Shubhra Pandit commited on
Commit
589ba0a
·
1 Parent(s): f026306

Upload model files

Browse files
added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
chat_template.json ADDED
@@ -0,0 +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
+ }
config.json ADDED
@@ -0,0 +1,977 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Qwen/Qwen2.5-VL-72B-Instruct",
3
+ "architectures": [
4
+ "Qwen2_5_VLForConditionalGeneration"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151645,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 8192,
11
+ "image_token_id": 151655,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 29568,
14
+ "max_position_embeddings": 128000,
15
+ "max_window_layers": 80,
16
+ "model_type": "qwen2_5_vl",
17
+ "num_attention_heads": 64,
18
+ "num_hidden_layers": 80,
19
+ "num_key_value_heads": 8,
20
+ "quantization_config": {
21
+ "config_groups": {
22
+ "group_0": {
23
+ "input_activations": null,
24
+ "output_activations": null,
25
+ "targets": [
26
+ "Linear"
27
+ ],
28
+ "weights": {
29
+ "actorder": null,
30
+ "block_structure": null,
31
+ "dynamic": false,
32
+ "group_size": 128,
33
+ "num_bits": 4,
34
+ "observer": "minmax",
35
+ "observer_kwargs": {},
36
+ "strategy": "group",
37
+ "symmetric": true,
38
+ "type": "int"
39
+ }
40
+ }
41
+ },
42
+ "format": "pack-quantized",
43
+ "global_compression_ratio": 1.5773885551629256,
44
+ "ignore": [
45
+ "visual.blocks.0.attn.qkv",
46
+ "visual.blocks.0.attn.proj",
47
+ "visual.blocks.0.mlp.gate_proj",
48
+ "visual.blocks.0.mlp.up_proj",
49
+ "visual.blocks.0.mlp.down_proj",
50
+ "visual.blocks.1.attn.qkv",
51
+ "visual.blocks.1.attn.proj",
52
+ "visual.blocks.1.mlp.gate_proj",
53
+ "visual.blocks.1.mlp.up_proj",
54
+ "visual.blocks.1.mlp.down_proj",
55
+ "visual.blocks.2.attn.qkv",
56
+ "visual.blocks.2.attn.proj",
57
+ "visual.blocks.2.mlp.gate_proj",
58
+ "visual.blocks.2.mlp.up_proj",
59
+ "visual.blocks.2.mlp.down_proj",
60
+ "visual.blocks.3.attn.qkv",
61
+ "visual.blocks.3.attn.proj",
62
+ "visual.blocks.3.mlp.gate_proj",
63
+ "visual.blocks.3.mlp.up_proj",
64
+ "visual.blocks.3.mlp.down_proj",
65
+ "visual.blocks.4.attn.qkv",
66
+ "visual.blocks.4.attn.proj",
67
+ "visual.blocks.4.mlp.gate_proj",
68
+ "visual.blocks.4.mlp.up_proj",
69
+ "visual.blocks.4.mlp.down_proj",
70
+ "visual.blocks.5.attn.qkv",
71
+ "visual.blocks.5.attn.proj",
72
+ "visual.blocks.5.mlp.gate_proj",
73
+ "visual.blocks.5.mlp.up_proj",
74
+ "visual.blocks.5.mlp.down_proj",
75
+ "visual.blocks.6.attn.qkv",
76
+ "visual.blocks.6.attn.proj",
77
+ "visual.blocks.6.mlp.gate_proj",
78
+ "visual.blocks.6.mlp.up_proj",
79
+ "visual.blocks.6.mlp.down_proj",
80
+ "visual.blocks.7.attn.qkv",
81
+ "visual.blocks.7.attn.proj",
82
+ "visual.blocks.7.mlp.gate_proj",
83
+ "visual.blocks.7.mlp.up_proj",
84
+ "visual.blocks.7.mlp.down_proj",
85
+ "visual.blocks.8.attn.qkv",
86
+ "visual.blocks.8.attn.proj",
87
+ "visual.blocks.8.mlp.gate_proj",
88
+ "visual.blocks.8.mlp.up_proj",
89
+ "visual.blocks.8.mlp.down_proj",
90
+ "visual.blocks.9.attn.qkv",
91
+ "visual.blocks.9.attn.proj",
92
+ "visual.blocks.9.mlp.gate_proj",
93
+ "visual.blocks.9.mlp.up_proj",
94
+ "visual.blocks.9.mlp.down_proj",
95
+ "visual.blocks.10.attn.qkv",
96
+ "visual.blocks.10.attn.proj",
97
+ "visual.blocks.10.mlp.gate_proj",
98
+ "visual.blocks.10.mlp.up_proj",
99
+ "visual.blocks.10.mlp.down_proj",
100
+ "visual.blocks.11.attn.qkv",
101
+ "visual.blocks.11.attn.proj",
102
+ "visual.blocks.11.mlp.gate_proj",
103
+ "visual.blocks.11.mlp.up_proj",
104
+ "visual.blocks.11.mlp.down_proj",
105
+ "visual.blocks.12.attn.qkv",
106
+ "visual.blocks.12.attn.proj",
107
+ "visual.blocks.12.mlp.gate_proj",
108
+ "visual.blocks.12.mlp.up_proj",
109
+ "visual.blocks.12.mlp.down_proj",
110
+ "visual.blocks.13.attn.qkv",
111
+ "visual.blocks.13.attn.proj",
112
+ "visual.blocks.13.mlp.gate_proj",
113
+ "visual.blocks.13.mlp.up_proj",
114
+ "visual.blocks.13.mlp.down_proj",
115
+ "visual.blocks.14.attn.qkv",
116
+ "visual.blocks.14.attn.proj",
117
+ "visual.blocks.14.mlp.gate_proj",
118
+ "visual.blocks.14.mlp.up_proj",
119
+ "visual.blocks.14.mlp.down_proj",
120
+ "visual.blocks.15.attn.qkv",
121
+ "visual.blocks.15.attn.proj",
122
+ "visual.blocks.15.mlp.gate_proj",
123
+ "visual.blocks.15.mlp.up_proj",
124
+ "visual.blocks.15.mlp.down_proj",
125
+ "visual.blocks.16.attn.qkv",
126
+ "visual.blocks.16.attn.proj",
127
+ "visual.blocks.16.mlp.gate_proj",
128
+ "visual.blocks.16.mlp.up_proj",
129
+ "visual.blocks.16.mlp.down_proj",
130
+ "visual.blocks.17.attn.qkv",
131
+ "visual.blocks.17.attn.proj",
132
+ "visual.blocks.17.mlp.gate_proj",
133
+ "visual.blocks.17.mlp.up_proj",
134
+ "visual.blocks.17.mlp.down_proj",
135
+ "visual.blocks.18.attn.qkv",
136
+ "visual.blocks.18.attn.proj",
137
+ "visual.blocks.18.mlp.gate_proj",
138
+ "visual.blocks.18.mlp.up_proj",
139
+ "visual.blocks.18.mlp.down_proj",
140
+ "visual.blocks.19.attn.qkv",
141
+ "visual.blocks.19.attn.proj",
142
+ "visual.blocks.19.mlp.gate_proj",
143
+ "visual.blocks.19.mlp.up_proj",
144
+ "visual.blocks.19.mlp.down_proj",
145
+ "visual.blocks.20.attn.qkv",
146
+ "visual.blocks.20.attn.proj",
147
+ "visual.blocks.20.mlp.gate_proj",
148
+ "visual.blocks.20.mlp.up_proj",
149
+ "visual.blocks.20.mlp.down_proj",
150
+ "visual.blocks.21.attn.qkv",
151
+ "visual.blocks.21.attn.proj",
152
+ "visual.blocks.21.mlp.gate_proj",
153
+ "visual.blocks.21.mlp.up_proj",
154
+ "visual.blocks.21.mlp.down_proj",
155
+ "visual.blocks.22.attn.qkv",
156
+ "visual.blocks.22.attn.proj",
157
+ "visual.blocks.22.mlp.gate_proj",
158
+ "visual.blocks.22.mlp.up_proj",
159
+ "visual.blocks.22.mlp.down_proj",
160
+ "visual.blocks.23.attn.qkv",
161
+ "visual.blocks.23.attn.proj",
162
+ "visual.blocks.23.mlp.gate_proj",
163
+ "visual.blocks.23.mlp.up_proj",
164
+ "visual.blocks.23.mlp.down_proj",
165
+ "visual.blocks.24.attn.qkv",
166
+ "visual.blocks.24.attn.proj",
167
+ "visual.blocks.24.mlp.gate_proj",
168
+ "visual.blocks.24.mlp.up_proj",
169
+ "visual.blocks.24.mlp.down_proj",
170
+ "visual.blocks.25.attn.qkv",
171
+ "visual.blocks.25.attn.proj",
172
+ "visual.blocks.25.mlp.gate_proj",
173
+ "visual.blocks.25.mlp.up_proj",
174
+ "visual.blocks.25.mlp.down_proj",
175
+ "visual.blocks.26.attn.qkv",
176
+ "visual.blocks.26.attn.proj",
177
+ "visual.blocks.26.mlp.gate_proj",
178
+ "visual.blocks.26.mlp.up_proj",
179
+ "visual.blocks.26.mlp.down_proj",
180
+ "visual.blocks.27.attn.qkv",
181
+ "visual.blocks.27.attn.proj",
182
+ "visual.blocks.27.mlp.gate_proj",
183
+ "visual.blocks.27.mlp.up_proj",
184
+ "visual.blocks.27.mlp.down_proj",
185
+ "visual.blocks.28.attn.qkv",
186
+ "visual.blocks.28.attn.proj",
187
+ "visual.blocks.28.mlp.gate_proj",
188
+ "visual.blocks.28.mlp.up_proj",
189
+ "visual.blocks.28.mlp.down_proj",
190
+ "visual.blocks.29.attn.qkv",
191
+ "visual.blocks.29.attn.proj",
192
+ "visual.blocks.29.mlp.gate_proj",
193
+ "visual.blocks.29.mlp.up_proj",
194
+ "visual.blocks.29.mlp.down_proj",
195
+ "visual.blocks.30.attn.qkv",
196
+ "visual.blocks.30.attn.proj",
197
+ "visual.blocks.30.mlp.gate_proj",
198
+ "visual.blocks.30.mlp.up_proj",
199
+ "visual.blocks.30.mlp.down_proj",
200
+ "visual.blocks.31.attn.qkv",
201
+ "visual.blocks.31.attn.proj",
202
+ "visual.blocks.31.mlp.gate_proj",
203
+ "visual.blocks.31.mlp.up_proj",
204
+ "visual.blocks.31.mlp.down_proj",
205
+ "visual.merger.mlp.0",
206
+ "visual.merger.mlp.2",
207
+ "lm_head"
208
+ ],
209
+ "kv_cache_scheme": null,
210
+ "quant_method": "compressed-tensors",
211
+ "quantization_status": "compressed",
212
+ "sparsity_config": {
213
+ "format": "dense",
214
+ "global_sparsity": 0.15522512352032172,
215
+ "ignore": [
216
+ "visual.blocks.0.attn.qkv",
217
+ "visual.blocks.0.attn.proj",
218
+ "visual.blocks.0.mlp.gate_proj",
219
+ "visual.blocks.0.mlp.up_proj",
220
+ "visual.blocks.0.mlp.down_proj",
221
+ "visual.blocks.1.attn.qkv",
222
+ "visual.blocks.1.attn.proj",
223
+ "visual.blocks.1.mlp.gate_proj",
224
+ "visual.blocks.1.mlp.up_proj",
225
+ "visual.blocks.1.mlp.down_proj",
226
+ "visual.blocks.2.attn.qkv",
227
+ "visual.blocks.2.attn.proj",
228
+ "visual.blocks.2.mlp.gate_proj",
229
+ "visual.blocks.2.mlp.up_proj",
230
+ "visual.blocks.2.mlp.down_proj",
231
+ "visual.blocks.3.attn.qkv",
232
+ "visual.blocks.3.attn.proj",
233
+ "visual.blocks.3.mlp.gate_proj",
234
+ "visual.blocks.3.mlp.up_proj",
235
+ "visual.blocks.3.mlp.down_proj",
236
+ "visual.blocks.4.attn.qkv",
237
+ "visual.blocks.4.attn.proj",
238
+ "visual.blocks.4.mlp.gate_proj",
239
+ "visual.blocks.4.mlp.up_proj",
240
+ "visual.blocks.4.mlp.down_proj",
241
+ "visual.blocks.5.attn.qkv",
242
+ "visual.blocks.5.attn.proj",
243
+ "visual.blocks.5.mlp.gate_proj",
244
+ "visual.blocks.5.mlp.up_proj",
245
+ "visual.blocks.5.mlp.down_proj",
246
+ "visual.blocks.6.attn.qkv",
247
+ "visual.blocks.6.attn.proj",
248
+ "visual.blocks.6.mlp.gate_proj",
249
+ "visual.blocks.6.mlp.up_proj",
250
+ "visual.blocks.6.mlp.down_proj",
251
+ "visual.blocks.7.attn.qkv",
252
+ "visual.blocks.7.attn.proj",
253
+ "visual.blocks.7.mlp.gate_proj",
254
+ "visual.blocks.7.mlp.up_proj",
255
+ "visual.blocks.7.mlp.down_proj",
256
+ "visual.blocks.8.attn.qkv",
257
+ "visual.blocks.8.attn.proj",
258
+ "visual.blocks.8.mlp.gate_proj",
259
+ "visual.blocks.8.mlp.up_proj",
260
+ "visual.blocks.8.mlp.down_proj",
261
+ "visual.blocks.9.attn.qkv",
262
+ "visual.blocks.9.attn.proj",
263
+ "visual.blocks.9.mlp.gate_proj",
264
+ "visual.blocks.9.mlp.up_proj",
265
+ "visual.blocks.9.mlp.down_proj",
266
+ "visual.blocks.10.attn.qkv",
267
+ "visual.blocks.10.attn.proj",
268
+ "visual.blocks.10.mlp.gate_proj",
269
+ "visual.blocks.10.mlp.up_proj",
270
+ "visual.blocks.10.mlp.down_proj",
271
+ "visual.blocks.11.attn.qkv",
272
+ "visual.blocks.11.attn.proj",
273
+ "visual.blocks.11.mlp.gate_proj",
274
+ "visual.blocks.11.mlp.up_proj",
275
+ "visual.blocks.11.mlp.down_proj",
276
+ "visual.blocks.12.attn.qkv",
277
+ "visual.blocks.12.attn.proj",
278
+ "visual.blocks.12.mlp.gate_proj",
279
+ "visual.blocks.12.mlp.up_proj",
280
+ "visual.blocks.12.mlp.down_proj",
281
+ "visual.blocks.13.attn.qkv",
282
+ "visual.blocks.13.attn.proj",
283
+ "visual.blocks.13.mlp.gate_proj",
284
+ "visual.blocks.13.mlp.up_proj",
285
+ "visual.blocks.13.mlp.down_proj",
286
+ "visual.blocks.14.attn.qkv",
287
+ "visual.blocks.14.attn.proj",
288
+ "visual.blocks.14.mlp.gate_proj",
289
+ "visual.blocks.14.mlp.up_proj",
290
+ "visual.blocks.14.mlp.down_proj",
291
+ "visual.blocks.15.attn.qkv",
292
+ "visual.blocks.15.attn.proj",
293
+ "visual.blocks.15.mlp.gate_proj",
294
+ "visual.blocks.15.mlp.up_proj",
295
+ "visual.blocks.15.mlp.down_proj",
296
+ "visual.blocks.16.attn.qkv",
297
+ "visual.blocks.16.attn.proj",
298
+ "visual.blocks.16.mlp.gate_proj",
299
+ "visual.blocks.16.mlp.up_proj",
300
+ "visual.blocks.16.mlp.down_proj",
301
+ "visual.blocks.17.attn.qkv",
302
+ "visual.blocks.17.attn.proj",
303
+ "visual.blocks.17.mlp.gate_proj",
304
+ "visual.blocks.17.mlp.up_proj",
305
+ "visual.blocks.17.mlp.down_proj",
306
+ "visual.blocks.18.attn.qkv",
307
+ "visual.blocks.18.attn.proj",
308
+ "visual.blocks.18.mlp.gate_proj",
309
+ "visual.blocks.18.mlp.up_proj",
310
+ "visual.blocks.18.mlp.down_proj",
311
+ "visual.blocks.19.attn.qkv",
312
+ "visual.blocks.19.attn.proj",
313
+ "visual.blocks.19.mlp.gate_proj",
314
+ "visual.blocks.19.mlp.up_proj",
315
+ "visual.blocks.19.mlp.down_proj",
316
+ "visual.blocks.20.attn.qkv",
317
+ "visual.blocks.20.attn.proj",
318
+ "visual.blocks.20.mlp.gate_proj",
319
+ "visual.blocks.20.mlp.up_proj",
320
+ "visual.blocks.20.mlp.down_proj",
321
+ "visual.blocks.21.attn.qkv",
322
+ "visual.blocks.21.attn.proj",
323
+ "visual.blocks.21.mlp.gate_proj",
324
+ "visual.blocks.21.mlp.up_proj",
325
+ "visual.blocks.21.mlp.down_proj",
326
+ "visual.blocks.22.attn.qkv",
327
+ "visual.blocks.22.attn.proj",
328
+ "visual.blocks.22.mlp.gate_proj",
329
+ "visual.blocks.22.mlp.up_proj",
330
+ "visual.blocks.22.mlp.down_proj",
331
+ "visual.blocks.23.attn.qkv",
332
+ "visual.blocks.23.attn.proj",
333
+ "visual.blocks.23.mlp.gate_proj",
334
+ "visual.blocks.23.mlp.up_proj",
335
+ "visual.blocks.23.mlp.down_proj",
336
+ "visual.blocks.24.attn.qkv",
337
+ "visual.blocks.24.attn.proj",
338
+ "visual.blocks.24.mlp.gate_proj",
339
+ "visual.blocks.24.mlp.up_proj",
340
+ "visual.blocks.24.mlp.down_proj",
341
+ "visual.blocks.25.attn.qkv",
342
+ "visual.blocks.25.attn.proj",
343
+ "visual.blocks.25.mlp.gate_proj",
344
+ "visual.blocks.25.mlp.up_proj",
345
+ "visual.blocks.25.mlp.down_proj",
346
+ "visual.blocks.26.attn.qkv",
347
+ "visual.blocks.26.attn.proj",
348
+ "visual.blocks.26.mlp.gate_proj",
349
+ "visual.blocks.26.mlp.up_proj",
350
+ "visual.blocks.26.mlp.down_proj",
351
+ "visual.blocks.27.attn.qkv",
352
+ "visual.blocks.27.attn.proj",
353
+ "visual.blocks.27.mlp.gate_proj",
354
+ "visual.blocks.27.mlp.up_proj",
355
+ "visual.blocks.27.mlp.down_proj",
356
+ "visual.blocks.28.attn.qkv",
357
+ "visual.blocks.28.attn.proj",
358
+ "visual.blocks.28.mlp.gate_proj",
359
+ "visual.blocks.28.mlp.up_proj",
360
+ "visual.blocks.28.mlp.down_proj",
361
+ "visual.blocks.29.attn.qkv",
362
+ "visual.blocks.29.attn.proj",
363
+ "visual.blocks.29.mlp.gate_proj",
364
+ "visual.blocks.29.mlp.up_proj",
365
+ "visual.blocks.29.mlp.down_proj",
366
+ "visual.blocks.30.attn.qkv",
367
+ "visual.blocks.30.attn.proj",
368
+ "visual.blocks.30.mlp.gate_proj",
369
+ "visual.blocks.30.mlp.up_proj",
370
+ "visual.blocks.30.mlp.down_proj",
371
+ "visual.blocks.31.attn.qkv",
372
+ "visual.blocks.31.attn.proj",
373
+ "visual.blocks.31.mlp.gate_proj",
374
+ "visual.blocks.31.mlp.up_proj",
375
+ "visual.blocks.31.mlp.down_proj",
376
+ "visual.merger.mlp.0",
377
+ "visual.merger.mlp.2",
378
+ "model.layers.0.self_attn.q_proj",
379
+ "model.layers.0.self_attn.k_proj",
380
+ "model.layers.0.self_attn.v_proj",
381
+ "model.layers.0.self_attn.o_proj",
382
+ "model.layers.0.mlp.gate_proj",
383
+ "model.layers.0.mlp.up_proj",
384
+ "model.layers.0.mlp.down_proj",
385
+ "model.layers.1.self_attn.v_proj",
386
+ "model.layers.1.self_attn.o_proj",
387
+ "model.layers.1.mlp.gate_proj",
388
+ "model.layers.1.mlp.up_proj",
389
+ "model.layers.1.mlp.down_proj",
390
+ "model.layers.2.self_attn.v_proj",
391
+ "model.layers.2.self_attn.o_proj",
392
+ "model.layers.2.mlp.gate_proj",
393
+ "model.layers.2.mlp.up_proj",
394
+ "model.layers.2.mlp.down_proj",
395
+ "model.layers.3.self_attn.k_proj",
396
+ "model.layers.3.self_attn.v_proj",
397
+ "model.layers.3.self_attn.o_proj",
398
+ "model.layers.3.mlp.gate_proj",
399
+ "model.layers.3.mlp.up_proj",
400
+ "model.layers.3.mlp.down_proj",
401
+ "model.layers.4.self_attn.q_proj",
402
+ "model.layers.4.self_attn.k_proj",
403
+ "model.layers.4.self_attn.v_proj",
404
+ "model.layers.4.self_attn.o_proj",
405
+ "model.layers.4.mlp.gate_proj",
406
+ "model.layers.4.mlp.up_proj",
407
+ "model.layers.4.mlp.down_proj",
408
+ "model.layers.5.self_attn.k_proj",
409
+ "model.layers.5.self_attn.v_proj",
410
+ "model.layers.5.self_attn.o_proj",
411
+ "model.layers.5.mlp.gate_proj",
412
+ "model.layers.5.mlp.up_proj",
413
+ "model.layers.5.mlp.down_proj",
414
+ "model.layers.6.self_attn.q_proj",
415
+ "model.layers.6.self_attn.k_proj",
416
+ "model.layers.6.self_attn.v_proj",
417
+ "model.layers.6.self_attn.o_proj",
418
+ "model.layers.6.mlp.gate_proj",
419
+ "model.layers.6.mlp.up_proj",
420
+ "model.layers.6.mlp.down_proj",
421
+ "model.layers.7.self_attn.q_proj",
422
+ "model.layers.7.self_attn.k_proj",
423
+ "model.layers.7.self_attn.v_proj",
424
+ "model.layers.7.self_attn.o_proj",
425
+ "model.layers.7.mlp.gate_proj",
426
+ "model.layers.7.mlp.up_proj",
427
+ "model.layers.7.mlp.down_proj",
428
+ "model.layers.8.self_attn.q_proj",
429
+ "model.layers.8.self_attn.k_proj",
430
+ "model.layers.8.self_attn.v_proj",
431
+ "model.layers.8.self_attn.o_proj",
432
+ "model.layers.8.mlp.gate_proj",
433
+ "model.layers.8.mlp.up_proj",
434
+ "model.layers.8.mlp.down_proj",
435
+ "model.layers.9.self_attn.q_proj",
436
+ "model.layers.9.self_attn.k_proj",
437
+ "model.layers.9.self_attn.v_proj",
438
+ "model.layers.9.self_attn.o_proj",
439
+ "model.layers.9.mlp.gate_proj",
440
+ "model.layers.9.mlp.up_proj",
441
+ "model.layers.9.mlp.down_proj",
442
+ "model.layers.10.self_attn.q_proj",
443
+ "model.layers.10.self_attn.k_proj",
444
+ "model.layers.10.self_attn.v_proj",
445
+ "model.layers.10.self_attn.o_proj",
446
+ "model.layers.10.mlp.gate_proj",
447
+ "model.layers.10.mlp.up_proj",
448
+ "model.layers.10.mlp.down_proj",
449
+ "model.layers.11.self_attn.q_proj",
450
+ "model.layers.11.self_attn.k_proj",
451
+ "model.layers.11.self_attn.v_proj",
452
+ "model.layers.11.self_attn.o_proj",
453
+ "model.layers.11.mlp.gate_proj",
454
+ "model.layers.11.mlp.up_proj",
455
+ "model.layers.11.mlp.down_proj",
456
+ "model.layers.12.self_attn.q_proj",
457
+ "model.layers.12.self_attn.k_proj",
458
+ "model.layers.12.self_attn.v_proj",
459
+ "model.layers.12.self_attn.o_proj",
460
+ "model.layers.12.mlp.gate_proj",
461
+ "model.layers.12.mlp.up_proj",
462
+ "model.layers.12.mlp.down_proj",
463
+ "model.layers.13.self_attn.q_proj",
464
+ "model.layers.13.self_attn.k_proj",
465
+ "model.layers.13.self_attn.v_proj",
466
+ "model.layers.13.self_attn.o_proj",
467
+ "model.layers.13.mlp.gate_proj",
468
+ "model.layers.13.mlp.up_proj",
469
+ "model.layers.13.mlp.down_proj",
470
+ "model.layers.14.self_attn.q_proj",
471
+ "model.layers.14.self_attn.k_proj",
472
+ "model.layers.14.self_attn.v_proj",
473
+ "model.layers.14.self_attn.o_proj",
474
+ "model.layers.14.mlp.gate_proj",
475
+ "model.layers.14.mlp.up_proj",
476
+ "model.layers.14.mlp.down_proj",
477
+ "model.layers.15.self_attn.q_proj",
478
+ "model.layers.15.self_attn.k_proj",
479
+ "model.layers.15.self_attn.v_proj",
480
+ "model.layers.15.self_attn.o_proj",
481
+ "model.layers.15.mlp.gate_proj",
482
+ "model.layers.15.mlp.up_proj",
483
+ "model.layers.15.mlp.down_proj",
484
+ "model.layers.16.self_attn.q_proj",
485
+ "model.layers.16.self_attn.k_proj",
486
+ "model.layers.16.self_attn.v_proj",
487
+ "model.layers.16.self_attn.o_proj",
488
+ "model.layers.16.mlp.gate_proj",
489
+ "model.layers.16.mlp.up_proj",
490
+ "model.layers.16.mlp.down_proj",
491
+ "model.layers.17.self_attn.q_proj",
492
+ "model.layers.17.self_attn.k_proj",
493
+ "model.layers.17.self_attn.v_proj",
494
+ "model.layers.17.self_attn.o_proj",
495
+ "model.layers.17.mlp.gate_proj",
496
+ "model.layers.17.mlp.up_proj",
497
+ "model.layers.17.mlp.down_proj",
498
+ "model.layers.18.self_attn.q_proj",
499
+ "model.layers.18.self_attn.k_proj",
500
+ "model.layers.18.self_attn.v_proj",
501
+ "model.layers.18.self_attn.o_proj",
502
+ "model.layers.18.mlp.gate_proj",
503
+ "model.layers.18.mlp.up_proj",
504
+ "model.layers.18.mlp.down_proj",
505
+ "model.layers.19.self_attn.q_proj",
506
+ "model.layers.19.self_attn.k_proj",
507
+ "model.layers.19.self_attn.v_proj",
508
+ "model.layers.19.self_attn.o_proj",
509
+ "model.layers.19.mlp.gate_proj",
510
+ "model.layers.19.mlp.up_proj",
511
+ "model.layers.19.mlp.down_proj",
512
+ "model.layers.20.self_attn.q_proj",
513
+ "model.layers.20.self_attn.k_proj",
514
+ "model.layers.20.self_attn.v_proj",
515
+ "model.layers.20.self_attn.o_proj",
516
+ "model.layers.20.mlp.gate_proj",
517
+ "model.layers.20.mlp.up_proj",
518
+ "model.layers.20.mlp.down_proj",
519
+ "model.layers.21.self_attn.q_proj",
520
+ "model.layers.21.self_attn.k_proj",
521
+ "model.layers.21.self_attn.v_proj",
522
+ "model.layers.21.self_attn.o_proj",
523
+ "model.layers.21.mlp.gate_proj",
524
+ "model.layers.21.mlp.up_proj",
525
+ "model.layers.21.mlp.down_proj",
526
+ "model.layers.22.self_attn.q_proj",
527
+ "model.layers.22.self_attn.k_proj",
528
+ "model.layers.22.self_attn.v_proj",
529
+ "model.layers.22.self_attn.o_proj",
530
+ "model.layers.22.mlp.gate_proj",
531
+ "model.layers.22.mlp.up_proj",
532
+ "model.layers.22.mlp.down_proj",
533
+ "model.layers.23.self_attn.q_proj",
534
+ "model.layers.23.self_attn.k_proj",
535
+ "model.layers.23.self_attn.v_proj",
536
+ "model.layers.23.self_attn.o_proj",
537
+ "model.layers.23.mlp.gate_proj",
538
+ "model.layers.23.mlp.up_proj",
539
+ "model.layers.23.mlp.down_proj",
540
+ "model.layers.24.self_attn.q_proj",
541
+ "model.layers.24.self_attn.k_proj",
542
+ "model.layers.24.self_attn.v_proj",
543
+ "model.layers.24.self_attn.o_proj",
544
+ "model.layers.24.mlp.gate_proj",
545
+ "model.layers.24.mlp.up_proj",
546
+ "model.layers.24.mlp.down_proj",
547
+ "model.layers.25.self_attn.q_proj",
548
+ "model.layers.25.self_attn.k_proj",
549
+ "model.layers.25.self_attn.v_proj",
550
+ "model.layers.25.self_attn.o_proj",
551
+ "model.layers.25.mlp.gate_proj",
552
+ "model.layers.25.mlp.up_proj",
553
+ "model.layers.25.mlp.down_proj",
554
+ "model.layers.26.self_attn.q_proj",
555
+ "model.layers.26.self_attn.k_proj",
556
+ "model.layers.26.self_attn.v_proj",
557
+ "model.layers.26.self_attn.o_proj",
558
+ "model.layers.26.mlp.gate_proj",
559
+ "model.layers.26.mlp.up_proj",
560
+ "model.layers.26.mlp.down_proj",
561
+ "model.layers.27.self_attn.q_proj",
562
+ "model.layers.27.self_attn.k_proj",
563
+ "model.layers.27.self_attn.v_proj",
564
+ "model.layers.27.self_attn.o_proj",
565
+ "model.layers.27.mlp.gate_proj",
566
+ "model.layers.27.mlp.up_proj",
567
+ "model.layers.27.mlp.down_proj",
568
+ "model.layers.28.self_attn.q_proj",
569
+ "model.layers.28.self_attn.k_proj",
570
+ "model.layers.28.self_attn.v_proj",
571
+ "model.layers.28.self_attn.o_proj",
572
+ "model.layers.28.mlp.gate_proj",
573
+ "model.layers.28.mlp.up_proj",
574
+ "model.layers.28.mlp.down_proj",
575
+ "model.layers.29.self_attn.q_proj",
576
+ "model.layers.29.self_attn.k_proj",
577
+ "model.layers.29.self_attn.v_proj",
578
+ "model.layers.29.self_attn.o_proj",
579
+ "model.layers.29.mlp.gate_proj",
580
+ "model.layers.29.mlp.up_proj",
581
+ "model.layers.29.mlp.down_proj",
582
+ "model.layers.30.self_attn.q_proj",
583
+ "model.layers.30.self_attn.k_proj",
584
+ "model.layers.30.self_attn.v_proj",
585
+ "model.layers.30.self_attn.o_proj",
586
+ "model.layers.30.mlp.gate_proj",
587
+ "model.layers.30.mlp.up_proj",
588
+ "model.layers.30.mlp.down_proj",
589
+ "model.layers.31.self_attn.q_proj",
590
+ "model.layers.31.self_attn.k_proj",
591
+ "model.layers.31.self_attn.v_proj",
592
+ "model.layers.31.self_attn.o_proj",
593
+ "model.layers.31.mlp.gate_proj",
594
+ "model.layers.31.mlp.up_proj",
595
+ "model.layers.31.mlp.down_proj",
596
+ "model.layers.32.self_attn.q_proj",
597
+ "model.layers.32.self_attn.k_proj",
598
+ "model.layers.32.self_attn.v_proj",
599
+ "model.layers.32.self_attn.o_proj",
600
+ "model.layers.32.mlp.gate_proj",
601
+ "model.layers.32.mlp.up_proj",
602
+ "model.layers.32.mlp.down_proj",
603
+ "model.layers.33.self_attn.q_proj",
604
+ "model.layers.33.self_attn.k_proj",
605
+ "model.layers.33.self_attn.v_proj",
606
+ "model.layers.33.self_attn.o_proj",
607
+ "model.layers.33.mlp.gate_proj",
608
+ "model.layers.33.mlp.up_proj",
609
+ "model.layers.33.mlp.down_proj",
610
+ "model.layers.34.self_attn.q_proj",
611
+ "model.layers.34.self_attn.k_proj",
612
+ "model.layers.34.self_attn.v_proj",
613
+ "model.layers.34.self_attn.o_proj",
614
+ "model.layers.34.mlp.gate_proj",
615
+ "model.layers.34.mlp.up_proj",
616
+ "model.layers.34.mlp.down_proj",
617
+ "model.layers.35.self_attn.q_proj",
618
+ "model.layers.35.self_attn.k_proj",
619
+ "model.layers.35.self_attn.v_proj",
620
+ "model.layers.35.self_attn.o_proj",
621
+ "model.layers.35.mlp.gate_proj",
622
+ "model.layers.35.mlp.up_proj",
623
+ "model.layers.35.mlp.down_proj",
624
+ "model.layers.36.self_attn.q_proj",
625
+ "model.layers.36.self_attn.k_proj",
626
+ "model.layers.36.self_attn.v_proj",
627
+ "model.layers.36.self_attn.o_proj",
628
+ "model.layers.36.mlp.gate_proj",
629
+ "model.layers.36.mlp.up_proj",
630
+ "model.layers.36.mlp.down_proj",
631
+ "model.layers.37.self_attn.q_proj",
632
+ "model.layers.37.self_attn.k_proj",
633
+ "model.layers.37.self_attn.v_proj",
634
+ "model.layers.37.self_attn.o_proj",
635
+ "model.layers.37.mlp.gate_proj",
636
+ "model.layers.37.mlp.up_proj",
637
+ "model.layers.37.mlp.down_proj",
638
+ "model.layers.38.self_attn.q_proj",
639
+ "model.layers.38.self_attn.k_proj",
640
+ "model.layers.38.self_attn.v_proj",
641
+ "model.layers.38.self_attn.o_proj",
642
+ "model.layers.38.mlp.gate_proj",
643
+ "model.layers.38.mlp.up_proj",
644
+ "model.layers.38.mlp.down_proj",
645
+ "model.layers.39.self_attn.q_proj",
646
+ "model.layers.39.self_attn.k_proj",
647
+ "model.layers.39.self_attn.v_proj",
648
+ "model.layers.39.self_attn.o_proj",
649
+ "model.layers.39.mlp.gate_proj",
650
+ "model.layers.39.mlp.up_proj",
651
+ "model.layers.39.mlp.down_proj",
652
+ "model.layers.40.self_attn.q_proj",
653
+ "model.layers.40.self_attn.k_proj",
654
+ "model.layers.40.self_attn.v_proj",
655
+ "model.layers.40.self_attn.o_proj",
656
+ "model.layers.40.mlp.gate_proj",
657
+ "model.layers.40.mlp.up_proj",
658
+ "model.layers.40.mlp.down_proj",
659
+ "model.layers.41.self_attn.q_proj",
660
+ "model.layers.41.self_attn.k_proj",
661
+ "model.layers.41.self_attn.v_proj",
662
+ "model.layers.41.self_attn.o_proj",
663
+ "model.layers.41.mlp.gate_proj",
664
+ "model.layers.41.mlp.up_proj",
665
+ "model.layers.41.mlp.down_proj",
666
+ "model.layers.42.self_attn.q_proj",
667
+ "model.layers.42.self_attn.k_proj",
668
+ "model.layers.42.self_attn.v_proj",
669
+ "model.layers.42.self_attn.o_proj",
670
+ "model.layers.42.mlp.gate_proj",
671
+ "model.layers.42.mlp.up_proj",
672
+ "model.layers.42.mlp.down_proj",
673
+ "model.layers.43.self_attn.q_proj",
674
+ "model.layers.43.self_attn.k_proj",
675
+ "model.layers.43.self_attn.v_proj",
676
+ "model.layers.43.self_attn.o_proj",
677
+ "model.layers.43.mlp.gate_proj",
678
+ "model.layers.43.mlp.up_proj",
679
+ "model.layers.43.mlp.down_proj",
680
+ "model.layers.44.self_attn.q_proj",
681
+ "model.layers.44.self_attn.k_proj",
682
+ "model.layers.44.self_attn.v_proj",
683
+ "model.layers.44.self_attn.o_proj",
684
+ "model.layers.44.mlp.gate_proj",
685
+ "model.layers.44.mlp.up_proj",
686
+ "model.layers.44.mlp.down_proj",
687
+ "model.layers.45.self_attn.q_proj",
688
+ "model.layers.45.self_attn.k_proj",
689
+ "model.layers.45.self_attn.v_proj",
690
+ "model.layers.45.self_attn.o_proj",
691
+ "model.layers.45.mlp.gate_proj",
692
+ "model.layers.45.mlp.up_proj",
693
+ "model.layers.45.mlp.down_proj",
694
+ "model.layers.46.self_attn.q_proj",
695
+ "model.layers.46.self_attn.k_proj",
696
+ "model.layers.46.self_attn.v_proj",
697
+ "model.layers.46.self_attn.o_proj",
698
+ "model.layers.46.mlp.gate_proj",
699
+ "model.layers.46.mlp.up_proj",
700
+ "model.layers.46.mlp.down_proj",
701
+ "model.layers.47.self_attn.q_proj",
702
+ "model.layers.47.self_attn.k_proj",
703
+ "model.layers.47.self_attn.v_proj",
704
+ "model.layers.47.self_attn.o_proj",
705
+ "model.layers.47.mlp.gate_proj",
706
+ "model.layers.47.mlp.up_proj",
707
+ "model.layers.47.mlp.down_proj",
708
+ "model.layers.48.self_attn.q_proj",
709
+ "model.layers.48.self_attn.k_proj",
710
+ "model.layers.48.self_attn.v_proj",
711
+ "model.layers.48.self_attn.o_proj",
712
+ "model.layers.48.mlp.gate_proj",
713
+ "model.layers.48.mlp.up_proj",
714
+ "model.layers.48.mlp.down_proj",
715
+ "model.layers.49.self_attn.q_proj",
716
+ "model.layers.49.self_attn.k_proj",
717
+ "model.layers.49.self_attn.v_proj",
718
+ "model.layers.49.self_attn.o_proj",
719
+ "model.layers.49.mlp.gate_proj",
720
+ "model.layers.49.mlp.up_proj",
721
+ "model.layers.49.mlp.down_proj",
722
+ "model.layers.50.self_attn.q_proj",
723
+ "model.layers.50.self_attn.k_proj",
724
+ "model.layers.50.self_attn.v_proj",
725
+ "model.layers.50.self_attn.o_proj",
726
+ "model.layers.50.mlp.gate_proj",
727
+ "model.layers.50.mlp.up_proj",
728
+ "model.layers.50.mlp.down_proj",
729
+ "model.layers.51.self_attn.q_proj",
730
+ "model.layers.51.self_attn.k_proj",
731
+ "model.layers.51.self_attn.v_proj",
732
+ "model.layers.51.self_attn.o_proj",
733
+ "model.layers.51.mlp.gate_proj",
734
+ "model.layers.51.mlp.up_proj",
735
+ "model.layers.51.mlp.down_proj",
736
+ "model.layers.52.self_attn.q_proj",
737
+ "model.layers.52.self_attn.k_proj",
738
+ "model.layers.52.self_attn.v_proj",
739
+ "model.layers.52.self_attn.o_proj",
740
+ "model.layers.52.mlp.gate_proj",
741
+ "model.layers.52.mlp.up_proj",
742
+ "model.layers.52.mlp.down_proj",
743
+ "model.layers.53.self_attn.q_proj",
744
+ "model.layers.53.self_attn.k_proj",
745
+ "model.layers.53.self_attn.v_proj",
746
+ "model.layers.53.self_attn.o_proj",
747
+ "model.layers.53.mlp.gate_proj",
748
+ "model.layers.53.mlp.up_proj",
749
+ "model.layers.53.mlp.down_proj",
750
+ "model.layers.54.self_attn.q_proj",
751
+ "model.layers.54.self_attn.k_proj",
752
+ "model.layers.54.self_attn.v_proj",
753
+ "model.layers.54.self_attn.o_proj",
754
+ "model.layers.54.mlp.gate_proj",
755
+ "model.layers.54.mlp.up_proj",
756
+ "model.layers.54.mlp.down_proj",
757
+ "model.layers.55.self_attn.q_proj",
758
+ "model.layers.55.self_attn.k_proj",
759
+ "model.layers.55.self_attn.v_proj",
760
+ "model.layers.55.self_attn.o_proj",
761
+ "model.layers.55.mlp.gate_proj",
762
+ "model.layers.55.mlp.up_proj",
763
+ "model.layers.55.mlp.down_proj",
764
+ "model.layers.56.self_attn.q_proj",
765
+ "model.layers.56.self_attn.k_proj",
766
+ "model.layers.56.self_attn.v_proj",
767
+ "model.layers.56.self_attn.o_proj",
768
+ "model.layers.56.mlp.gate_proj",
769
+ "model.layers.56.mlp.up_proj",
770
+ "model.layers.56.mlp.down_proj",
771
+ "model.layers.57.self_attn.q_proj",
772
+ "model.layers.57.self_attn.k_proj",
773
+ "model.layers.57.self_attn.v_proj",
774
+ "model.layers.57.self_attn.o_proj",
775
+ "model.layers.57.mlp.gate_proj",
776
+ "model.layers.57.mlp.up_proj",
777
+ "model.layers.57.mlp.down_proj",
778
+ "model.layers.58.self_attn.q_proj",
779
+ "model.layers.58.self_attn.k_proj",
780
+ "model.layers.58.self_attn.v_proj",
781
+ "model.layers.58.self_attn.o_proj",
782
+ "model.layers.58.mlp.gate_proj",
783
+ "model.layers.58.mlp.up_proj",
784
+ "model.layers.58.mlp.down_proj",
785
+ "model.layers.59.self_attn.q_proj",
786
+ "model.layers.59.self_attn.k_proj",
787
+ "model.layers.59.self_attn.v_proj",
788
+ "model.layers.59.self_attn.o_proj",
789
+ "model.layers.59.mlp.gate_proj",
790
+ "model.layers.59.mlp.up_proj",
791
+ "model.layers.59.mlp.down_proj",
792
+ "model.layers.60.self_attn.q_proj",
793
+ "model.layers.60.self_attn.k_proj",
794
+ "model.layers.60.self_attn.v_proj",
795
+ "model.layers.60.self_attn.o_proj",
796
+ "model.layers.60.mlp.gate_proj",
797
+ "model.layers.60.mlp.up_proj",
798
+ "model.layers.60.mlp.down_proj",
799
+ "model.layers.61.self_attn.q_proj",
800
+ "model.layers.61.self_attn.k_proj",
801
+ "model.layers.61.self_attn.v_proj",
802
+ "model.layers.61.self_attn.o_proj",
803
+ "model.layers.61.mlp.gate_proj",
804
+ "model.layers.61.mlp.up_proj",
805
+ "model.layers.61.mlp.down_proj",
806
+ "model.layers.62.self_attn.q_proj",
807
+ "model.layers.62.self_attn.k_proj",
808
+ "model.layers.62.self_attn.v_proj",
809
+ "model.layers.62.self_attn.o_proj",
810
+ "model.layers.62.mlp.gate_proj",
811
+ "model.layers.62.mlp.up_proj",
812
+ "model.layers.62.mlp.down_proj",
813
+ "model.layers.63.self_attn.q_proj",
814
+ "model.layers.63.self_attn.k_proj",
815
+ "model.layers.63.self_attn.v_proj",
816
+ "model.layers.63.self_attn.o_proj",
817
+ "model.layers.63.mlp.gate_proj",
818
+ "model.layers.63.mlp.up_proj",
819
+ "model.layers.63.mlp.down_proj",
820
+ "model.layers.64.self_attn.q_proj",
821
+ "model.layers.64.self_attn.k_proj",
822
+ "model.layers.64.self_attn.v_proj",
823
+ "model.layers.64.self_attn.o_proj",
824
+ "model.layers.64.mlp.gate_proj",
825
+ "model.layers.64.mlp.up_proj",
826
+ "model.layers.64.mlp.down_proj",
827
+ "model.layers.65.self_attn.q_proj",
828
+ "model.layers.65.self_attn.k_proj",
829
+ "model.layers.65.self_attn.v_proj",
830
+ "model.layers.65.self_attn.o_proj",
831
+ "model.layers.65.mlp.gate_proj",
832
+ "model.layers.65.mlp.up_proj",
833
+ "model.layers.65.mlp.down_proj",
834
+ "model.layers.66.self_attn.q_proj",
835
+ "model.layers.66.self_attn.k_proj",
836
+ "model.layers.66.self_attn.v_proj",
837
+ "model.layers.66.self_attn.o_proj",
838
+ "model.layers.66.mlp.gate_proj",
839
+ "model.layers.66.mlp.up_proj",
840
+ "model.layers.66.mlp.down_proj",
841
+ "model.layers.67.self_attn.q_proj",
842
+ "model.layers.67.self_attn.k_proj",
843
+ "model.layers.67.self_attn.v_proj",
844
+ "model.layers.67.self_attn.o_proj",
845
+ "model.layers.67.mlp.gate_proj",
846
+ "model.layers.67.mlp.up_proj",
847
+ "model.layers.67.mlp.down_proj",
848
+ "model.layers.68.self_attn.q_proj",
849
+ "model.layers.68.self_attn.k_proj",
850
+ "model.layers.68.self_attn.v_proj",
851
+ "model.layers.68.self_attn.o_proj",
852
+ "model.layers.68.mlp.gate_proj",
853
+ "model.layers.68.mlp.up_proj",
854
+ "model.layers.68.mlp.down_proj",
855
+ "model.layers.69.self_attn.q_proj",
856
+ "model.layers.69.self_attn.k_proj",
857
+ "model.layers.69.self_attn.v_proj",
858
+ "model.layers.69.self_attn.o_proj",
859
+ "model.layers.69.mlp.gate_proj",
860
+ "model.layers.69.mlp.up_proj",
861
+ "model.layers.69.mlp.down_proj",
862
+ "model.layers.70.self_attn.q_proj",
863
+ "model.layers.70.self_attn.k_proj",
864
+ "model.layers.70.self_attn.v_proj",
865
+ "model.layers.70.self_attn.o_proj",
866
+ "model.layers.70.mlp.gate_proj",
867
+ "model.layers.70.mlp.up_proj",
868
+ "model.layers.70.mlp.down_proj",
869
+ "model.layers.71.self_attn.q_proj",
870
+ "model.layers.71.self_attn.k_proj",
871
+ "model.layers.71.self_attn.v_proj",
872
+ "model.layers.71.self_attn.o_proj",
873
+ "model.layers.71.mlp.gate_proj",
874
+ "model.layers.71.mlp.up_proj",
875
+ "model.layers.71.mlp.down_proj",
876
+ "model.layers.72.self_attn.q_proj",
877
+ "model.layers.72.self_attn.k_proj",
878
+ "model.layers.72.self_attn.v_proj",
879
+ "model.layers.72.self_attn.o_proj",
880
+ "model.layers.72.mlp.gate_proj",
881
+ "model.layers.72.mlp.up_proj",
882
+ "model.layers.72.mlp.down_proj",
883
+ "model.layers.73.self_attn.q_proj",
884
+ "model.layers.73.self_attn.k_proj",
885
+ "model.layers.73.self_attn.v_proj",
886
+ "model.layers.73.self_attn.o_proj",
887
+ "model.layers.73.mlp.gate_proj",
888
+ "model.layers.73.mlp.up_proj",
889
+ "model.layers.73.mlp.down_proj",
890
+ "model.layers.74.self_attn.q_proj",
891
+ "model.layers.74.self_attn.k_proj",
892
+ "model.layers.74.self_attn.v_proj",
893
+ "model.layers.74.self_attn.o_proj",
894
+ "model.layers.74.mlp.gate_proj",
895
+ "model.layers.74.mlp.up_proj",
896
+ "model.layers.74.mlp.down_proj",
897
+ "model.layers.75.self_attn.q_proj",
898
+ "model.layers.75.self_attn.k_proj",
899
+ "model.layers.75.self_attn.v_proj",
900
+ "model.layers.75.self_attn.o_proj",
901
+ "model.layers.75.mlp.gate_proj",
902
+ "model.layers.75.mlp.up_proj",
903
+ "model.layers.75.mlp.down_proj",
904
+ "model.layers.76.self_attn.q_proj",
905
+ "model.layers.76.self_attn.k_proj",
906
+ "model.layers.76.self_attn.v_proj",
907
+ "model.layers.76.self_attn.o_proj",
908
+ "model.layers.76.mlp.gate_proj",
909
+ "model.layers.76.mlp.up_proj",
910
+ "model.layers.76.mlp.down_proj",
911
+ "model.layers.77.self_attn.q_proj",
912
+ "model.layers.77.self_attn.k_proj",
913
+ "model.layers.77.self_attn.v_proj",
914
+ "model.layers.77.self_attn.o_proj",
915
+ "model.layers.77.mlp.gate_proj",
916
+ "model.layers.77.mlp.up_proj",
917
+ "model.layers.77.mlp.down_proj",
918
+ "model.layers.78.self_attn.q_proj",
919
+ "model.layers.78.self_attn.k_proj",
920
+ "model.layers.78.self_attn.v_proj",
921
+ "model.layers.78.self_attn.o_proj",
922
+ "model.layers.78.mlp.gate_proj",
923
+ "model.layers.78.mlp.up_proj",
924
+ "model.layers.78.mlp.down_proj",
925
+ "model.layers.79.self_attn.q_proj",
926
+ "model.layers.79.self_attn.k_proj",
927
+ "model.layers.79.self_attn.v_proj",
928
+ "model.layers.79.self_attn.o_proj",
929
+ "model.layers.79.mlp.gate_proj",
930
+ "model.layers.79.mlp.up_proj",
931
+ "model.layers.79.mlp.down_proj",
932
+ "lm_head"
933
+ ],
934
+ "registry_requires_subclass": false,
935
+ "sparsity_structure": "unstructured",
936
+ "targets": [
937
+ "model.layers.1.self_attn.q_proj",
938
+ "model.layers.1.self_attn.k_proj",
939
+ "model.layers.2.self_attn.q_proj",
940
+ "model.layers.2.self_attn.k_proj",
941
+ "model.layers.3.self_attn.q_proj",
942
+ "model.layers.5.self_attn.q_proj"
943
+ ]
944
+ }
945
+ },
946
+ "rms_norm_eps": 1e-06,
947
+ "rope_scaling": {
948
+ "mrope_section": [
949
+ 16,
950
+ 24,
951
+ 24
952
+ ],
953
+ "rope_type": "default",
954
+ "type": "default"
955
+ },
956
+ "rope_theta": 1000000.0,
957
+ "sliding_window": 32768,
958
+ "tie_word_embeddings": false,
959
+ "torch_dtype": "bfloat16",
960
+ "transformers_version": "4.49.0.dev0",
961
+ "use_cache": true,
962
+ "use_sliding_window": false,
963
+ "video_token_id": 151656,
964
+ "vision_config": {
965
+ "hidden_size": 1280,
966
+ "in_chans": 3,
967
+ "intermediate_size": 3456,
968
+ "model_type": "qwen2_5_vl",
969
+ "out_hidden_size": 8192,
970
+ "spatial_patch_size": 14,
971
+ "tokens_per_second": 2
972
+ },
973
+ "vision_end_token_id": 151653,
974
+ "vision_start_token_id": 151652,
975
+ "vision_token_id": 151654,
976
+ "vocab_size": 152064
977
+ }
generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "pad_token_id": 151643,
9
+ "repetition_penalty": 1.05,
10
+ "top_k": 1,
11
+ "top_p": 0.001,
12
+ "transformers_version": "4.49.0.dev0"
13
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model-00001-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65ae6205182adb03619a5a259938f7ccc7e0de841d692c9ff43ed0b25d056e4d
3
+ size 4883683112
model-00002-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c59e382470f47fa6f49e97fffda25ef8d4030e509222e9b45dc46db976820fba
3
+ size 4978586528
model-00003-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fffadac0b79d5fcbda277a65228d1f3a8477d0e64d598aa81779520c7d8b6f0
3
+ size 4978586728
model-00004-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4347cc3cd726cc58348ffa4219936f57f4ef41e2ac032d7aa287a48f54295785
3
+ size 4978586728
model-00005-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2c6237c2d3dab8dc5231e1753cb48e524371cb16465fe16b50d0e2eedb35f9c7
3
+ size 4978586728
model-00006-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e4a9d890927c5c4dae6a19f3b759dc98a8543d58aafaf50021d29c0f6d7377f
3
+ size 4978586728
model-00007-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7498e9424272f765a649fde0e937779b232bddafc41ee3677930460d083896ae
3
+ size 4978586728
model-00008-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:687fcd6bf5a296efe10c8eab1d9141724aa2c71f567abe1967e1f022e8f37eb9
3
+ size 4978586728
model-00009-of-00009.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79becbef2f38683dc0340ed4da922bf41b76ef1171204d191613469b6e52386e
3
+ size 2866152816
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
preprocessor_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": true,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.48145466,
8
+ 0.4578275,
9
+ 0.40821073
10
+ ],
11
+ "image_processor_type": "Qwen2_5_VLImageProcessor",
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,
20
+ "patch_size": 14,
21
+ "processor_class": "Qwen2_5_VLProcessor",
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "longest_edge": 12845056,
26
+ "shortest_edge": 3136
27
+ },
28
+ "temporal_patch_size": 2
29
+ }
recipe.yaml ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ DEFAULT_stage:
2
+ DEFAULT_modifiers:
3
+ GPTQModifier:
4
+ sequential_targets: [Qwen2_5_VLDecoderLayer]
5
+ dampening_frac: 0.03
6
+ scheme: W4A16
7
+ targets: Linear
8
+ ignore: [lm_head, 're:visual.*']
special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:913950e4971737031da511cdd1b410daae4566f62eb845b3975bca5a102323d8
3
+ size 11421995
tokenizer_config.json ADDED
@@ -0,0 +1,209 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ }
181
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
199
+ "clean_up_tokenization_spaces": false,
200
+ "eos_token": "<|im_end|>",
201
+ "errors": "replace",
202
+ "extra_special_tokens": {},
203
+ "model_max_length": 131072,
204
+ "pad_token": "<|endoftext|>",
205
+ "processor_class": "Qwen2_5_VLProcessor",
206
+ "split_special_tokens": false,
207
+ "tokenizer_class": "Qwen2Tokenizer",
208
+ "unk_token": null
209
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff