_layernorm": false,
309
+ "language_model.model.layers.2.self_attn.q_proj": true,
310
+ "language_model.model.layers.2.self_attn.k_proj": true,
311
+ "language_model.model.layers.2.self_attn.v_proj": true,
312
+ "language_model.model.layers.2.self_attn.o_proj": true,
313
+ "language_model.model.layers.2.mlp.gate_proj": true,
314
+ "language_model.model.layers.2.mlp.down_proj": true,
315
+ "language_model.model.layers.2.mlp.up_proj": true,
316
+ "language_model.model.layers.2.input_layernorm": false,
317
+ "language_model.model.layers.2.post_attention_layernorm": false,
318
+ "language_model.model.layers.3.self_attn.q_proj": true,
319
+ "language_model.model.layers.3.self_attn.k_proj": true,
320
+ "language_model.model.layers.3.self_attn.v_proj": true,
321
+ "language_model.model.layers.3.self_attn.o_proj": true,
322
+ "language_model.model.layers.3.mlp.gate_proj": true,
323
+ "language_model.model.layers.3.mlp.down_proj": true,
324
+ "language_model.model.layers.3.mlp.up_proj": true,
325
+ "language_model.model.layers.3.input_layernorm": false,
326
+ "language_model.model.layers.3.post_attention_layernorm": false,
327
+ "language_model.model.layers.4.self_attn.q_proj": true,
328
+ "language_model.model.layers.4.self_attn.k_proj": true,
329
+ "language_model.model.layers.4.self_attn.v_proj": true,
330
+ "language_model.model.layers.4.self_attn.o_proj": true,
331
+ "language_model.model.layers.4.mlp.gate_proj": true,
332
+ "language_model.model.layers.4.mlp.down_proj": true,
333
+ "language_model.model.layers.4.mlp.up_proj": true,
334
+ "language_model.model.layers.4.input_layernorm": false,
335
+ "language_model.model.layers.4.post_attention_layernorm": false,
336
+ "language_model.model.layers.5.self_attn.q_proj": true,
337
+ "language_model.model.layers.5.self_attn.k_proj": true,
338
+ "language_model.model.layers.5.self_attn.v_proj": true,
339
+ "language_model.model.layers.5.self_attn.o_proj": true,
340
+ "language_model.model.layers.5.mlp.gate_proj": true,
341
+ "language_model.model.layers.5.mlp.down_proj": true,
342
+ "language_model.model.layers.5.mlp.up_proj": true,
343
+ "language_model.model.layers.5.input_layernorm": false,
344
+ "language_model.model.layers.5.post_attention_layernorm": false,
345
+ "language_model.model.layers.6.self_attn.q_proj": true,
346
+ "language_model.model.layers.6.self_attn.k_proj": true,
347
+ "language_model.model.layers.6.self_attn.v_proj": true,
348
+ "language_model.model.layers.6.self_attn.o_proj": true,
349
+ "language_model.model.layers.6.mlp.gate_proj": true,
350
+ "language_model.model.layers.6.mlp.down_proj": true,
351
+ "language_model.model.layers.6.mlp.up_proj": true,
352
+ "language_model.model.layers.6.input_layernorm": false,
353
+ "language_model.model.layers.6.post_attention_layernorm": false,
354
+ "language_model.model.layers.7.self_attn.q_proj": true,
355
+ "language_model.model.layers.7.self_attn.k_proj": true,
356
+ "language_model.model.layers.7.self_attn.v_proj": true,
357
+ "language_model.model.layers.7.self_attn.o_proj": true,
358
+ "language_model.model.layers.7.mlp.gate_proj": true,
359
+ "language_model.model.layers.7.mlp.down_proj": true,
360
+ "language_model.model.layers.7.mlp.up_proj": true,
361
+ "language_model.model.layers.7.input_layernorm": false,
362
+ "language_model.model.layers.7.post_attention_layernorm": false,
363
+ "language_model.model.layers.8.self_attn.q_proj": true,
364
+ "language_model.model.layers.8.self_attn.k_proj": true,
365
+ "language_model.model.layers.8.self_attn.v_proj": true,
366
+ "language_model.model.layers.8.self_attn.o_proj": true,
367
+ "language_model.model.layers.8.mlp.gate_proj": true,
368
+ "language_model.model.layers.8.mlp.down_proj": true,
369
+ "language_model.model.layers.8.mlp.up_proj": true,
370
+ "language_model.model.layers.8.input_layernorm": false,
371
+ "language_model.model.layers.8.post_attention_layernorm": false,
372
+ "language_model.model.layers.9.self_attn.q_proj": true,
373
+ "language_model.model.layers.9.self_attn.k_proj": true,
374
+ "language_model.model.layers.9.self_attn.v_proj": true,
375
+ "language_model.model.layers.9.self_attn.o_proj": true,
376
+ "language_model.model.layers.9.mlp.gate_proj": true,
377
+ "language_model.model.layers.9.mlp.down_proj": true,
378
+ "language_model.model.layers.9.mlp.up_proj": true,
379
+ "language_model.model.layers.9.input_layernorm": false,
380
+ "language_model.model.layers.9.post_attention_layernorm": false,
381
+ "language_model.model.layers.10.self_attn.q_proj": true,
382
+ "language_model.model.layers.10.self_attn.k_proj": true,
383
+ "language_model.model.layers.10.self_attn.v_proj": true,
384
+ "language_model.model.layers.10.self_attn.o_proj": true,
385
+ "language_model.model.layers.10.mlp.gate_proj": true,
386
+ "language_model.model.layers.10.mlp.down_proj": true,
387
+ "language_model.model.layers.10.mlp.up_proj": true,
388
+ "language_model.model.layers.10.input_layernorm": false,
389
+ "language_model.model.layers.10.post_attention_layernorm": false,
390
+ "language_model.model.layers.11.self_attn.q_proj": true,
391
+ "language_model.model.layers.11.self_attn.k_proj": true,
392
+ "language_model.model.layers.11.self_attn.v_proj": true,
393
+ "language_model.model.layers.11.self_attn.o_proj": true,
394
+ "language_model.model.layers.11.mlp.gate_proj": true,
395
+ "language_model.model.layers.11.mlp.down_proj": true,
396
+ "language_model.model.layers.11.mlp.up_proj": true,
397
+ "language_model.model.layers.11.input_layernorm": false,
398
+ "language_model.model.layers.11.post_attention_layernorm": false,
399
+ "language_model.model.layers.12.self_attn.q_proj": true,
400
+ "language_model.model.layers.12.self_attn.k_proj": true,
401
+ "language_model.model.layers.12.self_attn.v_proj": true,
402
+ "language_model.model.layers.12.self_attn.o_proj": true,
403
+ "language_model.model.layers.12.mlp.gate_proj": true,
404
+ "language_model.model.layers.12.mlp.down_proj": true,
405
+ "language_model.model.layers.12.mlp.up_proj": true,
406
+ "language_model.model.layers.12.input_layernorm": false,
407
+ "language_model.model.layers.12.post_attention_layernorm": false,
408
+ "language_model.model.layers.13.self_attn.q_proj": true,
409
+ "language_model.model.layers.13.self_attn.k_proj": true,
410
+ "language_model.model.layers.13.self_attn.v_proj": true,
411
+ "language_model.model.layers.13.self_attn.o_proj": true,
412
+ "language_model.model.layers.13.mlp.gate_proj": true,
413
+ "language_model.model.layers.13.mlp.down_proj": true,
414
+ "language_model.model.layers.13.mlp.up_proj": true,
415
+ "language_model.model.layers.13.input_layernorm": false,
416
+ "language_model.model.layers.13.post_attention_layernorm": false,
417
+ "language_model.model.layers.14.self_attn.q_proj": true,
418
+ "language_model.model.layers.14.self_attn.k_proj": true,
419
+ "language_model.model.layers.14.self_attn.v_proj": true,
420
+ "language_model.model.layers.14.self_attn.o_proj": true,
421
+ "language_model.model.layers.14.mlp.gate_proj": true,
422
+ "language_model.model.layers.14.mlp.down_proj": true,
423
+ "language_model.model.layers.14.mlp.up_proj": true,
424
+ "language_model.model.layers.14.input_layernorm": false,
425
+ "language_model.model.layers.14.post_attention_layernorm": false,
426
+ "language_model.model.layers.15.self_attn.q_proj": true,
427
+ "language_model.model.layers.15.self_attn.k_proj": true,
428
+ "language_model.model.layers.15.self_attn.v_proj": true,
429
+ "language_model.model.layers.15.self_attn.o_proj": true,
430
+ "language_model.model.layers.15.mlp.gate_proj": true,
431
+ "language_model.model.layers.15.mlp.down_proj": true,
432
+ "language_model.model.layers.15.mlp.up_proj": true,
433
+ "language_model.model.layers.15.input_layernorm": false,
434
+ "language_model.model.layers.15.post_attention_layernorm": false,
435
+ "language_model.model.layers.16.self_attn.q_proj": true,
436
+ "language_model.model.layers.16.self_attn.k_proj": true,
437
+ "language_model.model.layers.16.self_attn.v_proj": true,
438
+ "language_model.model.layers.16.self_attn.o_proj": true,
439
+ "language_model.model.layers.16.mlp.gate_proj": true,
440
+ "language_model.model.layers.16.mlp.down_proj": true,
441
+ "language_model.model.layers.16.mlp.up_proj": true,
442
+ "language_model.model.layers.16.input_layernorm": false,
443
+ "language_model.model.layers.16.post_attention_layernorm": false,
444
+ "language_model.model.layers.17.self_attn.q_proj": true,
445
+ "language_model.model.layers.17.self_attn.k_proj": true,
446
+ "language_model.model.layers.17.self_attn.v_proj": true,
447
+ "language_model.model.layers.17.self_attn.o_proj": true,
448
+ "language_model.model.layers.17.mlp.gate_proj": true,
449
+ "language_model.model.layers.17.mlp.down_proj": true,
450
+ "language_model.model.layers.17.mlp.up_proj": true,
451
+ "language_model.model.layers.17.input_layernorm": false,
452
+ "language_model.model.layers.17.post_attention_layernorm": false,
453
+ "language_model.model.layers.18.self_attn.q_proj": true,
454
+ "language_model.model.layers.18.self_attn.k_proj": true,
455
+ "language_model.model.layers.18.self_attn.v_proj": true,
456
+ "language_model.model.layers.18.self_attn.o_proj": true,
457
+ "language_model.model.layers.18.mlp.gate_proj": true,
458
+ "language_model.model.layers.18.mlp.down_proj": true,
459
+ "language_model.model.layers.18.mlp.up_proj": true,
460
+ "language_model.model.layers.18.input_layernorm": false,
461
+ "language_model.model.layers.18.post_attention_layernorm": false,
462
+ "language_model.model.layers.19.self_attn.q_proj": true,
463
+ "language_model.model.layers.19.self_attn.k_proj": true,
464
+ "language_model.model.layers.19.self_attn.v_proj": true,
465
+ "language_model.model.layers.19.self_attn.o_proj": true,
466
+ "language_model.model.layers.19.mlp.gate_proj": true,
467
+ "language_model.model.layers.19.mlp.down_proj": true,
468
+ "language_model.model.layers.19.mlp.up_proj": true,
469
+ "language_model.model.layers.19.input_layernorm": false,
470
+ "language_model.model.layers.19.post_attention_layernorm": false,
471
+ "language_model.model.layers.20.self_attn.q_proj": true,
472
+ "language_model.model.layers.20.self_attn.k_proj": true,
473
+ "language_model.model.layers.20.self_attn.v_proj": true,
474
+ "language_model.model.layers.20.self_attn.o_proj": true,
475
+ "language_model.model.layers.20.mlp.gate_proj": true,
476
+ "language_model.model.layers.20.mlp.down_proj": true,
477
+ "language_model.model.layers.20.mlp.up_proj": true,
478
+ "language_model.model.layers.20.input_layernorm": false,
479
+ "language_model.model.layers.20.post_attention_layernorm": false,
480
+ "language_model.model.layers.21.self_attn.q_proj": true,
481
+ "language_model.model.layers.21.self_attn.k_proj": true,
482
+ "language_model.model.layers.21.self_attn.v_proj": true,
483
+ "language_model.model.layers.21.self_attn.o_proj": true,
484
+ "language_model.model.layers.21.mlp.gate_proj": true,
485
+ "language_model.model.layers.21.mlp.down_proj": true,
486
+ "language_model.model.layers.21.mlp.up_proj": true,
487
+ "language_model.model.layers.21.input_layernorm": false,
488
+ "language_model.model.layers.21.post_attention_layernorm": false,
489
+ "language_model.model.layers.22.self_attn.q_proj": true,
490
+ "language_model.model.layers.22.self_attn.k_proj": true,
491
+ "language_model.model.layers.22.self_attn.v_proj": true,
492
+ "language_model.model.layers.22.self_attn.o_proj": true,
493
+ "language_model.model.layers.22.mlp.gate_proj": true,
494
+ "language_model.model.layers.22.mlp.down_proj": true,
495
+ "language_model.model.layers.22.mlp.up_proj": true,
496
+ "language_model.model.layers.22.input_layernorm": false,
497
+ "language_model.model.layers.22.post_attention_layernorm": false,
498
+ "language_model.model.layers.23.self_attn.q_proj": true,
499
+ "language_model.model.layers.23.self_attn.k_proj": true,
500
+ "language_model.model.layers.23.self_attn.v_proj": true,
501
+ "language_model.model.layers.23.self_attn.o_proj": true,
502
+ "language_model.model.layers.23.mlp.gate_proj": true,
503
+ "language_model.model.layers.23.mlp.down_proj": true,
504
+ "language_model.model.layers.23.mlp.up_proj": true,
505
+ "language_model.model.layers.23.input_layernorm": false,
506
+ "language_model.model.layers.23.post_attention_layernorm": false,
507
+ "language_model.model.layers.24.self_attn.q_proj": true,
508
+ "language_model.model.layers.24.self_attn.k_proj": true,
509
+ "language_model.model.layers.24.self_attn.v_proj": true,
510
+ "language_model.model.layers.24.self_attn.o_proj": true,
511
+ "language_model.model.layers.24.mlp.gate_proj": true,
512
+ "language_model.model.layers.24.mlp.down_proj": true,
513
+ "language_model.model.layers.24.mlp.up_proj": true,
514
+ "language_model.model.layers.24.input_layernorm": false,
515
+ "language_model.model.layers.24.post_attention_layernorm": false,
516
+ "language_model.model.layers.25.self_attn.q_proj": true,
517
+ "language_model.model.layers.25.self_attn.k_proj": true,
518
+ "language_model.model.layers.25.self_attn.v_proj": true,
519
+ "language_model.model.layers.25.self_attn.o_proj": true,
520
+ "language_model.model.layers.25.mlp.gate_proj": true,
521
+ "language_model.model.layers.25.mlp.down_proj": true,
522
+ "language_model.model.layers.25.mlp.up_proj": true,
523
+ "language_model.model.layers.25.input_layernorm": false,
524
+ "language_model.model.layers.25.post_attention_layernorm": false,
525
+ "language_model.model.layers.26.self_attn.q_proj": true,
526
+ "language_model.model.layers.26.self_attn.k_proj": true,
527
+ "language_model.model.layers.26.self_attn.v_proj": true,
528
+ "language_model.model.layers.26.self_attn.o_proj": true,
529
+ "language_model.model.layers.26.mlp.gate_proj": true,
530
+ "language_model.model.layers.26.mlp.down_proj": true,
531
+ "language_model.model.layers.26.mlp.up_proj": true,
532
+ "language_model.model.layers.26.input_layernorm": false,
533
+ "language_model.model.layers.26.post_attention_layernorm": false,
534
+ "language_model.model.layers.27.self_attn.q_proj": true,
535
+ "language_model.model.layers.27.self_attn.k_proj": true,
536
+ "language_model.model.layers.27.self_attn.v_proj": true,
537
+ "language_model.model.layers.27.self_attn.o_proj": true,
538
+ "language_model.model.layers.27.mlp.gate_proj": true,
539
+ "language_model.model.layers.27.mlp.down_proj": true,
540
+ "language_model.model.layers.27.mlp.up_proj": true,
541
+ "language_model.model.layers.27.input_layernorm": false,
542
+ "language_model.model.layers.27.post_attention_layernorm": false,
543
+ "language_model.model.layers.28.self_attn.q_proj": true,
544
+ "language_model.model.layers.28.self_attn.k_proj": true,
545
+ "language_model.model.layers.28.self_attn.v_proj": true,
546
+ "language_model.model.layers.28.self_attn.o_proj": true,
547
+ "language_model.model.layers.28.mlp.gate_proj": true,
548
+ "language_model.model.layers.28.mlp.down_proj": true,
549
+ "language_model.model.layers.28.mlp.up_proj": true,
550
+ "language_model.model.layers.28.input_layernorm": false,
551
+ "language_model.model.layers.28.post_attention_layernorm": false,
552
+ "language_model.model.layers.29.self_attn.q_proj": true,
553
+ "language_model.model.layers.29.self_attn.k_proj": true,
554
+ "language_model.model.layers.29.self_attn.v_proj": true,
555
+ "language_model.model.layers.29.self_attn.o_proj": true,
556
+ "language_model.model.layers.29.mlp.gate_proj": true,
557
+ "language_model.model.layers.29.mlp.down_proj": true,
558
+ "language_model.model.layers.29.mlp.up_proj": true,
559
+ "language_model.model.layers.29.input_layernorm": false,
560
+ "language_model.model.layers.29.post_attention_layernorm": false,
561
+ "language_model.model.layers.30.self_attn.q_proj": true,
562
+ "language_model.model.layers.30.self_attn.k_proj": true,
563
+ "language_model.model.layers.30.self_attn.v_proj": true,
564
+ "language_model.model.layers.30.self_attn.o_proj": true,
565
+ "language_model.model.layers.30.mlp.gate_proj": true,
566
+ "language_model.model.layers.30.mlp.down_proj": true,
567
+ "language_model.model.layers.30.mlp.up_proj": true,
568
+ "language_model.model.layers.30.input_layernorm": false,
569
+ "language_model.model.layers.30.post_attention_layernorm": false,
570
+ "language_model.model.layers.31.self_attn.q_proj": true,
571
+ "language_model.model.layers.31.self_attn.k_proj": true,
572
+ "language_model.model.layers.31.self_attn.v_proj": true,
573
+ "language_model.model.layers.31.self_attn.o_proj": true,
574
+ "language_model.model.layers.31.mlp.gate_proj": true,
575
+ "language_model.model.layers.31.mlp.down_proj": true,
576
+ "language_model.model.layers.31.mlp.up_proj": true,
577
+ "language_model.model.layers.31.input_layernorm": false,
578
+ "language_model.model.layers.31.post_attention_layernorm": false,
579
+ "language_model.model.layers.32.self_attn.q_proj": true,
580
+ "language_model.model.layers.32.self_attn.k_proj": true,
581
+ "language_model.model.layers.32.self_attn.v_proj": true,
582
+ "language_model.model.layers.32.self_attn.o_proj": true,
583
+ "language_model.model.layers.32.mlp.gate_proj": true,
584
+ "language_model.model.layers.32.mlp.down_proj": true,
585
+ "language_model.model.layers.32.mlp.up_proj": true,
586
+ "language_model.model.layers.32.input_layernorm": false,
587
+ "language_model.model.layers.32.post_attention_layernorm": false,
588
+ "language_model.model.layers.33.self_attn.q_proj": true,
589
+ "language_model.model.layers.33.self_attn.k_proj": true,
590
+ "language_model.model.layers.33.self_attn.v_proj": true,
591
+ "language_model.model.layers.33.self_attn.o_proj": true,
592
+ "language_model.model.layers.33.mlp.gate_proj": true,
593
+ "language_model.model.layers.33.mlp.down_proj": true,
594
+ "language_model.model.layers.33.mlp.up_proj": true,
595
+ "language_model.model.layers.33.input_layernorm": false,
596
+ "language_model.model.layers.33.post_attention_layernorm": false,
597
+ "language_model.model.layers.34.self_attn.q_proj": true,
598
+ "language_model.model.layers.34.self_attn.k_proj": true,
599
+ "language_model.model.layers.34.self_attn.v_proj": true,
600
+ "language_model.model.layers.34.self_attn.o_proj": true,
601
+ "language_model.model.layers.34.mlp.gate_proj": true,
602
+ "language_model.model.layers.34.mlp.down_proj": true,
603
+ "language_model.model.layers.34.mlp.up_proj": true,
604
+ "language_model.model.layers.34.input_layernorm": false,
605
+ "language_model.model.layers.34.post_attention_layernorm": false,
606
+ "language_model.model.layers.35.self_attn.q_proj": true,
607
+ "language_model.model.layers.35.self_attn.k_proj": true,
608
+ "language_model.model.layers.35.self_attn.v_proj": true,
609
+ "language_model.model.layers.35.self_attn.o_proj": true,
610
+ "language_model.model.layers.35.mlp.gate_proj": true,
611
+ "language_model.model.layers.35.mlp.down_proj": true,
612
+ "language_model.model.layers.35.mlp.up_proj": true,
613
+ "language_model.model.layers.35.input_layernorm": false,
614
+ "language_model.model.layers.35.post_attention_layernorm": false,
615
+ "language_model.model.norm": false
616
+ },
617
+ "quantization_config": {
618
+ "group_size": 64,
619
+ "bits": 8,
620
+ "vision_tower.patch_embed.proj": false,
621
+ "vision_tower.rotary_pos_emb": false,
622
+ "vision_tower.blocks.0.norm1": false,
623
+ "vision_tower.blocks.0.norm2": false,
624
+ "vision_tower.blocks.0.attn.qkv": false,
625
+ "vision_tower.blocks.0.attn.proj": false,
626
+ "vision_tower.blocks.0.mlp.gate_proj": false,
627
+ "vision_tower.blocks.0.mlp.up_proj": false,
628
+ "vision_tower.blocks.0.mlp.down_proj": false,
629
+ "vision_tower.blocks.1.norm1": false,
630
+ "vision_tower.blocks.1.norm2": false,
631
+ "vision_tower.blocks.1.attn.qkv": false,
632
+ "vision_tower.blocks.1.attn.proj": false,
633
+ "vision_tower.blocks.1.mlp.gate_proj": false,
634
+ "vision_tower.blocks.1.mlp.up_proj": false,
635
+ "vision_tower.blocks.1.mlp.down_proj": false,
636
+ "vision_tower.blocks.2.norm1": false,
637
+ "vision_tower.blocks.2.norm2": false,
638
+ "vision_tower.blocks.2.attn.qkv": false,
639
+ "vision_tower.blocks.2.attn.proj": false,
640
+ "vision_tower.blocks.2.mlp.gate_proj": false,
641
+ "vision_tower.blocks.2.mlp.up_proj": false,
642
+ "vision_tower.blocks.2.mlp.down_proj": false,
643
+ "vision_tower.blocks.3.norm1": false,
644
+ "vision_tower.blocks.3.norm2": false,
645
+ "vision_tower.blocks.3.attn.qkv": false,
646
+ "vision_tower.blocks.3.attn.proj": false,
647
+ "vision_tower.blocks.3.mlp.gate_proj": false,
648
+ "vision_tower.blocks.3.mlp.up_proj": false,
649
+ "vision_tower.blocks.3.mlp.down_proj": false,
650
+ "vision_tower.blocks.4.norm1": false,
651
+ "vision_tower.blocks.4.norm2": false,
652
+ "vision_tower.blocks.4.attn.qkv": false,
653
+ "vision_tower.blocks.4.attn.proj": false,
654
+ "vision_tower.blocks.4.mlp.gate_proj": false,
655
+ "vision_tower.blocks.4.mlp.up_proj": false,
656
+ "vision_tower.blocks.4.mlp.down_proj": false,
657
+ "vision_tower.blocks.5.norm1": false,
658
+ "vision_tower.blocks.5.norm2": false,
659
+ "vision_tower.blocks.5.attn.qkv": false,
660
+ "vision_tower.blocks.5.attn.proj": false,
661
+ "vision_tower.blocks.5.mlp.gate_proj": false,
662
+ "vision_tower.blocks.5.mlp.up_proj": false,
663
+ "vision_tower.blocks.5.mlp.down_proj": false,
664
+ "vision_tower.blocks.6.norm1": false,
665
+ "vision_tower.blocks.6.norm2": false,
666
+ "vision_tower.blocks.6.attn.qkv": false,
667
+ "vision_tower.blocks.6.attn.proj": false,
668
+ "vision_tower.blocks.6.mlp.gate_proj": false,
669
+ "vision_tower.blocks.6.mlp.up_proj": false,
670
+ "vision_tower.blocks.6.mlp.down_proj": false,
671
+ "vision_tower.blocks.7.norm1": false,
672
+ "vision_tower.blocks.7.norm2": false,
673
+ "vision_tower.blocks.7.attn.qkv": false,
674
+ "vision_tower.blocks.7.attn.proj": false,
675
+ "vision_tower.blocks.7.mlp.gate_proj": false,
676
+ "vision_tower.blocks.7.mlp.up_proj": false,
677
+ "vision_tower.blocks.7.mlp.down_proj": false,
678
+ "vision_tower.blocks.8.norm1": false,
679
+ "vision_tower.blocks.8.norm2": false,
680
+ "vision_tower.blocks.8.attn.qkv": false,
681
+ "vision_tower.blocks.8.attn.proj": false,
682
+ "vision_tower.blocks.8.mlp.gate_proj": false,
683
+ "vision_tower.blocks.8.mlp.up_proj": false,
684
+ "vision_tower.blocks.8.mlp.down_proj": false,
685
+ "vision_tower.blocks.9.norm1": false,
686
+ "vision_tower.blocks.9.norm2": false,
687
+ "vision_tower.blocks.9.attn.qkv": false,
688
+ "vision_tower.blocks.9.attn.proj": false,
689
+ "vision_tower.blocks.9.mlp.gate_proj": false,
690
+ "vision_tower.blocks.9.mlp.up_proj": false,
691
+ "vision_tower.blocks.9.mlp.down_proj": false,
692
+ "vision_tower.blocks.10.norm1": false,
693
+ "vision_tower.blocks.10.norm2": false,
694
+ "vision_tower.blocks.10.attn.qkv": false,
695
+ "vision_tower.blocks.10.attn.proj": false,
696
+ "vision_tower.blocks.10.mlp.gate_proj": false,
697
+ "vision_tower.blocks.10.mlp.up_proj": false,
698
+ "vision_tower.blocks.10.mlp.down_proj": false,
699
+ "vision_tower.blocks.11.norm1": false,
700
+ "vision_tower.blocks.11.norm2": false,
701
+ "vision_tower.blocks.11.attn.qkv": false,
702
+ "vision_tower.blocks.11.attn.proj": false,
703
+ "vision_tower.blocks.11.mlp.gate_proj": false,
704
+ "vision_tower.blocks.11.mlp.up_proj": false,
705
+ "vision_tower.blocks.11.mlp.down_proj": false,
706
+ "vision_tower.blocks.12.norm1": false,
707
+ "vision_tower.blocks.12.norm2": false,
708
+ "vision_tower.blocks.12.attn.qkv": false,
709
+ "vision_tower.blocks.12.attn.proj": false,
710
+ "vision_tower.blocks.12.mlp.gate_proj": false,
711
+ "vision_tower.blocks.12.mlp.up_proj": false,
712
+ "vision_tower.blocks.12.mlp.down_proj": false,
713
+ "vision_tower.blocks.13.norm1": false,
714
+ "vision_tower.blocks.13.norm2": false,
715
+ "vision_tower.blocks.13.attn.qkv": false,
716
+ "vision_tower.blocks.13.attn.proj": false,
717
+ "vision_tower.blocks.13.mlp.gate_proj": false,
718
+ "vision_tower.blocks.13.mlp.up_proj": false,
719
+ "vision_tower.blocks.13.mlp.down_proj": false,
720
+ "vision_tower.blocks.14.norm1": false,
721
+ "vision_tower.blocks.14.norm2": false,
722
+ "vision_tower.blocks.14.attn.qkv": false,
723
+ "vision_tower.blocks.14.attn.proj": false,
724
+ "vision_tower.blocks.14.mlp.gate_proj": false,
725
+ "vision_tower.blocks.14.mlp.up_proj": false,
726
+ "vision_tower.blocks.14.mlp.down_proj": false,
727
+ "vision_tower.blocks.15.norm1": false,
728
+ "vision_tower.blocks.15.norm2": false,
729
+ "vision_tower.blocks.15.attn.qkv": false,
730
+ "vision_tower.blocks.15.attn.proj": false,
731
+ "vision_tower.blocks.15.mlp.gate_proj": false,
732
+ "vision_tower.blocks.15.mlp.up_proj": false,
733
+ "vision_tower.blocks.15.mlp.down_proj": false,
734
+ "vision_tower.blocks.16.norm1": false,
735
+ "vision_tower.blocks.16.norm2": false,
736
+ "vision_tower.blocks.16.attn.qkv": false,
737
+ "vision_tower.blocks.16.attn.proj": false,
738
+ "vision_tower.blocks.16.mlp.gate_proj": false,
739
+ "vision_tower.blocks.16.mlp.up_proj": false,
740
+ "vision_tower.blocks.16.mlp.down_proj": false,
741
+ "vision_tower.blocks.17.norm1": false,
742
+ "vision_tower.blocks.17.norm2": false,
743
+ "vision_tower.blocks.17.attn.qkv": false,
744
+ "vision_tower.blocks.17.attn.proj": false,
745
+ "vision_tower.blocks.17.mlp.gate_proj": false,
746
+ "vision_tower.blocks.17.mlp.up_proj": false,
747
+ "vision_tower.blocks.17.mlp.down_proj": false,
748
+ "vision_tower.blocks.18.norm1": false,
749
+ "vision_tower.blocks.18.norm2": false,
750
+ "vision_tower.blocks.18.attn.qkv": false,
751
+ "vision_tower.blocks.18.attn.proj": false,
752
+ "vision_tower.blocks.18.mlp.gate_proj": false,
753
+ "vision_tower.blocks.18.mlp.up_proj": false,
754
+ "vision_tower.blocks.18.mlp.down_proj": false,
755
+ "vision_tower.blocks.19.norm1": false,
756
+ "vision_tower.blocks.19.norm2": false,
757
+ "vision_tower.blocks.19.attn.qkv": false,
758
+ "vision_tower.blocks.19.attn.proj": false,
759
+ "vision_tower.blocks.19.mlp.gate_proj": false,
760
+ "vision_tower.blocks.19.mlp.up_proj": false,
761
+ "vision_tower.blocks.19.mlp.down_proj": false,
762
+ "vision_tower.blocks.20.norm1": false,
763
+ "vision_tower.blocks.20.norm2": false,
764
+ "vision_tower.blocks.20.attn.qkv": false,
765
+ "vision_tower.blocks.20.attn.proj": false,
766
+ "vision_tower.blocks.20.mlp.gate_proj": false,
767
+ "vision_tower.blocks.20.mlp.up_proj": false,
768
+ "vision_tower.blocks.20.mlp.down_proj": false,
769
+ "vision_tower.blocks.21.norm1": false,
770
+ "vision_tower.blocks.21.norm2": false,
771
+ "vision_tower.blocks.21.attn.qkv": false,
772
+ "vision_tower.blocks.21.attn.proj": false,
773
+ "vision_tower.blocks.21.mlp.gate_proj": false,
774
+ "vision_tower.blocks.21.mlp.up_proj": false,
775
+ "vision_tower.blocks.21.mlp.down_proj": false,
776
+ "vision_tower.blocks.22.norm1": false,
777
+ "vision_tower.blocks.22.norm2": false,
778
+ "vision_tower.blocks.22.attn.qkv": false,
779
+ "vision_tower.blocks.22.attn.proj": false,
780
+ "vision_tower.blocks.22.mlp.gate_proj": false,
781
+ "vision_tower.blocks.22.mlp.up_proj": false,
782
+ "vision_tower.blocks.22.mlp.down_proj": false,
783
+ "vision_tower.blocks.23.norm1": false,
784
+ "vision_tower.blocks.23.norm2": false,
785
+ "vision_tower.blocks.23.attn.qkv": false,
786
+ "vision_tower.blocks.23.attn.proj": false,
787
+ "vision_tower.blocks.23.mlp.gate_proj": false,
788
+ "vision_tower.blocks.23.mlp.up_proj": false,
789
+ "vision_tower.blocks.23.mlp.down_proj": false,
790
+ "vision_tower.blocks.24.norm1": false,
791
+ "vision_tower.blocks.24.norm2": false,
792
+ "vision_tower.blocks.24.attn.qkv": false,
793
+ "vision_tower.blocks.24.attn.proj": false,
794
+ "vision_tower.blocks.24.mlp.gate_proj": false,
795
+ "vision_tower.blocks.24.mlp.up_proj": false,
796
+ "vision_tower.blocks.24.mlp.down_proj": false,
797
+ "vision_tower.blocks.25.norm1": false,
798
+ "vision_tower.blocks.25.norm2": false,
799
+ "vision_tower.blocks.25.attn.qkv": false,
800
+ "vision_tower.blocks.25.attn.proj": false,
801
+ "vision_tower.blocks.25.mlp.gate_proj": false,
802
+ "vision_tower.blocks.25.mlp.up_proj": false,
803
+ "vision_tower.blocks.25.mlp.down_proj": false,
804
+ "vision_tower.blocks.26.norm1": false,
805
+ "vision_tower.blocks.26.norm2": false,
806
+ "vision_tower.blocks.26.attn.qkv": false,
807
+ "vision_tower.blocks.26.attn.proj": false,
808
+ "vision_tower.blocks.26.mlp.gate_proj": false,
809
+ "vision_tower.blocks.26.mlp.up_proj": false,
810
+ "vision_tower.blocks.26.mlp.down_proj": false,
811
+ "vision_tower.blocks.27.norm1": false,
812
+ "vision_tower.blocks.27.norm2": false,
813
+ "vision_tower.blocks.27.attn.qkv": false,
814
+ "vision_tower.blocks.27.attn.proj": false,
815
+ "vision_tower.blocks.27.mlp.gate_proj": false,
816
+ "vision_tower.blocks.27.mlp.up_proj": false,
817
+ "vision_tower.blocks.27.mlp.down_proj": false,
818
+ "vision_tower.blocks.28.norm1": false,
819
+ "vision_tower.blocks.28.norm2": false,
820
+ "vision_tower.blocks.28.attn.qkv": false,
821
+ "vision_tower.blocks.28.attn.proj": false,
822
+ "vision_tower.blocks.28.mlp.gate_proj": false,
823
+ "vision_tower.blocks.28.mlp.up_proj": false,
824
+ "vision_tower.blocks.28.mlp.down_proj": false,
825
+ "vision_tower.blocks.29.norm1": false,
826
+ "vision_tower.blocks.29.norm2": false,
827
+ "vision_tower.blocks.29.attn.qkv": false,
828
+ "vision_tower.blocks.29.attn.proj": false,
829
+ "vision_tower.blocks.29.mlp.gate_proj": false,
830
+ "vision_tower.blocks.29.mlp.up_proj": false,
831
+ "vision_tower.blocks.29.mlp.down_proj": false,
832
+ "vision_tower.blocks.30.norm1": false,
833
+ "vision_tower.blocks.30.norm2": false,
834
+ "vision_tower.blocks.30.attn.qkv": false,
835
+ "vision_tower.blocks.30.attn.proj": false,
836
+ "vision_tower.blocks.30.mlp.gate_proj": false,
837
+ "vision_tower.blocks.30.mlp.up_proj": false,
838
+ "vision_tower.blocks.30.mlp.down_proj": false,
839
+ "vision_tower.blocks.31.norm1": false,
840
+ "vision_tower.blocks.31.norm2": false,
841
+ "vision_tower.blocks.31.attn.qkv": false,
842
+ "vision_tower.blocks.31.attn.proj": false,
843
+ "vision_tower.blocks.31.mlp.gate_proj": false,
844
+ "vision_tower.blocks.31.mlp.up_proj": false,
845
+ "vision_tower.blocks.31.mlp.down_proj": false,
846
+ "vision_tower.merger.ln_q": false,
847
+ "vision_tower.merger.mlp.0": false,
848
+ "vision_tower.merger.mlp.1": false,
849
+ "vision_tower.merger.mlp.2": false,
850
+ "language_model.model.embed_tokens": true,
851
+ "language_model.model.layers.0.self_attn.q_proj": true,
852
+ "language_model.model.layers.0.self_attn.k_proj": true,
853
+ "language_model.model.layers.0.self_attn.v_proj": true,
854
+ "language_model.model.layers.0.self_attn.o_proj": true,
855
+ "language_model.model.layers.0.mlp.gate_proj": true,
856
+ "language_model.model.layers.0.mlp.down_proj": true,
857
+ "language_model.model.layers.0.mlp.up_proj": true,
858
+ "language_model.model.layers.0.input_layernorm": false,
859
+ "language_model.model.layers.0.post_attention_layernorm": false,
860
+ "language_model.model.layers.1.self_attn.q_proj": true,
861
+ "language_model.model.layers.1.self_attn.k_proj": true,
862
+ "language_model.model.layers.1.self_attn.v_proj": true,
863
+ "language_model.model.layers.1.self_attn.o_proj": true,
864
+ "language_model.model.layers.1.mlp.gate_proj": true,
865
+ "language_model.model.layers.1.mlp.down_proj": true,
866
+ "language_model.model.layers.1.mlp.up_proj": true,
867
+ "language_model.model.layers.1.input_layernorm": false,
868
+ "language_model.model.layers.1.post_attention_layernorm": false,
869
+ "language_model.model.layers.2.self_attn.q_proj": true,
870
+ "language_model.model.layers.2.self_attn.k_proj": true,
871
+ "language_model.model.layers.2.self_attn.v_proj": true,
872
+ "language_model.model.layers.2.self_attn.o_proj": true,
873
+ "language_model.model.layers.2.mlp.gate_proj": true,
874
+ "language_model.model.layers.2.mlp.down_proj": true,
875
+ "language_model.model.layers.2.mlp.up_proj": true,
876
+ "language_model.model.layers.2.input_layernorm": false,
877
+ "language_model.model.layers.2.post_attention_layernorm": false,
878
+ "language_model.model.layers.3.self_attn.q_proj": true,
879
+ "language_model.model.layers.3.self_attn.k_proj": true,
880
+ "language_model.model.layers.3.self_attn.v_proj": true,
881
+ "language_model.model.layers.3.self_attn.o_proj": true,
882
+ "language_model.model.layers.3.mlp.gate_proj": true,
883
+ "language_model.model.layers.3.mlp.down_proj": true,
884
+ "language_model.model.layers.3.mlp.up_proj": true,
885
+ "language_model.model.layers.3.input_layernorm": false,
886
+ "language_model.model.layers.3.post_attention_layernorm": false,
887
+ "language_model.model.layers.4.self_attn.q_proj": true,
888
+ "language_model.model.layers.4.self_attn.k_proj": true,
889
+ "language_model.model.layers.4.self_attn.v_proj": true,
890
+ "language_model.model.layers.4.self_attn.o_proj": true,
891
+ "language_model.model.layers.4.mlp.gate_proj": true,
892
+ "language_model.model.layers.4.mlp.down_proj": true,
893
+ "language_model.model.layers.4.mlp.up_proj": true,
894
+ "language_model.model.layers.4.input_layernorm": false,
895
+ "language_model.model.layers.4.post_attention_layernorm": false,
896
+ "language_model.model.layers.5.self_attn.q_proj": true,
897
+ "language_model.model.layers.5.self_attn.k_proj": true,
898
+ "language_model.model.layers.5.self_attn.v_proj": true,
899
+ "language_model.model.layers.5.self_attn.o_proj": true,
900
+ "language_model.model.layers.5.mlp.gate_proj": true,
901
+ "language_model.model.layers.5.mlp.down_proj": true,
902
+ "language_model.model.layers.5.mlp.up_proj": true,
903
+ "language_model.model.layers.5.input_layernorm": false,
904
+ "language_model.model.layers.5.post_attention_layernorm": false,
905
+ "language_model.model.layers.6.self_attn.q_proj": true,
906
+ "language_model.model.layers.6.self_attn.k_proj": true,
907
+ "language_model.model.layers.6.self_attn.v_proj": true,
908
+ "language_model.model.layers.6.self_attn.o_proj": true,
909
+ "language_model.model.layers.6.mlp.gate_proj": true,
910
+ "language_model.model.layers.6.mlp.down_proj": true,
911
+ "language_model.model.layers.6.mlp.up_proj": true,
912
+ "language_model.model.layers.6.input_layernorm": false,
913
+ "language_model.model.layers.6.post_attention_layernorm": false,
914
+ "language_model.model.layers.7.self_attn.q_proj": true,
915
+ "language_model.model.layers.7.self_attn.k_proj": true,
916
+ "language_model.model.layers.7.self_attn.v_proj": true,
917
+ "language_model.model.layers.7.self_attn.o_proj": true,
918
+ "language_model.model.layers.7.mlp.gate_proj": true,
919
+ "language_model.model.layers.7.mlp.down_proj": true,
920
+ "language_model.model.layers.7.mlp.up_proj": true,
921
+ "language_model.model.layers.7.input_layernorm": false,
922
+ "language_model.model.layers.7.post_attention_layernorm": false,
923
+ "language_model.model.layers.8.self_attn.q_proj": true,
924
+ "language_model.model.layers.8.self_attn.k_proj": true,
925
+ "language_model.model.layers.8.self_attn.v_proj": true,
926
+ "language_model.model.layers.8.self_attn.o_proj": true,
927
+ "language_model.model.layers.8.mlp.gate_proj": true,
928
+ "language_model.model.layers.8.mlp.down_proj": true,
929
+ "language_model.model.layers.8.mlp.up_proj": true,
930
+ "language_model.model.layers.8.input_layernorm": false,
931
+ "language_model.model.layers.8.post_attention_layernorm": false,
932
+ "language_model.model.layers.9.self_attn.q_proj": true,
933
+ "language_model.model.layers.9.self_attn.k_proj": true,
934
+ "language_model.model.layers.9.self_attn.v_proj": true,
935
+ "language_model.model.layers.9.self_attn.o_proj": true,
936
+ "language_model.model.layers.9.mlp.gate_proj": true,
937
+ "language_model.model.layers.9.mlp.down_proj": true,
938
+ "language_model.model.layers.9.mlp.up_proj": true,
939
+ "language_model.model.layers.9.input_layernorm": false,
940
+ "language_model.model.layers.9.post_attention_layernorm": false,
941
+ "language_model.model.layers.10.self_attn.q_proj": true,
942
+ "language_model.model.layers.10.self_attn.k_proj": true,
943
+ "language_model.model.layers.10.self_attn.v_proj": true,
944
+ "language_model.model.layers.10.self_attn.o_proj": true,
945
+ "language_model.model.layers.10.mlp.gate_proj": true,
946
+ "language_model.model.layers.10.mlp.down_proj": true,
947
+ "language_model.model.layers.10.mlp.up_proj": true,
948
+ "language_model.model.layers.10.input_layernorm": false,
949
+ "language_model.model.layers.10.post_attention_layernorm": false,
950
+ "language_model.model.layers.11.self_attn.q_proj": true,
951
+ "language_model.model.layers.11.self_attn.k_proj": true,
952
+ "language_model.model.layers.11.self_attn.v_proj": true,
953
+ "language_model.model.layers.11.self_attn.o_proj": true,
954
+ "language_model.model.layers.11.mlp.gate_proj": true,
955
+ "language_model.model.layers.11.mlp.down_proj": true,
956
+ "language_model.model.layers.11.mlp.up_proj": true,
957
+ "language_model.model.layers.11.input_layernorm": false,
958
+ "language_model.model.layers.11.post_attention_layernorm": false,
959
+ "language_model.model.layers.12.self_attn.q_proj": true,
960
+ "language_model.model.layers.12.self_attn.k_proj": true,
961
+ "language_model.model.layers.12.self_attn.v_proj": true,
962
+ "language_model.model.layers.12.self_attn.o_proj": true,
963
+ "language_model.model.layers.12.mlp.gate_proj": true,
964
+ "language_model.model.layers.12.mlp.down_proj": true,
965
+ "language_model.model.layers.12.mlp.up_proj": true,
966
+ "language_model.model.layers.12.input_layernorm": false,
967
+ "language_model.model.layers.12.post_attention_layernorm": false,
968
+ "language_model.model.layers.13.self_attn.q_proj": true,
969
+ "language_model.model.layers.13.self_attn.k_proj": true,
970
+ "language_model.model.layers.13.self_attn.v_proj": true,
971
+ "language_model.model.layers.13.self_attn.o_proj": true,
972
+ "language_model.model.layers.13.mlp.gate_proj": true,
973
+ "language_model.model.layers.13.mlp.down_proj": true,
974
+ "language_model.model.layers.13.mlp.up_proj": true,
975
+ "language_model.model.layers.13.input_layernorm": false,
976
+ "language_model.model.layers.13.post_attention_layernorm": false,
977
+ "language_model.model.layers.14.self_attn.q_proj": true,
978
+ "language_model.model.layers.14.self_attn.k_proj": true,
979
+ "language_model.model.layers.14.self_attn.v_proj": true,
980
+ "language_model.model.layers.14.self_attn.o_proj": true,
981
+ "language_model.model.layers.14.mlp.gate_proj": true,
982
+ "language_model.model.layers.14.mlp.down_proj": true,
983
+ "language_model.model.layers.14.mlp.up_proj": true,
984
+ "language_model.model.layers.14.input_layernorm": false,
985
+ "language_model.model.layers.14.post_attention_layernorm": false,
986
+ "language_model.model.layers.15.self_attn.q_proj": true,
987
+ "language_model.model.layers.15.self_attn.k_proj": true,
988
+ "language_model.model.layers.15.self_attn.v_proj": true,
989
+ "language_model.model.layers.15.self_attn.o_proj": true,
990
+ "language_model.model.layers.15.mlp.gate_proj": true,
991
+ "language_model.model.layers.15.mlp.down_proj": true,
992
+ "language_model.model.layers.15.mlp.up_proj": true,
993
+ "language_model.model.layers.15.input_layernorm": false,
994
+ "language_model.model.layers.15.post_attention_layernorm": false,
995
+ "language_model.model.layers.16.self_attn.q_proj": true,
996
+ "language_model.model.layers.16.self_attn.k_proj": true,
997
+ "language_model.model.layers.16.self_attn.v_proj": true,
998
+ "language_model.model.layers.16.self_attn.o_proj": true,
999
+ "language_model.model.layers.16.mlp.gate_proj": true,
1000
+ "language_model.model.layers.16.mlp.down_proj": true,
1001
+ "language_model.model.layers.16.mlp.up_proj": true,
1002
+ "language_model.model.layers.16.input_layernorm": false,
1003
+ "language_model.model.layers.16.post_attention_layernorm": false,
1004
+ "language_model.model.layers.17.self_attn.q_proj": true,
1005
+ "language_model.model.layers.17.self_attn.k_proj": true,
1006
+ "language_model.model.layers.17.self_attn.v_proj": true,
1007
+ "language_model.model.layers.17.self_attn.o_proj": true,
1008
+ "language_model.model.layers.17.mlp.gate_proj": true,
1009
+ "language_model.model.layers.17.mlp.down_proj": true,
1010
+ "language_model.model.layers.17.mlp.up_proj": true,
1011
+ "language_model.model.layers.17.input_layernorm": false,
1012
+ "language_model.model.layers.17.post_attention_layernorm": false,
1013
+ "language_model.model.layers.18.self_attn.q_proj": true,
1014
+ "language_model.model.layers.18.self_attn.k_proj": true,
1015
+ "language_model.model.layers.18.self_attn.v_proj": true,
1016
+ "language_model.model.layers.18.self_attn.o_proj": true,
1017
+ "language_model.model.layers.18.mlp.gate_proj": true,
1018
+ "language_model.model.layers.18.mlp.down_proj": true,
1019
+ "language_model.model.layers.18.mlp.up_proj": true,
1020
+ "language_model.model.layers.18.input_layernorm": false,
1021
+ "language_model.model.layers.18.post_attention_layernorm": false,
1022
+ "language_model.model.layers.19.self_attn.q_proj": true,
1023
+ "language_model.model.layers.19.self_attn.k_proj": true,
1024
+ "language_model.model.layers.19.self_attn.v_proj": true,
1025
+ "language_model.model.layers.19.self_attn.o_proj": true,
1026
+ "language_model.model.layers.19.mlp.gate_proj": true,
1027
+ "language_model.model.layers.19.mlp.down_proj": true,
1028
+ "language_model.model.layers.19.mlp.up_proj": true,
1029
+ "language_model.model.layers.19.input_layernorm": false,
1030
+ "language_model.model.layers.19.post_attention_layernorm": false,
1031
+ "language_model.model.layers.20.self_attn.q_proj": true,
1032
+ "language_model.model.layers.20.self_attn.k_proj": true,
1033
+ "language_model.model.layers.20.self_attn.v_proj": true,
1034
+ "language_model.model.layers.20.self_attn.o_proj": true,
1035
+ "language_model.model.layers.20.mlp.gate_proj": true,
1036
+ "language_model.model.layers.20.mlp.down_proj": true,
1037
+ "language_model.model.layers.20.mlp.up_proj": true,
1038
+ "language_model.model.layers.20.input_layernorm": false,
1039
+ "language_model.model.layers.20.post_attention_layernorm": false,
1040
+ "language_model.model.layers.21.self_attn.q_proj": true,
1041
+ "language_model.model.layers.21.self_attn.k_proj": true,
1042
+ "language_model.model.layers.21.self_attn.v_proj": true,
1043
+ "language_model.model.layers.21.self_attn.o_proj": true,
1044
+ "language_model.model.layers.21.mlp.gate_proj": true,
1045
+ "language_model.model.layers.21.mlp.down_proj": true,
1046
+ "language_model.model.layers.21.mlp.up_proj": true,
1047
+ "language_model.model.layers.21.input_layernorm": false,
1048
+ "language_model.model.layers.21.post_attention_layernorm": false,
1049
+ "language_model.model.layers.22.self_attn.q_proj": true,
1050
+ "language_model.model.layers.22.self_attn.k_proj": true,
1051
+ "language_model.model.layers.22.self_attn.v_proj": true,
1052
+ "language_model.model.layers.22.self_attn.o_proj": true,
1053
+ "language_model.model.layers.22.mlp.gate_proj": true,
1054
+ "language_model.model.layers.22.mlp.down_proj": true,
1055
+ "language_model.model.layers.22.mlp.up_proj": true,
1056
+ "language_model.model.layers.22.input_layernorm": false,
1057
+ "language_model.model.layers.22.post_attention_layernorm": false,
1058
+ "language_model.model.layers.23.self_attn.q_proj": true,
1059
+ "language_model.model.layers.23.self_attn.k_proj": true,
1060
+ "language_model.model.layers.23.self_attn.v_proj": true,
1061
+ "language_model.model.layers.23.self_attn.o_proj": true,
1062
+ "language_model.model.layers.23.mlp.gate_proj": true,
1063
+ "language_model.model.layers.23.mlp.down_proj": true,
1064
+ "language_model.model.layers.23.mlp.up_proj": true,
1065
+ "language_model.model.layers.23.input_layernorm": false,
1066
+ "language_model.model.layers.23.post_attention_layernorm": false,
1067
+ "language_model.model.layers.24.self_attn.q_proj": true,
1068
+ "language_model.model.layers.24.self_attn.k_proj": true,
1069
+ "language_model.model.layers.24.self_attn.v_proj": true,
1070
+ "language_model.model.layers.24.self_attn.o_proj": true,
1071
+ "language_model.model.layers.24.mlp.gate_proj": true,
1072
+ "language_model.model.layers.24.mlp.down_proj": true,
1073
+ "language_model.model.layers.24.mlp.up_proj": true,
1074
+ "language_model.model.layers.24.input_layernorm": false,
1075
+ "language_model.model.layers.24.post_attention_layernorm": false,
1076
+ "language_model.model.layers.25.self_attn.q_proj": true,
1077
+ "language_model.model.layers.25.self_attn.k_proj": true,
1078
+ "language_model.model.layers.25.self_attn.v_proj": true,
1079
+ "language_model.model.layers.25.self_attn.o_proj": true,
1080
+ "language_model.model.layers.25.mlp.gate_proj": true,
1081
+ "language_model.model.layers.25.mlp.down_proj": true,
1082
+ "language_model.model.layers.25.mlp.up_proj": true,
1083
+ "language_model.model.layers.25.input_layernorm": false,
1084
+ "language_model.model.layers.25.post_attention_layernorm": false,
1085
+ "language_model.model.layers.26.self_attn.q_proj": true,
1086
+ "language_model.model.layers.26.self_attn.k_proj": true,
1087
+ "language_model.model.layers.26.self_attn.v_proj": true,
1088
+ "language_model.model.layers.26.self_attn.o_proj": true,
1089
+ "language_model.model.layers.26.mlp.gate_proj": true,
1090
+ "language_model.model.layers.26.mlp.down_proj": true,
1091
+ "language_model.model.layers.26.mlp.up_proj": true,
1092
+ "language_model.model.layers.26.input_layernorm": false,
1093
+ "language_model.model.layers.26.post_attention_layernorm": false,
1094
+ "language_model.model.layers.27.self_attn.q_proj": true,
1095
+ "language_model.model.layers.27.self_attn.k_proj": true,
1096
+ "language_model.model.layers.27.self_attn.v_proj": true,
1097
+ "language_model.model.layers.27.self_attn.o_proj": true,
1098
+ "language_model.model.layers.27.mlp.gate_proj": true,
1099
+ "language_model.model.layers.27.mlp.down_proj": true,
1100
+ "language_model.model.layers.27.mlp.up_proj": true,
1101
+ "language_model.model.layers.27.input_layernorm": false,
1102
+ "language_model.model.layers.27.post_attention_layernorm": false,
1103
+ "language_model.model.layers.28.self_attn.q_proj": true,
1104
+ "language_model.model.layers.28.self_attn.k_proj": true,
1105
+ "language_model.model.layers.28.self_attn.v_proj": true,
1106
+ "language_model.model.layers.28.self_attn.o_proj": true,
1107
+ "language_model.model.layers.28.mlp.gate_proj": true,
1108
+ "language_model.model.layers.28.mlp.down_proj": true,
1109
+ "language_model.model.layers.28.mlp.up_proj": true,
1110
+ "language_model.model.layers.28.input_layernorm": false,
1111
+ "language_model.model.layers.28.post_attention_layernorm": false,
1112
+ "language_model.model.layers.29.self_attn.q_proj": true,
1113
+ "language_model.model.layers.29.self_attn.k_proj": true,
1114
+ "language_model.model.layers.29.self_attn.v_proj": true,
1115
+ "language_model.model.layers.29.self_attn.o_proj": true,
1116
+ "language_model.model.layers.29.mlp.gate_proj": true,
1117
+ "language_model.model.layers.29.mlp.down_proj": true,
1118
+ "language_model.model.layers.29.mlp.up_proj": true,
1119
+ "language_model.model.layers.29.input_layernorm": false,
1120
+ "language_model.model.layers.29.post_attention_layernorm": false,
1121
+ "language_model.model.layers.30.self_attn.q_proj": true,
1122
+ "language_model.model.layers.30.self_attn.k_proj": true,
1123
+ "language_model.model.layers.30.self_attn.v_proj": true,
1124
+ "language_model.model.layers.30.self_attn.o_proj": true,
1125
+ "language_model.model.layers.30.mlp.gate_proj": true,
1126
+ "language_model.model.layers.30.mlp.down_proj": true,
1127
+ "language_model.model.layers.30.mlp.up_proj": true,
1128
+ "language_model.model.layers.30.input_layernorm": false,
1129
+ "language_model.model.layers.30.post_attention_layernorm": false,
1130
+ "language_model.model.layers.31.self_attn.q_proj": true,
1131
+ "language_model.model.layers.31.self_attn.k_proj": true,
1132
+ "language_model.model.layers.31.self_attn.v_proj": true,
1133
+ "language_model.model.layers.31.self_attn.o_proj": true,
1134
+ "language_model.model.layers.31.mlp.gate_proj": true,
1135
+ "language_model.model.layers.31.mlp.down_proj": true,
1136
+ "language_model.model.layers.31.mlp.up_proj": true,
1137
+ "language_model.model.layers.31.input_layernorm": false,
1138
+ "language_model.model.layers.31.post_attention_layernorm": false,
1139
+ "language_model.model.layers.32.self_attn.q_proj": true,
1140
+ "language_model.model.layers.32.self_attn.k_proj": true,
1141
+ "language_model.model.layers.32.self_attn.v_proj": true,
1142
+ "language_model.model.layers.32.self_attn.o_proj": true,
1143
+ "language_model.model.layers.32.mlp.gate_proj": true,
1144
+ "language_model.model.layers.32.mlp.down_proj": true,
1145
+ "language_model.model.layers.32.mlp.up_proj": true,
1146
+ "language_model.model.layers.32.input_layernorm": false,
1147
+ "language_model.model.layers.32.post_attention_layernorm": false,
1148
+ "language_model.model.layers.33.self_attn.q_proj": true,
1149
+ "language_model.model.layers.33.self_attn.k_proj": true,
1150
+ "language_model.model.layers.33.self_attn.v_proj": true,
1151
+ "language_model.model.layers.33.self_attn.o_proj": true,
1152
+ "language_model.model.layers.33.mlp.gate_proj": true,
1153
+ "language_model.model.layers.33.mlp.down_proj": true,
1154
+ "language_model.model.layers.33.mlp.up_proj": true,
1155
+ "language_model.model.layers.33.input_layernorm": false,
1156
+ "language_model.model.layers.33.post_attention_layernorm": false,
1157
+ "language_model.model.layers.34.self_attn.q_proj": true,
1158
+ "language_model.model.layers.34.self_attn.k_proj": true,
1159
+ "language_model.model.layers.34.self_attn.v_proj": true,
1160
+ "language_model.model.layers.34.self_attn.o_proj": true,
1161
+ "language_model.model.layers.34.mlp.gate_proj": true,
1162
+ "language_model.model.layers.34.mlp.down_proj": true,
1163
+ "language_model.model.layers.34.mlp.up_proj": true,
1164
+ "language_model.model.layers.34.input_layernorm": false,
1165
+ "language_model.model.layers.34.post_attention_layernorm": false,
1166
+ "language_model.model.layers.35.self_attn.q_proj": true,
1167
+ "language_model.model.layers.35.self_attn.k_proj": true,
1168
+ "language_model.model.layers.35.self_attn.v_proj": true,
1169
+ "language_model.model.layers.35.self_attn.o_proj": true,
1170
+ "language_model.model.layers.35.mlp.gate_proj": true,
1171
+ "language_model.model.layers.35.mlp.down_proj": true,
1172
+ "language_model.model.layers.35.mlp.up_proj": true,
1173
+ "language_model.model.layers.35.input_layernorm": false,
1174
+ "language_model.model.layers.35.post_attention_layernorm": false,
1175
+ "language_model.model.norm": false
1176
  },
1177
  "remove_invalid_values": false,
1178
  "repetition_penalty": 1.0,
 
1180
  "return_dict_in_generate": false,
1181
  "rms_norm_eps": 1e-06,
1182
  "rope_scaling": {
 
1183
  "mrope_section": [
1184
  16,
1185
  24,
1186
  24
1187
  ],
1188
+ "rope_type": "default",
1189
+ "type": "default"
1190
  },
1191
  "rope_theta": 1000000.0,
1192
  "sep_token_id": null,
 
1202
  "num_hidden_layers": 36,
1203
  "num_attention_heads": 16,
1204
  "use_sliding_window": false,
1205
+ "sliding_window": null,
1206
  "max_window_layers": 70,
1207
  "num_key_value_heads": 2,
1208
  "hidden_act": "silu",
 
1212
  "rope_theta": 1000000.0,
1213
  "attention_dropout": 0.0,
1214
  "rope_scaling": {
 
1215
  "mrope_section": [
1216
  16,
1217
  24,
1218
  24
1219
  ],
1220
+ "rope_type": "default",
1221
+ "type": "default"
1222
  },
1223
+ "layer_types": [
1224
+ "full_attention",
1225
+ "full_attention",
1226
+ "full_attention",
1227
+ "full_attention",
1228
+ "full_attention",
1229
+ "full_attention",
1230
+ "full_attention",
1231
+ "full_attention",
1232
+ "full_attention",
1233
+ "full_attention",
1234
+ "full_attention",
1235
+ "full_attention",
1236
+ "full_attention",
1237
+ "full_attention",
1238
+ "full_attention",
1239
+ "full_attention",
1240
+ "full_attention",
1241
+ "full_attention",
1242
+ "full_attention",
1243
+ "full_attention",
1244
+ "full_attention",
1245
+ "full_attention",
1246
+ "full_attention",
1247
+ "full_attention",
1248
+ "full_attention",
1249
+ "full_attention",
1250
+ "full_attention",
1251
+ "full_attention",
1252
+ "full_attention",
1253
+ "full_attention",
1254
+ "full_attention",
1255
+ "full_attention",
1256
+ "full_attention",
1257
+ "full_attention",
1258
+ "full_attention",
1259
+ "full_attention"
1260
+ ],
1261
+ "image_token_id": null,
1262
+ "video_token_id": null,
1263
  "return_dict": true,
1264
  "output_hidden_states": false,
 
1265
  "torchscript": false,
1266
  "torch_dtype": "bfloat16",
 
 
1267
  "pruned_heads": {},
1268
  "tie_word_embeddings": true,
1269
  "chunk_size_feed_forward": 0,
 
1272
  "cross_attention_hidden_size": null,
1273
  "add_cross_attention": false,
1274
  "tie_encoder_decoder": false,
1275
+ "architectures": [
1276
+ "Qwen2_5_VLForConditionalGeneration"
1277
+ ],
1278
+ "finetuning_task": null,
1279
+ "id2label": {
1280
+ "0": "LABEL_0",
1281
+ "1": "LABEL_1"
1282
+ },
1283
+ "label2id": {
1284
+ "LABEL_0": 0,
1285
+ "LABEL_1": 1
1286
+ },
1287
+ "task_specific_params": null,
1288
+ "problem_type": null,
1289
+ "tokenizer_class": null,
1290
+ "prefix": null,
1291
+ "bos_token_id": null,
1292
+ "pad_token_id": 151654,
1293
+ "eos_token_id": 151645,
1294
+ "sep_token_id": null,
1295
+ "decoder_start_token_id": null,
1296
  "max_length": 20,
1297
  "min_length": 0,
1298
  "do_sample": false,
 
1318
  "exponential_decay_length_penalty": null,
1319
  "suppress_tokens": null,
1320
  "begin_suppress_tokens": null,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1321
  "_name_or_path": "",
1322
+ "model_type": "qwen2_5_vl_text",
1323
+ "unsloth_fixed": true,
1324
  "vision_end_token_id": 151653,
1325
+ "vision_start_token_id": 151652,
1326
  "vision_token_id": 151654,
1327
+ "tf_legacy_loss": false,
1328
+ "use_bfloat16": false,
1329
+ "output_attentions": false
1330
  },
1331
  "tf_legacy_loss": false,
1332
  "tie_encoder_decoder": false,
 
1334
  "tokenizer_class": null,
1335
  "top_k": 50,
1336
  "top_p": 1.0,
 
1337
  "torchscript": false,
1338
+ "transformers_version": "4.55.0.dev0",
1339
  "typical_p": 1.0,
1340
+ "unsloth_fixed": true,
1341
  "use_bfloat16": false,
1342
  "use_cache": true,
1343
  "use_sliding_window": false,
 
1345
  "vision_config": {
1346
  "return_dict": true,
1347
  "output_hidden_states": false,
 
1348
  "torchscript": false,
1349
+ "torch_dtype": "bfloat16",
 
 
1350
  "pruned_heads": {},
1351
  "tie_word_embeddings": true,
1352
  "chunk_size_feed_forward": 0,
 
1355
  "cross_attention_hidden_size": null,
1356
  "add_cross_attention": false,
1357
  "tie_encoder_decoder": false,
1358
+ "architectures": null,
1359
+ "finetuning_task": null,
1360
+ "id2label": {
1361
+ "0": "LABEL_0",
1362
+ "1": "LABEL_1"
1363
+ },
1364
+ "label2id": {
1365
+ "LABEL_0": 0,
1366
+ "LABEL_1": 1
1367
+ },
1368
+ "task_specific_params": null,
1369
+ "problem_type": null,
1370
+ "tokenizer_class": null,
1371
+ "prefix": null,
1372
+ "bos_token_id": null,
1373
+ "pad_token_id": null,
1374
+ "eos_token_id": null,
1375
+ "sep_token_id": null,
1376
+ "decoder_start_token_id": null,
1377
  "max_length": 20,
1378
  "min_length": 0,
1379
  "do_sample": false,
 
1399
  "exponential_decay_length_penalty": null,
1400
  "suppress_tokens": null,
1401
  "begin_suppress_tokens": null,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1402
  "_name_or_path": "",
 
1403
  "in_chans": 3,
1404
+ "model_type": "qwen2_5_vl",
1405
  "spatial_patch_size": 14,
1406
+ "tf_legacy_loss": false,
1407
+ "use_bfloat16": false,
1408
  "depth": 32,
1409
  "hidden_size": 1280,
1410
  "hidden_act": "silu",
 
1424
  ],
1425
  "out_hidden_size": 2048,
1426
  "initializer_range": 0.02,
1427
+ "output_attentions": false
1428
  },
1429
  "vision_end_token_id": 151653,
1430
  "vision_start_token_id": 151652,
generation_config.json CHANGED
@@ -1,12 +1,13 @@
1
  {
2
  "bos_token_id": 151643,
3
- "pad_token_id": 151643,
4
  "do_sample": true,
5
  "eos_token_id": [
6
  151645,
7
  151643
8
  ],
 
 
9
  "repetition_penalty": 1.05,
10
- "temperature": 0.000001,
11
- "transformers_version": "4.49.0"
12
- }
 
1
  {
2
  "bos_token_id": 151643,
 
3
  "do_sample": true,
4
  "eos_token_id": [
5
  151645,
6
  151643
7
  ],
8
+ "max_length": 128000,
9
+ "pad_token_id": 151654,
10
  "repetition_penalty": 1.05,
11
+ "temperature": 1e-06,
12
+ "transformers_version": "4.51.3"
13
+ }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f541d3f5e5d7ab47e185c80b849a74a04bc9c52bb9864a7d9973eab01d15cdcc
3
- size 4122929834
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b206c60844ee7560a986bd28c1013bda562af5593a7162ec9ef40d591251640
3
+ size 4616570154
model.safetensors.index.json CHANGED
@@ -65,9 +65,9 @@
65
  "model.layers.12.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
66
  "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
67
  "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
68
- "model.layers.13.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
69
- "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
70
- "model.layers.13.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
71
  "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
72
  "model.layers.13.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
73
  "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
@@ -76,78 +76,78 @@
76
  "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
77
  "model.layers.13.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
78
  "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
79
- "model.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors",
80
- "model.layers.14.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
81
- "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
82
- "model.layers.14.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
83
- "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
84
- "model.layers.14.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
85
- "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
86
- "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
87
- "model.layers.14.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
88
- "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
89
- "model.layers.14.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
90
- "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
91
- "model.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors",
92
- "model.layers.15.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
93
- "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
94
- "model.layers.15.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
95
- "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
96
- "model.layers.15.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
97
- "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
98
- "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
99
- "model.layers.15.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
100
- "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
101
- "model.layers.15.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
102
- "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
103
- "model.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors",
104
- "model.layers.16.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
105
- "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
106
- "model.layers.16.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
107
- "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
108
- "model.layers.16.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
109
- "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
110
- "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
111
- "model.layers.16.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
112
- "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
113
- "model.layers.16.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
114
- "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
115
- "model.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors",
116
- "model.layers.17.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
117
- "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
118
- "model.layers.17.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
119
- "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
120
- "model.layers.17.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
121
- "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
122
- "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
123
- "model.layers.17.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
124
- "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
125
- "model.layers.17.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
126
- "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
127
- "model.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors",
128
- "model.layers.18.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
129
- "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
130
- "model.layers.18.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
131
- "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
132
- "model.layers.18.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
133
- "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
134
- "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
135
- "model.layers.18.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
136
- "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
137
- "model.layers.18.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
138
- "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
139
  "model.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors",
140
  "model.layers.19.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
141
- "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
142
- "model.layers.19.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
143
  "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
144
- "model.layers.19.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
145
- "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
146
- "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
147
- "model.layers.19.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
148
- "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
149
- "model.layers.19.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
150
- "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
151
  "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
152
  "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
153
  "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
 
65
  "model.layers.12.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
66
  "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
67
  "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
68
+ "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
69
+ "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
70
+ "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
71
  "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
72
  "model.layers.13.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
73
  "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
 
76
  "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
77
  "model.layers.13.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
78
  "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
79
+ "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
80
+ "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
81
+ "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
82
+ "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
83
+ "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
84
+ "model.layers.14.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
85
+ "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
86
+ "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
87
+ "model.layers.14.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
88
+ "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
89
+ "model.layers.14.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
90
+ "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
91
+ "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
92
+ "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
93
+ "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
94
+ "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
95
+ "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
96
+ "model.layers.15.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
97
+ "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
98
+ "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
99
+ "model.layers.15.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
100
+ "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
101
+ "model.layers.15.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
102
+ "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
103
+ "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
104
+ "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
105
+ "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
106
+ "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
107
+ "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
108
+ "model.layers.16.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
109
+ "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
110
+ "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
111
+ "model.layers.16.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
112
+ "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
113
+ "model.layers.16.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
114
+ "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
115
+ "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
116
+ "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
117
+ "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
118
+ "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
119
+ "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
120
+ "model.layers.17.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
121
+ "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
122
+ "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
123
+ "model.layers.17.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
124
+ "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
125
+ "model.layers.17.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
126
+ "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
127
+ "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
128
+ "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
129
+ "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
130
+ "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
131
+ "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
132
+ "model.layers.18.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
133
+ "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
134
+ "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
135
+ "model.layers.18.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
136
+ "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
137
+ "model.layers.18.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
138
+ "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
139
  "model.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors",
140
  "model.layers.19.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
141
+ "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
142
+ "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
143
  "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
144
+ "model.layers.19.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
145
+ "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
146
+ "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
147
+ "model.layers.19.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
148
+ "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
149
+ "model.layers.19.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
150
+ "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
151
  "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
152
  "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
153
  "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
preprocessor_config.json CHANGED
@@ -1,4 +1,10 @@
1
  {
 
 
 
 
 
 
2
  "do_convert_rgb": true,
3
  "do_normalize": true,
4
  "do_rescale": true,
@@ -8,12 +14,13 @@
8
  0.4578275,
9
  0.40821073
10
  ],
11
- "image_processor_type": "Qwen2VLImageProcessor",
12
  "image_std": [
13
  0.26862954,
14
  0.26130258,
15
  0.27577711
16
  ],
 
17
  "max_pixels": 12845056,
18
  "merge_size": 2,
19
  "min_pixels": 3136,
@@ -21,6 +28,7 @@
21
  "processor_class": "Qwen2_5_VLProcessor",
22
  "resample": 3,
23
  "rescale_factor": 0.00392156862745098,
 
24
  "size": {
25
  "longest_edge": 12845056,
26
  "shortest_edge": 3136
 
1
  {
2
+ "crop_size": null,
3
+ "data_format": "channels_first",
4
+ "default_to_square": true,
5
+ "device": null,
6
+ "disable_grouping": null,
7
+ "do_center_crop": null,
8
  "do_convert_rgb": true,
9
  "do_normalize": true,
10
  "do_rescale": true,
 
14
  0.4578275,
15
  0.40821073
16
  ],
17
+ "image_processor_type": "Qwen2VLImageProcessorFast",
18
  "image_std": [
19
  0.26862954,
20
  0.26130258,
21
  0.27577711
22
  ],
23
+ "input_data_format": null,
24
  "max_pixels": 12845056,
25
  "merge_size": 2,
26
  "min_pixels": 3136,
 
28
  "processor_class": "Qwen2_5_VLProcessor",
29
  "resample": 3,
30
  "rescale_factor": 0.00392156862745098,
31
+ "return_tensors": null,
32
  "size": {
33
  "longest_edge": 12845056,
34
  "shortest_edge": 3136
special_tokens_map.json CHANGED
@@ -22,7 +22,7 @@
22
  "single_word": false
23
  },
24
  "pad_token": {
25
- "content": "<|endoftext|>",
26
  "lstrip": false,
27
  "normalized": false,
28
  "rstrip": false,
 
22
  "single_word": false
23
  },
24
  "pad_token": {
25
+ "content": "<|vision_pad|>",
26
  "lstrip": false,
27
  "normalized": false,
28
  "rstrip": false,
tokenizer_config.json CHANGED
@@ -199,8 +199,9 @@
199
  "eos_token": "<|im_end|>",
200
  "errors": "replace",
201
  "extra_special_tokens": {},
202
- "model_max_length": 131072,
203
- "pad_token": "<|endoftext|>",
 
204
  "processor_class": "Qwen2_5_VLProcessor",
205
  "split_special_tokens": false,
206
  "tokenizer_class": "Qwen2Tokenizer",
 
199
  "eos_token": "<|im_end|>",
200
  "errors": "replace",
201
  "extra_special_tokens": {},
202
+ "model_max_length": 128000,
203
+ "pad_token": "<|vision_pad|>",
204
+ "padding_side": "left",
205
  "processor_class": "Qwen2_5_VLProcessor",
206
  "split_special_tokens": false,
207
  "tokenizer_class": "Qwen2Tokenizer",
video_preprocessor_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": null,
3
+ "data_format": "channels_first",
4
+ "default_to_square": true,
5
+ "device": null,
6
+ "do_center_crop": null,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_pad": null,
10
+ "do_rescale": true,
11
+ "do_resize": true,
12
+ "do_sample_frames": false,
13
+ "fps": null,
14
+ "image_mean": [
15
+ 0.48145466,
16
+ 0.4578275,
17
+ 0.40821073
18
+ ],
19
+ "image_std": [
20
+ 0.26862954,
21
+ 0.26130258,
22
+ 0.27577711
23
+ ],
24
+ "input_data_format": null,
25
+ "max_frames": 768,
26
+ "max_pixels": 12845056,
27
+ "merge_size": 2,
28
+ "min_frames": 4,
29
+ "min_pixels": 3136,
30
+ "num_frames": null,
31
+ "patch_size": 14,
32
+ "processor_class": "Qwen2_5_VLProcessor",
33
+ "resample": 3,
34
+ "rescale_factor": 0.00392156862745098,
35
+ "size": {
36
+ "longest_edge": 12845056,
37
+ "shortest_edge": 3136
38
+ },
39
+ "size_divisor": null,
40
+ "temporal_patch_size": 2,
41
+ "video_metadata": null,
42
+ "video_processor_type": "Qwen2VLVideoProcessor"
43
+ }