Hugging Face's logo

mlx-community
/
Lingshu-32B-bf16

Image-Text-to-Text
Transformers
Safetensors
MLX
qwen2_5_vl
image-to-text
medical
multimodal
report generation
radiology
clinical-reasoning
MRI
CT
Histopathology
X-ray
Fundus
conversational
text-generation-inference
Model card Files
xet
Community
warshanks commited on
Commit
127daaf
·
verified ·
1 Parent(s): 812db48

Upload folder using huggingface_hub

Browse files
Files changed (29) hide show
  1. .gitattributes +1 -0
  2. README.md +30 -0
  3. added_tokens.json +24 -0
  4. chat_template.jinja +7 -0
  5. chat_template.json +3 -0
  6. config.json +273 -0
  7. generation_config.json +12 -0
  8. merges.txt +0 -0
  9. model-00001-of-00013.safetensors +3 -0
  10. model-00002-of-00013.safetensors +3 -0
  11. model-00003-of-00013.safetensors +3 -0
  12. model-00004-of-00013.safetensors +3 -0
  13. model-00005-of-00013.safetensors +3 -0
  14. model-00006-of-00013.safetensors +3 -0
  15. model-00007-of-00013.safetensors +3 -0
  16. model-00008-of-00013.safetensors +3 -0
  17. model-00009-of-00013.safetensors +3 -0
  18. model-00010-of-00013.safetensors +3 -0
  19. model-00011-of-00013.safetensors +3 -0
  20. model-00012-of-00013.safetensors +3 -0
  21. model-00013-of-00013.safetensors +3 -0
  22. model.safetensors.index.json +1168 -0
  23. preprocessor_config.json +29 -0
  24. special_tokens_map.json +31 -0
  25. tokenizer.json +3 -0
  26. tokenizer_config.json +209 -0
  27. video_preprocessor_config.json +86 -0
  28. vocab.json +0 -0
  29. zero_to_fp32.py +674 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ library_name: transformers
4
+ pipeline_tag: image-text-to-text
5
+ tags:
6
+ - medical
7
+ - multimodal
8
+ - report generation
9
+ - radiology
10
+ - clinical-reasoning
11
+ - MRI
12
+ - CT
13
+ - Histopathology
14
+ - X-ray
15
+ - Fundus
16
+ - mlx
17
+ ---
18
+
19
+ # mlx-community/Lingshu-32B-bf16
20
+ This model was converted to MLX format from [`lingshu-medical-mllm/Lingshu-32B`]() using mlx-vlm version **0.1.27**.
21
+ Refer to the [original model card](https://huggingface.co/lingshu-medical-mllm/Lingshu-32B) for more details on the model.
22
+ ## Use with mlx
23
+
24
+ ```bash
25
+ pip install -U mlx-vlm
26
+ ```
27
+
28
+ ```bash
29
+ python -m mlx_vlm.generate --model mlx-community/Lingshu-32B-bf16 --max-tokens 100 --temperature 0.0 --prompt "Describe this image." --image <path_to_image>
30
+ ```
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.jinja ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {% 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
2
+ You are a helpful assistant.<|im_end|>
3
+ {% endif %}<|im_start|>{{ message['role'] }}
4
+ {% if message['content'] is string %}{{ message['content'] }}<|im_end|>
5
+ {% 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|>
6
+ {% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant
7
+ {% endif %}
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 +1,273 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "Qwen2_5_VLForConditionalGeneration"
5
+ ],
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,
13
+ "diversity_penalty": 0.0,
14
+ "do_sample": false,
15
+ "early_stopping": false,
16
+ "encoder_no_repeat_ngram_size": 0,
17
+ "eos_token_id": 151645,
18
+ "exponential_decay_length_penalty": null,
19
+ "finetuning_task": null,
20
+ "forced_bos_token_id": null,
21
+ "forced_eos_token_id": null,
22
+ "hidden_act": "silu",
23
+ "hidden_size": 5120,
24
+ "id2label": {
25
+ "0": "LABEL_0",
26
+ "1": "LABEL_1"
27
+ },
28
+ "image_token_id": 151655,
29
+ "initializer_range": 0.02,
30
+ "intermediate_size": 27648,
31
+ "is_decoder": false,
32
+ "is_encoder_decoder": false,
33
+ "label2id": {
34
+ "LABEL_0": 0,
35
+ "LABEL_1": 1
36
+ },
37
+ "length_penalty": 1.0,
38
+ "max_length": 20,
39
+ "max_position_embeddings": 128000,
40
+ "max_window_layers": 64,
41
+ "min_length": 0,
42
+ "model_type": "qwen2_5_vl",
43
+ "no_repeat_ngram_size": 0,
44
+ "num_attention_heads": 40,
45
+ "num_beam_groups": 1,
46
+ "num_beams": 1,
47
+ "num_hidden_layers": 64,
48
+ "num_key_value_heads": 8,
49
+ "num_return_sequences": 1,
50
+ "output_attentions": false,
51
+ "output_hidden_states": false,
52
+ "output_scores": false,
53
+ "pad_token_id": 151643,
54
+ "prefix": null,
55
+ "problem_type": null,
56
+ "pruned_heads": {},
57
+ "remove_invalid_values": false,
58
+ "repetition_penalty": 1.0,
59
+ "return_dict": true,
60
+ "return_dict_in_generate": false,
61
+ "rms_norm_eps": 1e-06,
62
+ "rope_scaling": {
63
+ "mrope_section": [
64
+ 16,
65
+ 24,
66
+ 24
67
+ ],
68
+ "rope_type": "default",
69
+ "type": "default"
70
+ },
71
+ "rope_theta": 1000000.0,
72
+ "sep_token_id": null,
73
+ "sliding_window": 32768,
74
+ "suppress_tokens": null,
75
+ "task_specific_params": null,
76
+ "temperature": 1.0,
77
+ "text_config": {
78
+ "vocab_size": 152064,
79
+ "max_position_embeddings": 128000,
80
+ "hidden_size": 5120,
81
+ "intermediate_size": 27648,
82
+ "num_hidden_layers": 64,
83
+ "num_attention_heads": 40,
84
+ "use_sliding_window": false,
85
+ "sliding_window": 32768,
86
+ "max_window_layers": 64,
87
+ "num_key_value_heads": 8,
88
+ "hidden_act": "silu",
89
+ "initializer_range": 0.02,
90
+ "rms_norm_eps": 1e-06,
91
+ "use_cache": false,
92
+ "rope_theta": 1000000.0,
93
+ "attention_dropout": 0.0,
94
+ "rope_scaling": {
95
+ "mrope_section": [
96
+ 16,
97
+ 24,
98
+ 24
99
+ ],
100
+ "rope_type": "default",
101
+ "type": "default"
102
+ },
103
+ "image_token_id": null,
104
+ "video_token_id": null,
105
+ "return_dict": true,
106
+ "output_hidden_states": false,
107
+ "output_attentions": false,
108
+ "torchscript": false,
109
+ "torch_dtype": "bfloat16",
110
+ "use_bfloat16": false,
111
+ "tf_legacy_loss": false,
112
+ "pruned_heads": {},
113
+ "tie_word_embeddings": false,
114
+ "chunk_size_feed_forward": 0,
115
+ "is_encoder_decoder": false,
116
+ "is_decoder": false,
117
+ "cross_attention_hidden_size": null,
118
+ "add_cross_attention": false,
119
+ "tie_encoder_decoder": false,
120
+ "max_length": 20,
121
+ "min_length": 0,
122
+ "do_sample": false,
123
+ "early_stopping": false,
124
+ "num_beams": 1,
125
+ "num_beam_groups": 1,
126
+ "diversity_penalty": 0.0,
127
+ "temperature": 1.0,
128
+ "top_k": 50,
129
+ "top_p": 1.0,
130
+ "typical_p": 1.0,
131
+ "repetition_penalty": 1.0,
132
+ "length_penalty": 1.0,
133
+ "no_repeat_ngram_size": 0,
134
+ "encoder_no_repeat_ngram_size": 0,
135
+ "bad_words_ids": null,
136
+ "num_return_sequences": 1,
137
+ "output_scores": false,
138
+ "return_dict_in_generate": false,
139
+ "forced_bos_token_id": null,
140
+ "forced_eos_token_id": null,
141
+ "remove_invalid_values": false,
142
+ "exponential_decay_length_penalty": null,
143
+ "suppress_tokens": null,
144
+ "begin_suppress_tokens": null,
145
+ "architectures": [
146
+ "Qwen2_5_VLForConditionalGeneration"
147
+ ],
148
+ "finetuning_task": null,
149
+ "id2label": {
150
+ "0": "LABEL_0",
151
+ "1": "LABEL_1"
152
+ },
153
+ "label2id": {
154
+ "LABEL_0": 0,
155
+ "LABEL_1": 1
156
+ },
157
+ "tokenizer_class": null,
158
+ "prefix": null,
159
+ "bos_token_id": null,
160
+ "pad_token_id": 151643,
161
+ "eos_token_id": 151645,
162
+ "sep_token_id": null,
163
+ "decoder_start_token_id": null,
164
+ "task_specific_params": null,
165
+ "problem_type": null,
166
+ "_name_or_path": "",
167
+ "model_type": "qwen2_5_vl_text",
168
+ "vision_end_token_id": 151653,
169
+ "vision_start_token_id": 151652,
170
+ "vision_token_id": 151654
171
+ },
172
+ "tf_legacy_loss": false,
173
+ "tie_encoder_decoder": false,
174
+ "tie_word_embeddings": false,
175
+ "tokenizer_class": null,
176
+ "top_k": 50,
177
+ "top_p": 1.0,
178
+ "torchscript": false,
179
+ "transformers_version": "4.52.4",
180
+ "typical_p": 1.0,
181
+ "use_bfloat16": false,
182
+ "use_cache": false,
183
+ "use_sliding_window": false,
184
+ "video_token_id": 151656,
185
+ "vision_config": {
186
+ "return_dict": true,
187
+ "output_hidden_states": false,
188
+ "output_attentions": false,
189
+ "torchscript": false,
190
+ "torch_dtype": "float32",
191
+ "use_bfloat16": false,
192
+ "tf_legacy_loss": false,
193
+ "pruned_heads": {},
194
+ "tie_word_embeddings": true,
195
+ "chunk_size_feed_forward": 0,
196
+ "is_encoder_decoder": false,
197
+ "is_decoder": false,
198
+ "cross_attention_hidden_size": null,
199
+ "add_cross_attention": false,
200
+ "tie_encoder_decoder": false,
201
+ "max_length": 20,
202
+ "min_length": 0,
203
+ "do_sample": false,
204
+ "early_stopping": false,
205
+ "num_beams": 1,
206
+ "num_beam_groups": 1,
207
+ "diversity_penalty": 0.0,
208
+ "temperature": 1.0,
209
+ "top_k": 50,
210
+ "top_p": 1.0,
211
+ "typical_p": 1.0,
212
+ "repetition_penalty": 1.0,
213
+ "length_penalty": 1.0,
214
+ "no_repeat_ngram_size": 0,
215
+ "encoder_no_repeat_ngram_size": 0,
216
+ "bad_words_ids": null,
217
+ "num_return_sequences": 1,
218
+ "output_scores": false,
219
+ "return_dict_in_generate": false,
220
+ "forced_bos_token_id": null,
221
+ "forced_eos_token_id": null,
222
+ "remove_invalid_values": false,
223
+ "exponential_decay_length_penalty": null,
224
+ "suppress_tokens": null,
225
+ "begin_suppress_tokens": null,
226
+ "architectures": null,
227
+ "finetuning_task": null,
228
+ "id2label": {
229
+ "0": "LABEL_0",
230
+ "1": "LABEL_1"
231
+ },
232
+ "label2id": {
233
+ "LABEL_0": 0,
234
+ "LABEL_1": 1
235
+ },
236
+ "tokenizer_class": null,
237
+ "prefix": null,
238
+ "bos_token_id": null,
239
+ "pad_token_id": null,
240
+ "eos_token_id": null,
241
+ "sep_token_id": null,
242
+ "decoder_start_token_id": null,
243
+ "task_specific_params": null,
244
+ "problem_type": null,
245
+ "_name_or_path": "",
246
+ "in_chans": 3,
247
+ "model_type": "qwen2_5_vl",
248
+ "spatial_patch_size": 14,
249
+ "depth": 32,
250
+ "hidden_size": 1280,
251
+ "hidden_act": "silu",
252
+ "intermediate_size": 3456,
253
+ "num_heads": 16,
254
+ "in_channels": 3,
255
+ "patch_size": 14,
256
+ "spatial_merge_size": 2,
257
+ "temporal_patch_size": 2,
258
+ "tokens_per_second": 2,
259
+ "window_size": 112,
260
+ "fullatt_block_indexes": [
261
+ 7,
262
+ 15,
263
+ 23,
264
+ 31
265
+ ],
266
+ "out_hidden_size": 5120,
267
+ "initializer_range": 0.02
268
+ },
269
+ "vision_end_token_id": 151653,
270
+ "vision_start_token_id": 151652,
271
+ "vision_token_id": 151654,
272
+ "vocab_size": 152064
273
+ }
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "temperature": 1e-06,
11
+ "transformers_version": "4.51.3"
12
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model-00001-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2747150edbae148e50761a9591131ff122f9ee10d6b66e7d79f0f6e1d90bb762
3
+ size 5294248257
model-00002-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0af16f208df07c1e61556efd0a96c94ddcfab54ec6468430212bbb095e41792
3
+ size 5159175904
model-00003-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b38fad0b5e078d21800c024a73675739bf5d6c09876032ae8577f9352c144ea7
3
+ size 5285041023
model-00004-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5dd284f2fdb2dd14f36de6193ede79cdd16934767985fa2eb0bdbbed7b54e305
3
+ size 5159175955
model-00005-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e841f1f82786332e8bfb481eec8eac3f4797ea3e4058363e647a936f45aaff1a
3
+ size 5159175943
model-00006-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07e3a1ae5a86376a7655f068d871dcc6a57212b1c761ce3fbc6c03efdf4dad34
3
+ size 5285041044
model-00007-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97a4881979f5338aad0d41d2d54cfa54c50f20692b452fb9ffed6dc40c084cac
3
+ size 5159175935
model-00008-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98ef042b8d194fd84a833cfb7eb4faf31af71d6ae15702f3048c2823a1977dde
3
+ size 5159175959
model-00009-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e264528573c8a650d21573369f22d54588bfd32d9cbe9b13e3739309bd52100d
3
+ size 5285041062
model-00010-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aaa43a79591ec312877b0677959c9c63ab6e6415d05150f52f88f4748198b3f2
3
+ size 5159175931
model-00011-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5c4633b4164926f18d7e8d593727415b51b5c79c443aba3370a05ea783185b4
3
+ size 5159175919
model-00012-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e37b9d17641e61a2a01c77bc9f7de5e1452ba1c64252b8586fa27b05b90a38b6
3
+ size 5285041052
model-00013-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1ec5ef11afc6cd852b936d36a4c301bffb50476b67d881d98d7065f5c51ef6b
3
+ size 4356937662
model.safetensors.index.json ADDED
@@ -0,0 +1,1168 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 66905436672
4
+ },
5
+ "weight_map": {
6
+ "lm_head.weight": "model-00014-of-00014.safetensors",
7
+ "model.embed_tokens.weight": "model-00001-of-00014.safetensors",
8
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00014.safetensors",
9
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
10
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
11
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
12
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00014.safetensors",
13
+ "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00014.safetensors",
14
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00014.safetensors",
15
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00014.safetensors",
16
+ "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00014.safetensors",
17
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00014.safetensors",
18
+ "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00014.safetensors",
19
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00014.safetensors",
20
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00014.safetensors",
21
+ "model.layers.1.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
22
+ "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
23
+ "model.layers.1.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
24
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00014.safetensors",
25
+ "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00014.safetensors",
26
+ "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00014.safetensors",
27
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00014.safetensors",
28
+ "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00014.safetensors",
29
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00014.safetensors",
30
+ "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00014.safetensors",
31
+ "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00014.safetensors",
32
+ "model.layers.10.input_layernorm.weight": "model-00003-of-00014.safetensors",
33
+ "model.layers.10.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
34
+ "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
35
+ "model.layers.10.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
36
+ "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
37
+ "model.layers.10.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
38
+ "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
39
+ "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
40
+ "model.layers.10.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
41
+ "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
42
+ "model.layers.10.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
43
+ "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
44
+ "model.layers.11.input_layernorm.weight": "model-00003-of-00014.safetensors",
45
+ "model.layers.11.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
46
+ "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
47
+ "model.layers.11.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
48
+ "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
49
+ "model.layers.11.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
50
+ "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
51
+ "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
52
+ "model.layers.11.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
53
+ "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
54
+ "model.layers.11.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
55
+ "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
56
+ "model.layers.12.input_layernorm.weight": "model-00004-of-00014.safetensors",
57
+ "model.layers.12.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
58
+ "model.layers.12.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
59
+ "model.layers.12.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
60
+ "model.layers.12.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
61
+ "model.layers.12.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
62
+ "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
63
+ "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
64
+ "model.layers.12.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
65
+ "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
66
+ "model.layers.12.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
67
+ "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
68
+ "model.layers.13.input_layernorm.weight": "model-00004-of-00014.safetensors",
69
+ "model.layers.13.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
70
+ "model.layers.13.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
71
+ "model.layers.13.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
72
+ "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
73
+ "model.layers.13.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
74
+ "model.layers.13.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
75
+ "model.layers.13.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
76
+ "model.layers.13.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
77
+ "model.layers.13.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
78
+ "model.layers.13.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
79
+ "model.layers.13.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
80
+ "model.layers.14.input_layernorm.weight": "model-00004-of-00014.safetensors",
81
+ "model.layers.14.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
82
+ "model.layers.14.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
83
+ "model.layers.14.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
84
+ "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
85
+ "model.layers.14.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
86
+ "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
87
+ "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
88
+ "model.layers.14.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
89
+ "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
90
+ "model.layers.14.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
91
+ "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
92
+ "model.layers.15.input_layernorm.weight": "model-00004-of-00014.safetensors",
93
+ "model.layers.15.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
94
+ "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
95
+ "model.layers.15.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
96
+ "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
97
+ "model.layers.15.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
98
+ "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
99
+ "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
100
+ "model.layers.15.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
101
+ "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
102
+ "model.layers.15.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
103
+ "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
104
+ "model.layers.16.input_layernorm.weight": "model-00004-of-00014.safetensors",
105
+ "model.layers.16.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
106
+ "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
107
+ "model.layers.16.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
108
+ "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
109
+ "model.layers.16.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
110
+ "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
111
+ "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
112
+ "model.layers.16.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
113
+ "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
114
+ "model.layers.16.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
115
+ "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
116
+ "model.layers.17.input_layernorm.weight": "model-00005-of-00014.safetensors",
117
+ "model.layers.17.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
118
+ "model.layers.17.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
119
+ "model.layers.17.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
120
+ "model.layers.17.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
121
+ "model.layers.17.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
122
+ "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
123
+ "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
124
+ "model.layers.17.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
125
+ "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
126
+ "model.layers.17.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
127
+ "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
128
+ "model.layers.18.input_layernorm.weight": "model-00005-of-00014.safetensors",
129
+ "model.layers.18.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
130
+ "model.layers.18.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
131
+ "model.layers.18.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
132
+ "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
133
+ "model.layers.18.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
134
+ "model.layers.18.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
135
+ "model.layers.18.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
136
+ "model.layers.18.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
137
+ "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
138
+ "model.layers.18.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
139
+ "model.layers.18.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
140
+ "model.layers.19.input_layernorm.weight": "model-00005-of-00014.safetensors",
141
+ "model.layers.19.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
142
+ "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
143
+ "model.layers.19.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
144
+ "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
145
+ "model.layers.19.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
146
+ "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
147
+ "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
148
+ "model.layers.19.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
149
+ "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
150
+ "model.layers.19.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
151
+ "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
152
+ "model.layers.2.input_layernorm.weight": "model-00002-of-00014.safetensors",
153
+ "model.layers.2.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
154
+ "model.layers.2.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
155
+ "model.layers.2.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
156
+ "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
157
+ "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00014.safetensors",
158
+ "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00014.safetensors",
159
+ "model.layers.2.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
160
+ "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00014.safetensors",
161
+ "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00014.safetensors",
162
+ "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00014.safetensors",
163
+ "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00014.safetensors",
164
+ "model.layers.20.input_layernorm.weight": "model-00005-of-00014.safetensors",
165
+ "model.layers.20.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
166
+ "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
167
+ "model.layers.20.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
168
+ "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
169
+ "model.layers.20.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
170
+ "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
171
+ "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
172
+ "model.layers.20.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
173
+ "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
174
+ "model.layers.20.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
175
+ "model.layers.20.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
176
+ "model.layers.21.input_layernorm.weight": "model-00005-of-00014.safetensors",
177
+ "model.layers.21.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
178
+ "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
179
+ "model.layers.21.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
180
+ "model.layers.21.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
181
+ "model.layers.21.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
182
+ "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
183
+ "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
184
+ "model.layers.21.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
185
+ "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
186
+ "model.layers.21.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
187
+ "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
188
+ "model.layers.22.input_layernorm.weight": "model-00006-of-00014.safetensors",
189
+ "model.layers.22.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
190
+ "model.layers.22.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
191
+ "model.layers.22.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
192
+ "model.layers.22.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
193
+ "model.layers.22.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
194
+ "model.layers.22.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
195
+ "model.layers.22.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
196
+ "model.layers.22.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
197
+ "model.layers.22.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
198
+ "model.layers.22.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
199
+ "model.layers.22.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
200
+ "model.layers.23.input_layernorm.weight": "model-00006-of-00014.safetensors",
201
+ "model.layers.23.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
202
+ "model.layers.23.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
203
+ "model.layers.23.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
204
+ "model.layers.23.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
205
+ "model.layers.23.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
206
+ "model.layers.23.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
207
+ "model.layers.23.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
208
+ "model.layers.23.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
209
+ "model.layers.23.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
210
+ "model.layers.23.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
211
+ "model.layers.23.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
212
+ "model.layers.24.input_layernorm.weight": "model-00006-of-00014.safetensors",
213
+ "model.layers.24.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
214
+ "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
215
+ "model.layers.24.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
216
+ "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
217
+ "model.layers.24.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
218
+ "model.layers.24.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
219
+ "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
220
+ "model.layers.24.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
221
+ "model.layers.24.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
222
+ "model.layers.24.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
223
+ "model.layers.24.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
224
+ "model.layers.25.input_layernorm.weight": "model-00006-of-00014.safetensors",
225
+ "model.layers.25.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
226
+ "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
227
+ "model.layers.25.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
228
+ "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
229
+ "model.layers.25.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
230
+ "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
231
+ "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
232
+ "model.layers.25.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
233
+ "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
234
+ "model.layers.25.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
235
+ "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
236
+ "model.layers.26.input_layernorm.weight": "model-00006-of-00014.safetensors",
237
+ "model.layers.26.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
238
+ "model.layers.26.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
239
+ "model.layers.26.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
240
+ "model.layers.26.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
241
+ "model.layers.26.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
242
+ "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
243
+ "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
244
+ "model.layers.26.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
245
+ "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
246
+ "model.layers.26.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
247
+ "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
248
+ "model.layers.27.input_layernorm.weight": "model-00007-of-00014.safetensors",
249
+ "model.layers.27.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
250
+ "model.layers.27.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
251
+ "model.layers.27.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
252
+ "model.layers.27.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
253
+ "model.layers.27.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
254
+ "model.layers.27.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
255
+ "model.layers.27.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
256
+ "model.layers.27.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
257
+ "model.layers.27.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
258
+ "model.layers.27.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
259
+ "model.layers.27.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
260
+ "model.layers.28.input_layernorm.weight": "model-00007-of-00014.safetensors",
261
+ "model.layers.28.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
262
+ "model.layers.28.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
263
+ "model.layers.28.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
264
+ "model.layers.28.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
265
+ "model.layers.28.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
266
+ "model.layers.28.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
267
+ "model.layers.28.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
268
+ "model.layers.28.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
269
+ "model.layers.28.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
270
+ "model.layers.28.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
271
+ "model.layers.28.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
272
+ "model.layers.29.input_layernorm.weight": "model-00007-of-00014.safetensors",
273
+ "model.layers.29.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
274
+ "model.layers.29.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
275
+ "model.layers.29.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
276
+ "model.layers.29.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
277
+ "model.layers.29.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
278
+ "model.layers.29.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
279
+ "model.layers.29.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
280
+ "model.layers.29.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
281
+ "model.layers.29.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
282
+ "model.layers.29.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
283
+ "model.layers.29.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
284
+ "model.layers.3.input_layernorm.weight": "model-00002-of-00014.safetensors",
285
+ "model.layers.3.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
286
+ "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
287
+ "model.layers.3.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
288
+ "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
289
+ "model.layers.3.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
290
+ "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
291
+ "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
292
+ "model.layers.3.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
293
+ "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
294
+ "model.layers.3.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
295
+ "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
296
+ "model.layers.30.input_layernorm.weight": "model-00007-of-00014.safetensors",
297
+ "model.layers.30.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
298
+ "model.layers.30.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
299
+ "model.layers.30.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
300
+ "model.layers.30.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
301
+ "model.layers.30.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
302
+ "model.layers.30.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
303
+ "model.layers.30.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
304
+ "model.layers.30.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
305
+ "model.layers.30.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
306
+ "model.layers.30.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
307
+ "model.layers.30.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
308
+ "model.layers.31.input_layernorm.weight": "model-00007-of-00014.safetensors",
309
+ "model.layers.31.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
310
+ "model.layers.31.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
311
+ "model.layers.31.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
312
+ "model.layers.31.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
313
+ "model.layers.31.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
314
+ "model.layers.31.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
315
+ "model.layers.31.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
316
+ "model.layers.31.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
317
+ "model.layers.31.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
318
+ "model.layers.31.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
319
+ "model.layers.31.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
320
+ "model.layers.32.input_layernorm.weight": "model-00008-of-00014.safetensors",
321
+ "model.layers.32.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
322
+ "model.layers.32.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
323
+ "model.layers.32.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
324
+ "model.layers.32.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
325
+ "model.layers.32.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
326
+ "model.layers.32.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
327
+ "model.layers.32.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
328
+ "model.layers.32.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
329
+ "model.layers.32.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
330
+ "model.layers.32.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
331
+ "model.layers.32.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
332
+ "model.layers.33.input_layernorm.weight": "model-00008-of-00014.safetensors",
333
+ "model.layers.33.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
334
+ "model.layers.33.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
335
+ "model.layers.33.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
336
+ "model.layers.33.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
337
+ "model.layers.33.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
338
+ "model.layers.33.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
339
+ "model.layers.33.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
340
+ "model.layers.33.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
341
+ "model.layers.33.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
342
+ "model.layers.33.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
343
+ "model.layers.33.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
344
+ "model.layers.34.input_layernorm.weight": "model-00008-of-00014.safetensors",
345
+ "model.layers.34.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
346
+ "model.layers.34.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
347
+ "model.layers.34.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
348
+ "model.layers.34.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
349
+ "model.layers.34.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
350
+ "model.layers.34.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
351
+ "model.layers.34.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
352
+ "model.layers.34.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
353
+ "model.layers.34.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
354
+ "model.layers.34.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
355
+ "model.layers.34.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
356
+ "model.layers.35.input_layernorm.weight": "model-00008-of-00014.safetensors",
357
+ "model.layers.35.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
358
+ "model.layers.35.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
359
+ "model.layers.35.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
360
+ "model.layers.35.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
361
+ "model.layers.35.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
362
+ "model.layers.35.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
363
+ "model.layers.35.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
364
+ "model.layers.35.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
365
+ "model.layers.35.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
366
+ "model.layers.35.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
367
+ "model.layers.35.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
368
+ "model.layers.36.input_layernorm.weight": "model-00008-of-00014.safetensors",
369
+ "model.layers.36.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
370
+ "model.layers.36.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
371
+ "model.layers.36.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
372
+ "model.layers.36.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
373
+ "model.layers.36.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
374
+ "model.layers.36.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
375
+ "model.layers.36.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
376
+ "model.layers.36.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
377
+ "model.layers.36.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
378
+ "model.layers.36.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
379
+ "model.layers.36.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
380
+ "model.layers.37.input_layernorm.weight": "model-00009-of-00014.safetensors",
381
+ "model.layers.37.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
382
+ "model.layers.37.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
383
+ "model.layers.37.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
384
+ "model.layers.37.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
385
+ "model.layers.37.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
386
+ "model.layers.37.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
387
+ "model.layers.37.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
388
+ "model.layers.37.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
389
+ "model.layers.37.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
390
+ "model.layers.37.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
391
+ "model.layers.37.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
392
+ "model.layers.38.input_layernorm.weight": "model-00009-of-00014.safetensors",
393
+ "model.layers.38.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
394
+ "model.layers.38.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
395
+ "model.layers.38.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
396
+ "model.layers.38.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
397
+ "model.layers.38.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
398
+ "model.layers.38.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
399
+ "model.layers.38.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
400
+ "model.layers.38.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
401
+ "model.layers.38.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
402
+ "model.layers.38.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
403
+ "model.layers.38.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
404
+ "model.layers.39.input_layernorm.weight": "model-00009-of-00014.safetensors",
405
+ "model.layers.39.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
406
+ "model.layers.39.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
407
+ "model.layers.39.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
408
+ "model.layers.39.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
409
+ "model.layers.39.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
410
+ "model.layers.39.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
411
+ "model.layers.39.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
412
+ "model.layers.39.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
413
+ "model.layers.39.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
414
+ "model.layers.39.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
415
+ "model.layers.39.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
416
+ "model.layers.4.input_layernorm.weight": "model-00002-of-00014.safetensors",
417
+ "model.layers.4.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
418
+ "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
419
+ "model.layers.4.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
420
+ "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
421
+ "model.layers.4.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
422
+ "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
423
+ "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
424
+ "model.layers.4.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
425
+ "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
426
+ "model.layers.4.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
427
+ "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
428
+ "model.layers.40.input_layernorm.weight": "model-00009-of-00014.safetensors",
429
+ "model.layers.40.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
430
+ "model.layers.40.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
431
+ "model.layers.40.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
432
+ "model.layers.40.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
433
+ "model.layers.40.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
434
+ "model.layers.40.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
435
+ "model.layers.40.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
436
+ "model.layers.40.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
437
+ "model.layers.40.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
438
+ "model.layers.40.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
439
+ "model.layers.40.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
440
+ "model.layers.41.input_layernorm.weight": "model-00009-of-00014.safetensors",
441
+ "model.layers.41.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
442
+ "model.layers.41.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
443
+ "model.layers.41.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
444
+ "model.layers.41.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
445
+ "model.layers.41.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
446
+ "model.layers.41.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
447
+ "model.layers.41.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
448
+ "model.layers.41.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
449
+ "model.layers.41.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
450
+ "model.layers.41.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
451
+ "model.layers.41.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
452
+ "model.layers.42.input_layernorm.weight": "model-00010-of-00014.safetensors",
453
+ "model.layers.42.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
454
+ "model.layers.42.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
455
+ "model.layers.42.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
456
+ "model.layers.42.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
457
+ "model.layers.42.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
458
+ "model.layers.42.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
459
+ "model.layers.42.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
460
+ "model.layers.42.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
461
+ "model.layers.42.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
462
+ "model.layers.42.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
463
+ "model.layers.42.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
464
+ "model.layers.43.input_layernorm.weight": "model-00010-of-00014.safetensors",
465
+ "model.layers.43.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
466
+ "model.layers.43.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
467
+ "model.layers.43.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
468
+ "model.layers.43.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
469
+ "model.layers.43.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
470
+ "model.layers.43.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
471
+ "model.layers.43.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
472
+ "model.layers.43.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
473
+ "model.layers.43.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
474
+ "model.layers.43.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
475
+ "model.layers.43.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
476
+ "model.layers.44.input_layernorm.weight": "model-00010-of-00014.safetensors",
477
+ "model.layers.44.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
478
+ "model.layers.44.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
479
+ "model.layers.44.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
480
+ "model.layers.44.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
481
+ "model.layers.44.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
482
+ "model.layers.44.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
483
+ "model.layers.44.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
484
+ "model.layers.44.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
485
+ "model.layers.44.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
486
+ "model.layers.44.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
487
+ "model.layers.44.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
488
+ "model.layers.45.input_layernorm.weight": "model-00010-of-00014.safetensors",
489
+ "model.layers.45.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
490
+ "model.layers.45.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
491
+ "model.layers.45.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
492
+ "model.layers.45.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
493
+ "model.layers.45.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
494
+ "model.layers.45.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
495
+ "model.layers.45.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
496
+ "model.layers.45.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
497
+ "model.layers.45.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
498
+ "model.layers.45.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
499
+ "model.layers.45.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
500
+ "model.layers.46.input_layernorm.weight": "model-00010-of-00014.safetensors",
501
+ "model.layers.46.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
502
+ "model.layers.46.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
503
+ "model.layers.46.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
504
+ "model.layers.46.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
505
+ "model.layers.46.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
506
+ "model.layers.46.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
507
+ "model.layers.46.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
508
+ "model.layers.46.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
509
+ "model.layers.46.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
510
+ "model.layers.46.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
511
+ "model.layers.46.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
512
+ "model.layers.47.input_layernorm.weight": "model-00011-of-00014.safetensors",
513
+ "model.layers.47.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
514
+ "model.layers.47.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
515
+ "model.layers.47.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
516
+ "model.layers.47.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
517
+ "model.layers.47.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
518
+ "model.layers.47.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
519
+ "model.layers.47.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
520
+ "model.layers.47.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
521
+ "model.layers.47.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
522
+ "model.layers.47.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
523
+ "model.layers.47.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
524
+ "model.layers.48.input_layernorm.weight": "model-00011-of-00014.safetensors",
525
+ "model.layers.48.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
526
+ "model.layers.48.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
527
+ "model.layers.48.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
528
+ "model.layers.48.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
529
+ "model.layers.48.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
530
+ "model.layers.48.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
531
+ "model.layers.48.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
532
+ "model.layers.48.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
533
+ "model.layers.48.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
534
+ "model.layers.48.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
535
+ "model.layers.48.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
536
+ "model.layers.49.input_layernorm.weight": "model-00011-of-00014.safetensors",
537
+ "model.layers.49.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
538
+ "model.layers.49.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
539
+ "model.layers.49.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
540
+ "model.layers.49.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
541
+ "model.layers.49.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
542
+ "model.layers.49.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
543
+ "model.layers.49.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
544
+ "model.layers.49.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
545
+ "model.layers.49.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
546
+ "model.layers.49.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
547
+ "model.layers.49.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
548
+ "model.layers.5.input_layernorm.weight": "model-00002-of-00014.safetensors",
549
+ "model.layers.5.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
550
+ "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
551
+ "model.layers.5.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
552
+ "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
553
+ "model.layers.5.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
554
+ "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
555
+ "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
556
+ "model.layers.5.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
557
+ "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
558
+ "model.layers.5.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
559
+ "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
560
+ "model.layers.50.input_layernorm.weight": "model-00011-of-00014.safetensors",
561
+ "model.layers.50.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
562
+ "model.layers.50.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
563
+ "model.layers.50.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
564
+ "model.layers.50.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
565
+ "model.layers.50.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
566
+ "model.layers.50.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
567
+ "model.layers.50.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
568
+ "model.layers.50.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
569
+ "model.layers.50.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
570
+ "model.layers.50.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
571
+ "model.layers.50.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
572
+ "model.layers.51.input_layernorm.weight": "model-00011-of-00014.safetensors",
573
+ "model.layers.51.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
574
+ "model.layers.51.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
575
+ "model.layers.51.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
576
+ "model.layers.51.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
577
+ "model.layers.51.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
578
+ "model.layers.51.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
579
+ "model.layers.51.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
580
+ "model.layers.51.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
581
+ "model.layers.51.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
582
+ "model.layers.51.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
583
+ "model.layers.51.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
584
+ "model.layers.52.input_layernorm.weight": "model-00012-of-00014.safetensors",
585
+ "model.layers.52.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
586
+ "model.layers.52.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
587
+ "model.layers.52.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
588
+ "model.layers.52.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
589
+ "model.layers.52.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
590
+ "model.layers.52.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
591
+ "model.layers.52.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
592
+ "model.layers.52.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
593
+ "model.layers.52.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
594
+ "model.layers.52.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
595
+ "model.layers.52.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
596
+ "model.layers.53.input_layernorm.weight": "model-00012-of-00014.safetensors",
597
+ "model.layers.53.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
598
+ "model.layers.53.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
599
+ "model.layers.53.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
600
+ "model.layers.53.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
601
+ "model.layers.53.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
602
+ "model.layers.53.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
603
+ "model.layers.53.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
604
+ "model.layers.53.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
605
+ "model.layers.53.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
606
+ "model.layers.53.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
607
+ "model.layers.53.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
608
+ "model.layers.54.input_layernorm.weight": "model-00012-of-00014.safetensors",
609
+ "model.layers.54.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
610
+ "model.layers.54.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
611
+ "model.layers.54.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
612
+ "model.layers.54.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
613
+ "model.layers.54.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
614
+ "model.layers.54.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
615
+ "model.layers.54.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
616
+ "model.layers.54.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
617
+ "model.layers.54.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
618
+ "model.layers.54.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
619
+ "model.layers.54.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
620
+ "model.layers.55.input_layernorm.weight": "model-00012-of-00014.safetensors",
621
+ "model.layers.55.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
622
+ "model.layers.55.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
623
+ "model.layers.55.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
624
+ "model.layers.55.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
625
+ "model.layers.55.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
626
+ "model.layers.55.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
627
+ "model.layers.55.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
628
+ "model.layers.55.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
629
+ "model.layers.55.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
630
+ "model.layers.55.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
631
+ "model.layers.55.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
632
+ "model.layers.56.input_layernorm.weight": "model-00012-of-00014.safetensors",
633
+ "model.layers.56.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
634
+ "model.layers.56.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
635
+ "model.layers.56.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
636
+ "model.layers.56.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
637
+ "model.layers.56.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
638
+ "model.layers.56.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
639
+ "model.layers.56.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
640
+ "model.layers.56.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
641
+ "model.layers.56.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
642
+ "model.layers.56.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
643
+ "model.layers.56.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
644
+ "model.layers.57.input_layernorm.weight": "model-00013-of-00014.safetensors",
645
+ "model.layers.57.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
646
+ "model.layers.57.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
647
+ "model.layers.57.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
648
+ "model.layers.57.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
649
+ "model.layers.57.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
650
+ "model.layers.57.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
651
+ "model.layers.57.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
652
+ "model.layers.57.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
653
+ "model.layers.57.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
654
+ "model.layers.57.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
655
+ "model.layers.57.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
656
+ "model.layers.58.input_layernorm.weight": "model-00013-of-00014.safetensors",
657
+ "model.layers.58.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
658
+ "model.layers.58.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
659
+ "model.layers.58.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
660
+ "model.layers.58.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
661
+ "model.layers.58.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
662
+ "model.layers.58.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
663
+ "model.layers.58.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
664
+ "model.layers.58.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
665
+ "model.layers.58.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
666
+ "model.layers.58.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
667
+ "model.layers.58.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
668
+ "model.layers.59.input_layernorm.weight": "model-00013-of-00014.safetensors",
669
+ "model.layers.59.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
670
+ "model.layers.59.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
671
+ "model.layers.59.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
672
+ "model.layers.59.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
673
+ "model.layers.59.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
674
+ "model.layers.59.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
675
+ "model.layers.59.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
676
+ "model.layers.59.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
677
+ "model.layers.59.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
678
+ "model.layers.59.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
679
+ "model.layers.59.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
680
+ "model.layers.6.input_layernorm.weight": "model-00002-of-00014.safetensors",
681
+ "model.layers.6.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
682
+ "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
683
+ "model.layers.6.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
684
+ "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
685
+ "model.layers.6.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
686
+ "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
687
+ "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
688
+ "model.layers.6.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
689
+ "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
690
+ "model.layers.6.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
691
+ "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
692
+ "model.layers.60.input_layernorm.weight": "model-00013-of-00014.safetensors",
693
+ "model.layers.60.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
694
+ "model.layers.60.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
695
+ "model.layers.60.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
696
+ "model.layers.60.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
697
+ "model.layers.60.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
698
+ "model.layers.60.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
699
+ "model.layers.60.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
700
+ "model.layers.60.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
701
+ "model.layers.60.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
702
+ "model.layers.60.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
703
+ "model.layers.60.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
704
+ "model.layers.61.input_layernorm.weight": "model-00013-of-00014.safetensors",
705
+ "model.layers.61.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
706
+ "model.layers.61.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
707
+ "model.layers.61.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
708
+ "model.layers.61.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
709
+ "model.layers.61.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
710
+ "model.layers.61.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
711
+ "model.layers.61.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
712
+ "model.layers.61.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
713
+ "model.layers.61.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
714
+ "model.layers.61.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
715
+ "model.layers.61.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
716
+ "model.layers.62.input_layernorm.weight": "model-00014-of-00014.safetensors",
717
+ "model.layers.62.mlp.down_proj.weight": "model-00014-of-00014.safetensors",
718
+ "model.layers.62.mlp.gate_proj.weight": "model-00014-of-00014.safetensors",
719
+ "model.layers.62.mlp.up_proj.weight": "model-00014-of-00014.safetensors",
720
+ "model.layers.62.post_attention_layernorm.weight": "model-00014-of-00014.safetensors",
721
+ "model.layers.62.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
722
+ "model.layers.62.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
723
+ "model.layers.62.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
724
+ "model.layers.62.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
725
+ "model.layers.62.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
726
+ "model.layers.62.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
727
+ "model.layers.62.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
728
+ "model.layers.63.input_layernorm.weight": "model-00014-of-00014.safetensors",
729
+ "model.layers.63.mlp.down_proj.weight": "model-00014-of-00014.safetensors",
730
+ "model.layers.63.mlp.gate_proj.weight": "model-00014-of-00014.safetensors",
731
+ "model.layers.63.mlp.up_proj.weight": "model-00014-of-00014.safetensors",
732
+ "model.layers.63.post_attention_layernorm.weight": "model-00014-of-00014.safetensors",
733
+ "model.layers.63.self_attn.k_proj.bias": "model-00014-of-00014.safetensors",
734
+ "model.layers.63.self_attn.k_proj.weight": "model-00014-of-00014.safetensors",
735
+ "model.layers.63.self_attn.o_proj.weight": "model-00014-of-00014.safetensors",
736
+ "model.layers.63.self_attn.q_proj.bias": "model-00014-of-00014.safetensors",
737
+ "model.layers.63.self_attn.q_proj.weight": "model-00014-of-00014.safetensors",
738
+ "model.layers.63.self_attn.v_proj.bias": "model-00014-of-00014.safetensors",
739
+ "model.layers.63.self_attn.v_proj.weight": "model-00014-of-00014.safetensors",
740
+ "model.layers.7.input_layernorm.weight": "model-00003-of-00014.safetensors",
741
+ "model.layers.7.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
742
+ "model.layers.7.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
743
+ "model.layers.7.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
744
+ "model.layers.7.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
745
+ "model.layers.7.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
746
+ "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
747
+ "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
748
+ "model.layers.7.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
749
+ "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
750
+ "model.layers.7.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
751
+ "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
752
+ "model.layers.8.input_layernorm.weight": "model-00003-of-00014.safetensors",
753
+ "model.layers.8.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
754
+ "model.layers.8.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
755
+ "model.layers.8.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
756
+ "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
757
+ "model.layers.8.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
758
+ "model.layers.8.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
759
+ "model.layers.8.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
760
+ "model.layers.8.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
761
+ "model.layers.8.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
762
+ "model.layers.8.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
763
+ "model.layers.8.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
764
+ "model.layers.9.input_layernorm.weight": "model-00003-of-00014.safetensors",
765
+ "model.layers.9.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
766
+ "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
767
+ "model.layers.9.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
768
+ "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
769
+ "model.layers.9.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
770
+ "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
771
+ "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
772
+ "model.layers.9.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
773
+ "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
774
+ "model.layers.9.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
775
+ "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
776
+ "model.norm.weight": "model-00014-of-00014.safetensors",
777
+ "visual.blocks.0.attn.proj.bias": "model-00001-of-00014.safetensors",
778
+ "visual.blocks.0.attn.proj.weight": "model-00001-of-00014.safetensors",
779
+ "visual.blocks.0.attn.qkv.bias": "model-00001-of-00014.safetensors",
780
+ "visual.blocks.0.attn.qkv.weight": "model-00001-of-00014.safetensors",
781
+ "visual.blocks.0.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
782
+ "visual.blocks.0.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
783
+ "visual.blocks.0.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
784
+ "visual.blocks.0.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
785
+ "visual.blocks.0.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
786
+ "visual.blocks.0.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
787
+ "visual.blocks.0.norm1.weight": "model-00001-of-00014.safetensors",
788
+ "visual.blocks.0.norm2.weight": "model-00001-of-00014.safetensors",
789
+ "visual.blocks.1.attn.proj.bias": "model-00001-of-00014.safetensors",
790
+ "visual.blocks.1.attn.proj.weight": "model-00001-of-00014.safetensors",
791
+ "visual.blocks.1.attn.qkv.bias": "model-00001-of-00014.safetensors",
792
+ "visual.blocks.1.attn.qkv.weight": "model-00001-of-00014.safetensors",
793
+ "visual.blocks.1.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
794
+ "visual.blocks.1.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
795
+ "visual.blocks.1.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
796
+ "visual.blocks.1.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
797
+ "visual.blocks.1.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
798
+ "visual.blocks.1.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
799
+ "visual.blocks.1.norm1.weight": "model-00001-of-00014.safetensors",
800
+ "visual.blocks.1.norm2.weight": "model-00001-of-00014.safetensors",
801
+ "visual.blocks.10.attn.proj.bias": "model-00001-of-00014.safetensors",
802
+ "visual.blocks.10.attn.proj.weight": "model-00001-of-00014.safetensors",
803
+ "visual.blocks.10.attn.qkv.bias": "model-00001-of-00014.safetensors",
804
+ "visual.blocks.10.attn.qkv.weight": "model-00001-of-00014.safetensors",
805
+ "visual.blocks.10.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
806
+ "visual.blocks.10.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
807
+ "visual.blocks.10.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
808
+ "visual.blocks.10.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
809
+ "visual.blocks.10.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
810
+ "visual.blocks.10.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
811
+ "visual.blocks.10.norm1.weight": "model-00001-of-00014.safetensors",
812
+ "visual.blocks.10.norm2.weight": "model-00001-of-00014.safetensors",
813
+ "visual.blocks.11.attn.proj.bias": "model-00001-of-00014.safetensors",
814
+ "visual.blocks.11.attn.proj.weight": "model-00001-of-00014.safetensors",
815
+ "visual.blocks.11.attn.qkv.bias": "model-00001-of-00014.safetensors",
816
+ "visual.blocks.11.attn.qkv.weight": "model-00001-of-00014.safetensors",
817
+ "visual.blocks.11.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
818
+ "visual.blocks.11.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
819
+ "visual.blocks.11.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
820
+ "visual.blocks.11.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
821
+ "visual.blocks.11.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
822
+ "visual.blocks.11.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
823
+ "visual.blocks.11.norm1.weight": "model-00001-of-00014.safetensors",
824
+ "visual.blocks.11.norm2.weight": "model-00001-of-00014.safetensors",
825
+ "visual.blocks.12.attn.proj.bias": "model-00001-of-00014.safetensors",
826
+ "visual.blocks.12.attn.proj.weight": "model-00001-of-00014.safetensors",
827
+ "visual.blocks.12.attn.qkv.bias": "model-00001-of-00014.safetensors",
828
+ "visual.blocks.12.attn.qkv.weight": "model-00001-of-00014.safetensors",
829
+ "visual.blocks.12.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
830
+ "visual.blocks.12.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
831
+ "visual.blocks.12.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
832
+ "visual.blocks.12.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
833
+ "visual.blocks.12.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
834
+ "visual.blocks.12.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
835
+ "visual.blocks.12.norm1.weight": "model-00001-of-00014.safetensors",
836
+ "visual.blocks.12.norm2.weight": "model-00001-of-00014.safetensors",
837
+ "visual.blocks.13.attn.proj.bias": "model-00001-of-00014.safetensors",
838
+ "visual.blocks.13.attn.proj.weight": "model-00001-of-00014.safetensors",
839
+ "visual.blocks.13.attn.qkv.bias": "model-00001-of-00014.safetensors",
840
+ "visual.blocks.13.attn.qkv.weight": "model-00001-of-00014.safetensors",
841
+ "visual.blocks.13.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
842
+ "visual.blocks.13.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
843
+ "visual.blocks.13.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
844
+ "visual.blocks.13.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
845
+ "visual.blocks.13.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
846
+ "visual.blocks.13.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
847
+ "visual.blocks.13.norm1.weight": "model-00001-of-00014.safetensors",
848
+ "visual.blocks.13.norm2.weight": "model-00001-of-00014.safetensors",
849
+ "visual.blocks.14.attn.proj.bias": "model-00001-of-00014.safetensors",
850
+ "visual.blocks.14.attn.proj.weight": "model-00001-of-00014.safetensors",
851
+ "visual.blocks.14.attn.qkv.bias": "model-00001-of-00014.safetensors",
852
+ "visual.blocks.14.attn.qkv.weight": "model-00001-of-00014.safetensors",
853
+ "visual.blocks.14.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
854
+ "visual.blocks.14.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
855
+ "visual.blocks.14.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
856
+ "visual.blocks.14.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
857
+ "visual.blocks.14.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
858
+ "visual.blocks.14.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
859
+ "visual.blocks.14.norm1.weight": "model-00001-of-00014.safetensors",
860
+ "visual.blocks.14.norm2.weight": "model-00001-of-00014.safetensors",
861
+ "visual.blocks.15.attn.proj.bias": "model-00001-of-00014.safetensors",
862
+ "visual.blocks.15.attn.proj.weight": "model-00001-of-00014.safetensors",
863
+ "visual.blocks.15.attn.qkv.bias": "model-00001-of-00014.safetensors",
864
+ "visual.blocks.15.attn.qkv.weight": "model-00001-of-00014.safetensors",
865
+ "visual.blocks.15.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
866
+ "visual.blocks.15.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
867
+ "visual.blocks.15.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
868
+ "visual.blocks.15.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
869
+ "visual.blocks.15.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
870
+ "visual.blocks.15.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
871
+ "visual.blocks.15.norm1.weight": "model-00001-of-00014.safetensors",
872
+ "visual.blocks.15.norm2.weight": "model-00001-of-00014.safetensors",
873
+ "visual.blocks.16.attn.proj.bias": "model-00001-of-00014.safetensors",
874
+ "visual.blocks.16.attn.proj.weight": "model-00001-of-00014.safetensors",
875
+ "visual.blocks.16.attn.qkv.bias": "model-00001-of-00014.safetensors",
876
+ "visual.blocks.16.attn.qkv.weight": "model-00001-of-00014.safetensors",
877
+ "visual.blocks.16.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
878
+ "visual.blocks.16.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
879
+ "visual.blocks.16.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
880
+ "visual.blocks.16.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
881
+ "visual.blocks.16.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
882
+ "visual.blocks.16.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
883
+ "visual.blocks.16.norm1.weight": "model-00001-of-00014.safetensors",
884
+ "visual.blocks.16.norm2.weight": "model-00001-of-00014.safetensors",
885
+ "visual.blocks.17.attn.proj.bias": "model-00001-of-00014.safetensors",
886
+ "visual.blocks.17.attn.proj.weight": "model-00001-of-00014.safetensors",
887
+ "visual.blocks.17.attn.qkv.bias": "model-00001-of-00014.safetensors",
888
+ "visual.blocks.17.attn.qkv.weight": "model-00001-of-00014.safetensors",
889
+ "visual.blocks.17.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
890
+ "visual.blocks.17.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
891
+ "visual.blocks.17.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
892
+ "visual.blocks.17.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
893
+ "visual.blocks.17.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
894
+ "visual.blocks.17.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
895
+ "visual.blocks.17.norm1.weight": "model-00001-of-00014.safetensors",
896
+ "visual.blocks.17.norm2.weight": "model-00001-of-00014.safetensors",
897
+ "visual.blocks.18.attn.proj.bias": "model-00001-of-00014.safetensors",
898
+ "visual.blocks.18.attn.proj.weight": "model-00001-of-00014.safetensors",
899
+ "visual.blocks.18.attn.qkv.bias": "model-00001-of-00014.safetensors",
900
+ "visual.blocks.18.attn.qkv.weight": "model-00001-of-00014.safetensors",
901
+ "visual.blocks.18.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
902
+ "visual.blocks.18.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
903
+ "visual.blocks.18.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
904
+ "visual.blocks.18.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
905
+ "visual.blocks.18.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
906
+ "visual.blocks.18.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
907
+ "visual.blocks.18.norm1.weight": "model-00001-of-00014.safetensors",
908
+ "visual.blocks.18.norm2.weight": "model-00001-of-00014.safetensors",
909
+ "visual.blocks.19.attn.proj.bias": "model-00001-of-00014.safetensors",
910
+ "visual.blocks.19.attn.proj.weight": "model-00001-of-00014.safetensors",
911
+ "visual.blocks.19.attn.qkv.bias": "model-00001-of-00014.safetensors",
912
+ "visual.blocks.19.attn.qkv.weight": "model-00001-of-00014.safetensors",
913
+ "visual.blocks.19.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
914
+ "visual.blocks.19.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
915
+ "visual.blocks.19.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
916
+ "visual.blocks.19.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
917
+ "visual.blocks.19.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
918
+ "visual.blocks.19.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
919
+ "visual.blocks.19.norm1.weight": "model-00001-of-00014.safetensors",
920
+ "visual.blocks.19.norm2.weight": "model-00001-of-00014.safetensors",
921
+ "visual.blocks.2.attn.proj.bias": "model-00001-of-00014.safetensors",
922
+ "visual.blocks.2.attn.proj.weight": "model-00001-of-00014.safetensors",
923
+ "visual.blocks.2.attn.qkv.bias": "model-00001-of-00014.safetensors",
924
+ "visual.blocks.2.attn.qkv.weight": "model-00001-of-00014.safetensors",
925
+ "visual.blocks.2.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
926
+ "visual.blocks.2.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
927
+ "visual.blocks.2.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
928
+ "visual.blocks.2.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
929
+ "visual.blocks.2.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
930
+ "visual.blocks.2.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
931
+ "visual.blocks.2.norm1.weight": "model-00001-of-00014.safetensors",
932
+ "visual.blocks.2.norm2.weight": "model-00001-of-00014.safetensors",
933
+ "visual.blocks.20.attn.proj.bias": "model-00001-of-00014.safetensors",
934
+ "visual.blocks.20.attn.proj.weight": "model-00001-of-00014.safetensors",
935
+ "visual.blocks.20.attn.qkv.bias": "model-00001-of-00014.safetensors",
936
+ "visual.blocks.20.attn.qkv.weight": "model-00001-of-00014.safetensors",
937
+ "visual.blocks.20.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
938
+ "visual.blocks.20.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
939
+ "visual.blocks.20.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
940
+ "visual.blocks.20.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
941
+ "visual.blocks.20.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
942
+ "visual.blocks.20.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
943
+ "visual.blocks.20.norm1.weight": "model-00001-of-00014.safetensors",
944
+ "visual.blocks.20.norm2.weight": "model-00001-of-00014.safetensors",
945
+ "visual.blocks.21.attn.proj.bias": "model-00001-of-00014.safetensors",
946
+ "visual.blocks.21.attn.proj.weight": "model-00001-of-00014.safetensors",
947
+ "visual.blocks.21.attn.qkv.bias": "model-00001-of-00014.safetensors",
948
+ "visual.blocks.21.attn.qkv.weight": "model-00001-of-00014.safetensors",
949
+ "visual.blocks.21.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
950
+ "visual.blocks.21.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
951
+ "visual.blocks.21.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
952
+ "visual.blocks.21.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
953
+ "visual.blocks.21.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
954
+ "visual.blocks.21.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
955
+ "visual.blocks.21.norm1.weight": "model-00001-of-00014.safetensors",
956
+ "visual.blocks.21.norm2.weight": "model-00001-of-00014.safetensors",
957
+ "visual.blocks.22.attn.proj.bias": "model-00001-of-00014.safetensors",
958
+ "visual.blocks.22.attn.proj.weight": "model-00001-of-00014.safetensors",
959
+ "visual.blocks.22.attn.qkv.bias": "model-00001-of-00014.safetensors",
960
+ "visual.blocks.22.attn.qkv.weight": "model-00001-of-00014.safetensors",
961
+ "visual.blocks.22.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
962
+ "visual.blocks.22.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
963
+ "visual.blocks.22.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
964
+ "visual.blocks.22.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
965
+ "visual.blocks.22.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
966
+ "visual.blocks.22.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
967
+ "visual.blocks.22.norm1.weight": "model-00001-of-00014.safetensors",
968
+ "visual.blocks.22.norm2.weight": "model-00001-of-00014.safetensors",
969
+ "visual.blocks.23.attn.proj.bias": "model-00001-of-00014.safetensors",
970
+ "visual.blocks.23.attn.proj.weight": "model-00001-of-00014.safetensors",
971
+ "visual.blocks.23.attn.qkv.bias": "model-00001-of-00014.safetensors",
972
+ "visual.blocks.23.attn.qkv.weight": "model-00001-of-00014.safetensors",
973
+ "visual.blocks.23.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
974
+ "visual.blocks.23.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
975
+ "visual.blocks.23.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
976
+ "visual.blocks.23.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
977
+ "visual.blocks.23.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
978
+ "visual.blocks.23.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
979
+ "visual.blocks.23.norm1.weight": "model-00001-of-00014.safetensors",
980
+ "visual.blocks.23.norm2.weight": "model-00001-of-00014.safetensors",
981
+ "visual.blocks.24.attn.proj.bias": "model-00001-of-00014.safetensors",
982
+ "visual.blocks.24.attn.proj.weight": "model-00001-of-00014.safetensors",
983
+ "visual.blocks.24.attn.qkv.bias": "model-00001-of-00014.safetensors",
984
+ "visual.blocks.24.attn.qkv.weight": "model-00001-of-00014.safetensors",
985
+ "visual.blocks.24.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
986
+ "visual.blocks.24.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
987
+ "visual.blocks.24.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
988
+ "visual.blocks.24.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
989
+ "visual.blocks.24.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
990
+ "visual.blocks.24.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
991
+ "visual.blocks.24.norm1.weight": "model-00001-of-00014.safetensors",
992
+ "visual.blocks.24.norm2.weight": "model-00001-of-00014.safetensors",
993
+ "visual.blocks.25.attn.proj.bias": "model-00001-of-00014.safetensors",
994
+ "visual.blocks.25.attn.proj.weight": "model-00001-of-00014.safetensors",
995
+ "visual.blocks.25.attn.qkv.bias": "model-00001-of-00014.safetensors",
996
+ "visual.blocks.25.attn.qkv.weight": "model-00001-of-00014.safetensors",
997
+ "visual.blocks.25.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
998
+ "visual.blocks.25.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
999
+ "visual.blocks.25.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1000
+ "visual.blocks.25.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1001
+ "visual.blocks.25.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1002
+ "visual.blocks.25.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1003
+ "visual.blocks.25.norm1.weight": "model-00001-of-00014.safetensors",
1004
+ "visual.blocks.25.norm2.weight": "model-00001-of-00014.safetensors",
1005
+ "visual.blocks.26.attn.proj.bias": "model-00001-of-00014.safetensors",
1006
+ "visual.blocks.26.attn.proj.weight": "model-00001-of-00014.safetensors",
1007
+ "visual.blocks.26.attn.qkv.bias": "model-00001-of-00014.safetensors",
1008
+ "visual.blocks.26.attn.qkv.weight": "model-00001-of-00014.safetensors",
1009
+ "visual.blocks.26.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1010
+ "visual.blocks.26.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1011
+ "visual.blocks.26.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1012
+ "visual.blocks.26.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1013
+ "visual.blocks.26.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1014
+ "visual.blocks.26.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1015
+ "visual.blocks.26.norm1.weight": "model-00001-of-00014.safetensors",
1016
+ "visual.blocks.26.norm2.weight": "model-00001-of-00014.safetensors",
1017
+ "visual.blocks.27.attn.proj.bias": "model-00001-of-00014.safetensors",
1018
+ "visual.blocks.27.attn.proj.weight": "model-00001-of-00014.safetensors",
1019
+ "visual.blocks.27.attn.qkv.bias": "model-00001-of-00014.safetensors",
1020
+ "visual.blocks.27.attn.qkv.weight": "model-00001-of-00014.safetensors",
1021
+ "visual.blocks.27.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1022
+ "visual.blocks.27.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1023
+ "visual.blocks.27.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1024
+ "visual.blocks.27.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1025
+ "visual.blocks.27.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1026
+ "visual.blocks.27.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1027
+ "visual.blocks.27.norm1.weight": "model-00001-of-00014.safetensors",
1028
+ "visual.blocks.27.norm2.weight": "model-00001-of-00014.safetensors",
1029
+ "visual.blocks.28.attn.proj.bias": "model-00001-of-00014.safetensors",
1030
+ "visual.blocks.28.attn.proj.weight": "model-00001-of-00014.safetensors",
1031
+ "visual.blocks.28.attn.qkv.bias": "model-00001-of-00014.safetensors",
1032
+ "visual.blocks.28.attn.qkv.weight": "model-00001-of-00014.safetensors",
1033
+ "visual.blocks.28.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1034
+ "visual.blocks.28.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1035
+ "visual.blocks.28.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1036
+ "visual.blocks.28.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1037
+ "visual.blocks.28.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1038
+ "visual.blocks.28.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1039
+ "visual.blocks.28.norm1.weight": "model-00001-of-00014.safetensors",
1040
+ "visual.blocks.28.norm2.weight": "model-00001-of-00014.safetensors",
1041
+ "visual.blocks.29.attn.proj.bias": "model-00001-of-00014.safetensors",
1042
+ "visual.blocks.29.attn.proj.weight": "model-00001-of-00014.safetensors",
1043
+ "visual.blocks.29.attn.qkv.bias": "model-00001-of-00014.safetensors",
1044
+ "visual.blocks.29.attn.qkv.weight": "model-00001-of-00014.safetensors",
1045
+ "visual.blocks.29.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1046
+ "visual.blocks.29.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1047
+ "visual.blocks.29.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1048
+ "visual.blocks.29.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1049
+ "visual.blocks.29.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1050
+ "visual.blocks.29.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1051
+ "visual.blocks.29.norm1.weight": "model-00001-of-00014.safetensors",
1052
+ "visual.blocks.29.norm2.weight": "model-00001-of-00014.safetensors",
1053
+ "visual.blocks.3.attn.proj.bias": "model-00001-of-00014.safetensors",
1054
+ "visual.blocks.3.attn.proj.weight": "model-00001-of-00014.safetensors",
1055
+ "visual.blocks.3.attn.qkv.bias": "model-00001-of-00014.safetensors",
1056
+ "visual.blocks.3.attn.qkv.weight": "model-00001-of-00014.safetensors",
1057
+ "visual.blocks.3.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1058
+ "visual.blocks.3.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1059
+ "visual.blocks.3.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1060
+ "visual.blocks.3.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1061
+ "visual.blocks.3.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1062
+ "visual.blocks.3.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1063
+ "visual.blocks.3.norm1.weight": "model-00001-of-00014.safetensors",
1064
+ "visual.blocks.3.norm2.weight": "model-00001-of-00014.safetensors",
1065
+ "visual.blocks.30.attn.proj.bias": "model-00001-of-00014.safetensors",
1066
+ "visual.blocks.30.attn.proj.weight": "model-00001-of-00014.safetensors",
1067
+ "visual.blocks.30.attn.qkv.bias": "model-00001-of-00014.safetensors",
1068
+ "visual.blocks.30.attn.qkv.weight": "model-00001-of-00014.safetensors",
1069
+ "visual.blocks.30.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1070
+ "visual.blocks.30.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1071
+ "visual.blocks.30.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1072
+ "visual.blocks.30.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1073
+ "visual.blocks.30.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1074
+ "visual.blocks.30.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1075
+ "visual.blocks.30.norm1.weight": "model-00001-of-00014.safetensors",
1076
+ "visual.blocks.30.norm2.weight": "model-00001-of-00014.safetensors",
1077
+ "visual.blocks.31.attn.proj.bias": "model-00001-of-00014.safetensors",
1078
+ "visual.blocks.31.attn.proj.weight": "model-00001-of-00014.safetensors",
1079
+ "visual.blocks.31.attn.qkv.bias": "model-00001-of-00014.safetensors",
1080
+ "visual.blocks.31.attn.qkv.weight": "model-00001-of-00014.safetensors",
1081
+ "visual.blocks.31.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1082
+ "visual.blocks.31.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1083
+ "visual.blocks.31.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1084
+ "visual.blocks.31.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1085
+ "visual.blocks.31.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1086
+ "visual.blocks.31.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1087
+ "visual.blocks.31.norm1.weight": "model-00001-of-00014.safetensors",
1088
+ "visual.blocks.31.norm2.weight": "model-00001-of-00014.safetensors",
1089
+ "visual.blocks.4.attn.proj.bias": "model-00001-of-00014.safetensors",
1090
+ "visual.blocks.4.attn.proj.weight": "model-00001-of-00014.safetensors",
1091
+ "visual.blocks.4.attn.qkv.bias": "model-00001-of-00014.safetensors",
1092
+ "visual.blocks.4.attn.qkv.weight": "model-00001-of-00014.safetensors",
1093
+ "visual.blocks.4.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1094
+ "visual.blocks.4.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1095
+ "visual.blocks.4.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1096
+ "visual.blocks.4.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1097
+ "visual.blocks.4.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1098
+ "visual.blocks.4.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1099
+ "visual.blocks.4.norm1.weight": "model-00001-of-00014.safetensors",
1100
+ "visual.blocks.4.norm2.weight": "model-00001-of-00014.safetensors",
1101
+ "visual.blocks.5.attn.proj.bias": "model-00001-of-00014.safetensors",
1102
+ "visual.blocks.5.attn.proj.weight": "model-00001-of-00014.safetensors",
1103
+ "visual.blocks.5.attn.qkv.bias": "model-00001-of-00014.safetensors",
1104
+ "visual.blocks.5.attn.qkv.weight": "model-00001-of-00014.safetensors",
1105
+ "visual.blocks.5.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1106
+ "visual.blocks.5.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1107
+ "visual.blocks.5.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1108
+ "visual.blocks.5.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1109
+ "visual.blocks.5.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1110
+ "visual.blocks.5.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1111
+ "visual.blocks.5.norm1.weight": "model-00001-of-00014.safetensors",
1112
+ "visual.blocks.5.norm2.weight": "model-00001-of-00014.safetensors",
1113
+ "visual.blocks.6.attn.proj.bias": "model-00001-of-00014.safetensors",
1114
+ "visual.blocks.6.attn.proj.weight": "model-00001-of-00014.safetensors",
1115
+ "visual.blocks.6.attn.qkv.bias": "model-00001-of-00014.safetensors",
1116
+ "visual.blocks.6.attn.qkv.weight": "model-00001-of-00014.safetensors",
1117
+ "visual.blocks.6.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1118
+ "visual.blocks.6.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1119
+ "visual.blocks.6.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1120
+ "visual.blocks.6.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1121
+ "visual.blocks.6.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1122
+ "visual.blocks.6.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1123
+ "visual.blocks.6.norm1.weight": "model-00001-of-00014.safetensors",
1124
+ "visual.blocks.6.norm2.weight": "model-00001-of-00014.safetensors",
1125
+ "visual.blocks.7.attn.proj.bias": "model-00001-of-00014.safetensors",
1126
+ "visual.blocks.7.attn.proj.weight": "model-00001-of-00014.safetensors",
1127
+ "visual.blocks.7.attn.qkv.bias": "model-00001-of-00014.safetensors",
1128
+ "visual.blocks.7.attn.qkv.weight": "model-00001-of-00014.safetensors",
1129
+ "visual.blocks.7.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1130
+ "visual.blocks.7.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1131
+ "visual.blocks.7.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1132
+ "visual.blocks.7.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1133
+ "visual.blocks.7.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1134
+ "visual.blocks.7.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1135
+ "visual.blocks.7.norm1.weight": "model-00001-of-00014.safetensors",
1136
+ "visual.blocks.7.norm2.weight": "model-00001-of-00014.safetensors",
1137
+ "visual.blocks.8.attn.proj.bias": "model-00001-of-00014.safetensors",
1138
+ "visual.blocks.8.attn.proj.weight": "model-00001-of-00014.safetensors",
1139
+ "visual.blocks.8.attn.qkv.bias": "model-00001-of-00014.safetensors",
1140
+ "visual.blocks.8.attn.qkv.weight": "model-00001-of-00014.safetensors",
1141
+ "visual.blocks.8.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1142
+ "visual.blocks.8.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1143
+ "visual.blocks.8.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1144
+ "visual.blocks.8.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1145
+ "visual.blocks.8.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1146
+ "visual.blocks.8.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1147
+ "visual.blocks.8.norm1.weight": "model-00001-of-00014.safetensors",
1148
+ "visual.blocks.8.norm2.weight": "model-00001-of-00014.safetensors",
1149
+ "visual.blocks.9.attn.proj.bias": "model-00001-of-00014.safetensors",
1150
+ "visual.blocks.9.attn.proj.weight": "model-00001-of-00014.safetensors",
1151
+ "visual.blocks.9.attn.qkv.bias": "model-00001-of-00014.safetensors",
1152
+ "visual.blocks.9.attn.qkv.weight": "model-00001-of-00014.safetensors",
1153
+ "visual.blocks.9.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1154
+ "visual.blocks.9.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1155
+ "visual.blocks.9.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1156
+ "visual.blocks.9.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1157
+ "visual.blocks.9.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1158
+ "visual.blocks.9.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1159
+ "visual.blocks.9.norm1.weight": "model-00001-of-00014.safetensors",
1160
+ "visual.blocks.9.norm2.weight": "model-00001-of-00014.safetensors",
1161
+ "visual.merger.ln_q.weight": "model-00001-of-00014.safetensors",
1162
+ "visual.merger.mlp.0.bias": "model-00001-of-00014.safetensors",
1163
+ "visual.merger.mlp.0.weight": "model-00001-of-00014.safetensors",
1164
+ "visual.merger.mlp.2.bias": "model-00001-of-00014.safetensors",
1165
+ "visual.merger.mlp.2.weight": "model-00001-of-00014.safetensors",
1166
+ "visual.patch_embed.proj.weight": "model-00001-of-00014.safetensors"
1167
+ }
1168
+ }
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": "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,
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
+ }
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 sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
3
+ size 11421896
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
+ "clean_up_tokenization_spaces": false,
199
+ "eos_token": "<|im_end|>",
200
+ "errors": "replace",
201
+ "extra_special_tokens": {},
202
+ "model_max_length": 131072,
203
+ "pad_token": "<|endoftext|>",
204
+ "padding_side": "right",
205
+ "processor_class": "Qwen2_5_VLProcessor",
206
+ "split_special_tokens": false,
207
+ "tokenizer_class": "Qwen2Tokenizer",
208
+ "unk_token": null
209
+ }
video_preprocessor_config.json ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_valid_kwargs_names": [
3
+ "do_convert_rgb",
4
+ "do_resize",
5
+ "size",
6
+ "size_divisor",
7
+ "default_to_square",
8
+ "resample",
9
+ "do_rescale",
10
+ "rescale_factor",
11
+ "do_normalize",
12
+ "image_mean",
13
+ "image_std",
14
+ "do_pad",
15
+ "do_center_crop",
16
+ "crop_size",
17
+ "data_format",
18
+ "input_data_format",
19
+ "device",
20
+ "min_pixels",
21
+ "max_pixels",
22
+ "patch_size",
23
+ "temporal_patch_size",
24
+ "merge_size"
25
+ ],
26
+ "crop_size": null,
27
+ "data_format": "channels_first",
28
+ "default_to_square": true,
29
+ "device": null,
30
+ "do_center_crop": null,
31
+ "do_convert_rgb": true,
32
+ "do_normalize": true,
33
+ "do_pad": null,
34
+ "do_rescale": true,
35
+ "do_resize": true,
36
+ "image_mean": [
37
+ 0.48145466,
38
+ 0.4578275,
39
+ 0.40821073
40
+ ],
41
+ "image_processor_type": "Qwen2VLImageProcessor",
42
+ "image_std": [
43
+ 0.26862954,
44
+ 0.26130258,
45
+ 0.27577711
46
+ ],
47
+ "input_data_format": null,
48
+ "max_pixels": 12845056,
49
+ "merge_size": 2,
50
+ "min_pixels": 3136,
51
+ "model_valid_processing_keys": [
52
+ "do_convert_rgb",
53
+ "do_resize",
54
+ "size",
55
+ "size_divisor",
56
+ "default_to_square",
57
+ "resample",
58
+ "do_rescale",
59
+ "rescale_factor",
60
+ "do_normalize",
61
+ "image_mean",
62
+ "image_std",
63
+ "do_pad",
64
+ "do_center_crop",
65
+ "crop_size",
66
+ "data_format",
67
+ "input_data_format",
68
+ "device",
69
+ "min_pixels",
70
+ "max_pixels",
71
+ "patch_size",
72
+ "temporal_patch_size",
73
+ "merge_size"
74
+ ],
75
+ "patch_size": 14,
76
+ "processor_class": "Qwen2_5_VLProcessor",
77
+ "resample": 3,
78
+ "rescale_factor": 0.00392156862745098,
79
+ "size": {
80
+ "longest_edge": 12845056,
81
+ "shortest_edge": 3136
82
+ },
83
+ "size_divisor": null,
84
+ "temporal_patch_size": 2,
85
+ "video_processor_type": "Qwen2VLVideoProcessor"
86
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
zero_to_fp32.py ADDED
@@ -0,0 +1,674 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env python
2
+
3
+ # Copyright (c) Microsoft Corporation.
4
+ # SPDX-License-Identifier: Apache-2.0
5
+
6
+ # DeepSpeed Team
7
+
8
+ # This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
9
+ # copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
10
+ # the future. Once extracted, the weights don't require DeepSpeed and can be used in any
11
+ # application.
12
+ #
13
+ # example:
14
+ # python zero_to_fp32.py . output_dir/
15
+ # or
16
+ # python zero_to_fp32.py . output_dir/ --safe_serialization
17
+
18
+ import argparse
19
+ import torch
20
+ import glob
21
+ import math
22
+ import os
23
+ import re
24
+ import json
25
+ from tqdm import tqdm
26
+ from collections import OrderedDict
27
+ from dataclasses import dataclass
28
+
29
+ # while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
30
+ # DeepSpeed data structures it has to be available in the current python environment.
31
+ from deepspeed.utils import logger
32
+ from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
33
+ FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
34
+ FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
35
+
36
+
37
+ @dataclass
38
+ class zero_model_state:
39
+ buffers: dict()
40
+ param_shapes: dict()
41
+ shared_params: list
42
+ ds_version: int
43
+ frozen_param_shapes: dict()
44
+ frozen_param_fragments: dict()
45
+
46
+
47
+ debug = 0
48
+
49
+ # load to cpu
50
+ device = torch.device('cpu')
51
+
52
+
53
+ def atoi(text):
54
+ return int(text) if text.isdigit() else text
55
+
56
+
57
+ def natural_keys(text):
58
+ '''
59
+ alist.sort(key=natural_keys) sorts in human order
60
+ http://nedbatchelder.com/blog/200712/human_sorting.html
61
+ (See Toothy's implementation in the comments)
62
+ '''
63
+ return [atoi(c) for c in re.split(r'(\d+)', text)]
64
+
65
+
66
+ def get_model_state_file(checkpoint_dir, zero_stage):
67
+ if not os.path.isdir(checkpoint_dir):
68
+ raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
69
+
70
+ # there should be only one file
71
+ if zero_stage <= 2:
72
+ file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
73
+ elif zero_stage == 3:
74
+ file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
75
+
76
+ if not os.path.exists(file):
77
+ raise FileNotFoundError(f"can't find model states file at '{file}'")
78
+
79
+ return file
80
+
81
+
82
+ def get_checkpoint_files(checkpoint_dir, glob_pattern):
83
+ # XXX: need to test that this simple glob rule works for multi-node setup too
84
+ ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
85
+
86
+ if len(ckpt_files) == 0:
87
+ raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
88
+
89
+ return ckpt_files
90
+
91
+
92
+ def get_optim_files(checkpoint_dir):
93
+ return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
94
+
95
+
96
+ def get_model_state_files(checkpoint_dir):
97
+ return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
98
+
99
+
100
+ def parse_model_states(files):
101
+ zero_model_states = []
102
+ for file in files:
103
+ state_dict = torch.load(file, map_location=device)
104
+
105
+ if BUFFER_NAMES not in state_dict:
106
+ raise ValueError(f"{file} is not a model state checkpoint")
107
+ buffer_names = state_dict[BUFFER_NAMES]
108
+ if debug:
109
+ print("Found buffers:", buffer_names)
110
+
111
+ # recover just the buffers while restoring them to fp32 if they were saved in fp16
112
+ buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
113
+ param_shapes = state_dict[PARAM_SHAPES]
114
+
115
+ # collect parameters that are included in param_shapes
116
+ param_names = []
117
+ for s in param_shapes:
118
+ for name in s.keys():
119
+ param_names.append(name)
120
+
121
+ # update with frozen parameters
122
+ frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
123
+ if frozen_param_shapes is not None:
124
+ if debug:
125
+ print(f"Found frozen_param_shapes: {frozen_param_shapes}")
126
+ param_names += list(frozen_param_shapes.keys())
127
+
128
+ # handle shared params
129
+ shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
130
+
131
+ ds_version = state_dict.get(DS_VERSION, None)
132
+
133
+ frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
134
+
135
+ z_model_state = zero_model_state(buffers=buffers,
136
+ param_shapes=param_shapes,
137
+ shared_params=shared_params,
138
+ ds_version=ds_version,
139
+ frozen_param_shapes=frozen_param_shapes,
140
+ frozen_param_fragments=frozen_param_fragments)
141
+ zero_model_states.append(z_model_state)
142
+
143
+ return zero_model_states
144
+
145
+
146
+ def parse_optim_states(files, ds_checkpoint_dir):
147
+ total_files = len(files)
148
+ state_dicts = []
149
+ for f in files:
150
+ state_dict = torch.load(f, map_location=device)
151
+ # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
152
+ # and also handle the case where it was already removed by another helper script
153
+ state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
154
+ state_dicts.append(state_dict)
155
+
156
+ if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
157
+ raise ValueError(f"{files[0]} is not a zero checkpoint")
158
+ zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
159
+ world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
160
+
161
+ # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
162
+ # parameters can be different from data parallelism for non-expert parameters. So we can just
163
+ # use the max of the partition_count to get the dp world_size.
164
+
165
+ if type(world_size) is list:
166
+ world_size = max(world_size)
167
+
168
+ if world_size != total_files:
169
+ raise ValueError(
170
+ f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
171
+ "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
172
+ )
173
+
174
+ # the groups are named differently in each stage
175
+ if zero_stage <= 2:
176
+ fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
177
+ elif zero_stage == 3:
178
+ fp32_groups_key = FP32_FLAT_GROUPS
179
+ else:
180
+ raise ValueError(f"unknown zero stage {zero_stage}")
181
+
182
+ if zero_stage <= 2:
183
+ fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
184
+ elif zero_stage == 3:
185
+ # if there is more than one param group, there will be multiple flattened tensors - one
186
+ # flattened tensor per group - for simplicity merge them into a single tensor
187
+ #
188
+ # XXX: could make the script more memory efficient for when there are multiple groups - it
189
+ # will require matching the sub-lists of param_shapes for each param group flattened tensor
190
+
191
+ fp32_flat_groups = [
192
+ torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
193
+ ]
194
+
195
+ return zero_stage, world_size, fp32_flat_groups
196
+
197
+
198
+ def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
199
+ """
200
+ Returns fp32 state_dict reconstructed from ds checkpoint
201
+
202
+ Args:
203
+ - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
204
+
205
+ """
206
+ print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
207
+
208
+ optim_files = get_optim_files(ds_checkpoint_dir)
209
+ zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
210
+ print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
211
+
212
+ model_files = get_model_state_files(ds_checkpoint_dir)
213
+
214
+ zero_model_states = parse_model_states(model_files)
215
+ print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
216
+
217
+ if zero_stage <= 2:
218
+ return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
219
+ exclude_frozen_parameters)
220
+ elif zero_stage == 3:
221
+ return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
222
+ exclude_frozen_parameters)
223
+
224
+
225
+ def _zero2_merge_frozen_params(state_dict, zero_model_states):
226
+ if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
227
+ return
228
+
229
+ frozen_param_shapes = zero_model_states[0].frozen_param_shapes
230
+ frozen_param_fragments = zero_model_states[0].frozen_param_fragments
231
+
232
+ if debug:
233
+ num_elem = sum(s.numel() for s in frozen_param_shapes.values())
234
+ print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
235
+
236
+ wanted_params = len(frozen_param_shapes)
237
+ wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
238
+ avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
239
+ print(f'Frozen params: Have {avail_numel} numels to process.')
240
+ print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
241
+
242
+ total_params = 0
243
+ total_numel = 0
244
+ for name, shape in frozen_param_shapes.items():
245
+ total_params += 1
246
+ unpartitioned_numel = shape.numel()
247
+ total_numel += unpartitioned_numel
248
+
249
+ state_dict[name] = frozen_param_fragments[name]
250
+
251
+ if debug:
252
+ print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
253
+
254
+ print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
255
+
256
+
257
+ def _has_callable(obj, fn):
258
+ attr = getattr(obj, fn, None)
259
+ return callable(attr)
260
+
261
+
262
+ def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
263
+ param_shapes = zero_model_states[0].param_shapes
264
+
265
+ # Reconstruction protocol:
266
+ #
267
+ # XXX: document this
268
+
269
+ if debug:
270
+ for i in range(world_size):
271
+ for j in range(len(fp32_flat_groups[0])):
272
+ print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
273
+
274
+ # XXX: memory usage doubles here (zero2)
275
+ num_param_groups = len(fp32_flat_groups[0])
276
+ merged_single_partition_of_fp32_groups = []
277
+ for i in range(num_param_groups):
278
+ merged_partitions = [sd[i] for sd in fp32_flat_groups]
279
+ full_single_fp32_vector = torch.cat(merged_partitions, 0)
280
+ merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
281
+ avail_numel = sum(
282
+ [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
283
+
284
+ if debug:
285
+ wanted_params = sum([len(shapes) for shapes in param_shapes])
286
+ wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
287
+ # not asserting if there is a mismatch due to possible padding
288
+ print(f"Have {avail_numel} numels to process.")
289
+ print(f"Need {wanted_numel} numels in {wanted_params} params.")
290
+
291
+ # params
292
+ # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
293
+ # out-of-core computing solution
294
+ total_numel = 0
295
+ total_params = 0
296
+ for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
297
+ offset = 0
298
+ avail_numel = full_single_fp32_vector.numel()
299
+ for name, shape in shapes.items():
300
+
301
+ unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
302
+ total_numel += unpartitioned_numel
303
+ total_params += 1
304
+
305
+ if debug:
306
+ print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
307
+ state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
308
+ offset += unpartitioned_numel
309
+
310
+ # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
311
+ # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
312
+ # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
313
+ # live optimizer object, so we are checking that the numbers are within the right range
314
+ align_to = 2 * world_size
315
+
316
+ def zero2_align(x):
317
+ return align_to * math.ceil(x / align_to)
318
+
319
+ if debug:
320
+ print(f"original offset={offset}, avail_numel={avail_numel}")
321
+
322
+ offset = zero2_align(offset)
323
+ avail_numel = zero2_align(avail_numel)
324
+
325
+ if debug:
326
+ print(f"aligned offset={offset}, avail_numel={avail_numel}")
327
+
328
+ # Sanity check
329
+ if offset != avail_numel:
330
+ raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
331
+
332
+ print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
333
+
334
+
335
+ def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
336
+ exclude_frozen_parameters):
337
+ state_dict = OrderedDict()
338
+
339
+ # buffers
340
+ buffers = zero_model_states[0].buffers
341
+ state_dict.update(buffers)
342
+ if debug:
343
+ print(f"added {len(buffers)} buffers")
344
+
345
+ if not exclude_frozen_parameters:
346
+ _zero2_merge_frozen_params(state_dict, zero_model_states)
347
+
348
+ _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
349
+
350
+ # recover shared parameters
351
+ for pair in zero_model_states[0].shared_params:
352
+ if pair[1] in state_dict:
353
+ state_dict[pair[0]] = state_dict[pair[1]]
354
+
355
+ return state_dict
356
+
357
+
358
+ def zero3_partitioned_param_info(unpartitioned_numel, world_size):
359
+ remainder = unpartitioned_numel % world_size
360
+ padding_numel = (world_size - remainder) if remainder else 0
361
+ partitioned_numel = math.ceil(unpartitioned_numel / world_size)
362
+ return partitioned_numel, padding_numel
363
+
364
+
365
+ def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
366
+ if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
367
+ return
368
+
369
+ if debug:
370
+ for i in range(world_size):
371
+ num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
372
+ print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
373
+
374
+ frozen_param_shapes = zero_model_states[0].frozen_param_shapes
375
+ wanted_params = len(frozen_param_shapes)
376
+ wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
377
+ avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
378
+ print(f'Frozen params: Have {avail_numel} numels to process.')
379
+ print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
380
+
381
+ total_params = 0
382
+ total_numel = 0
383
+ for name, shape in zero_model_states[0].frozen_param_shapes.items():
384
+ total_params += 1
385
+ unpartitioned_numel = shape.numel()
386
+ total_numel += unpartitioned_numel
387
+
388
+ param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
389
+ state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
390
+
391
+ partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
392
+
393
+ if debug:
394
+ print(
395
+ f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
396
+ )
397
+
398
+ print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
399
+
400
+
401
+ def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
402
+ param_shapes = zero_model_states[0].param_shapes
403
+ avail_numel = fp32_flat_groups[0].numel() * world_size
404
+ # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
405
+ # param, re-consolidating each param, while dealing with padding if any
406
+
407
+ # merge list of dicts, preserving order
408
+ param_shapes = {k: v for d in param_shapes for k, v in d.items()}
409
+
410
+ if debug:
411
+ for i in range(world_size):
412
+ print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
413
+
414
+ wanted_params = len(param_shapes)
415
+ wanted_numel = sum(shape.numel() for shape in param_shapes.values())
416
+ # not asserting if there is a mismatch due to possible padding
417
+ avail_numel = fp32_flat_groups[0].numel() * world_size
418
+ print(f"Trainable params: Have {avail_numel} numels to process.")
419
+ print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
420
+
421
+ # params
422
+ # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
423
+ # out-of-core computing solution
424
+ offset = 0
425
+ total_numel = 0
426
+ total_params = 0
427
+ for name, shape in tqdm(param_shapes.items(), desc='Gathering Sharded Weights'):
428
+ unpartitioned_numel = shape.numel()
429
+ total_numel += unpartitioned_numel
430
+ total_params += 1
431
+ partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
432
+
433
+ if debug:
434
+ print(
435
+ f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
436
+ )
437
+
438
+ # XXX: memory usage doubles here
439
+ state_dict[name] = torch.cat(
440
+ tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
441
+ 0).narrow(0, 0, unpartitioned_numel).view(shape)
442
+ offset += partitioned_numel
443
+
444
+ offset *= world_size
445
+
446
+ # Sanity check
447
+ if offset != avail_numel:
448
+ raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
449
+
450
+ print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
451
+
452
+
453
+ def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
454
+ exclude_frozen_parameters):
455
+ state_dict = OrderedDict()
456
+
457
+ # buffers
458
+ buffers = zero_model_states[0].buffers
459
+ state_dict.update(buffers)
460
+ if debug:
461
+ print(f"added {len(buffers)} buffers")
462
+
463
+ if not exclude_frozen_parameters:
464
+ _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
465
+
466
+ _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
467
+
468
+ # recover shared parameters
469
+ for pair in zero_model_states[0].shared_params:
470
+ if pair[1] in state_dict:
471
+ state_dict[pair[0]] = state_dict[pair[1]]
472
+
473
+ return state_dict
474
+
475
+
476
+ def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
477
+ """
478
+ Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
479
+ ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
480
+ via a model hub.
481
+
482
+ Args:
483
+ - ``checkpoint_dir``: path to the desired checkpoint folder
484
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
485
+ - ``exclude_frozen_parameters``: exclude frozen parameters
486
+
487
+ Returns:
488
+ - pytorch ``state_dict``
489
+
490
+ Note: this approach may not work if your application doesn't have sufficient free CPU memory and
491
+ you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
492
+ the checkpoint.
493
+
494
+ A typical usage might be ::
495
+
496
+ from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
497
+ # do the training and checkpoint saving
498
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
499
+ model = model.cpu() # move to cpu
500
+ model.load_state_dict(state_dict)
501
+ # submit to model hub or save the model to share with others
502
+
503
+ In this example the ``model`` will no longer be usable in the deepspeed context of the same
504
+ application. i.e. you will need to re-initialize the deepspeed engine, since
505
+ ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
506
+
507
+ If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
508
+
509
+ """
510
+ if tag is None:
511
+ latest_path = os.path.join(checkpoint_dir, 'latest')
512
+ if os.path.isfile(latest_path):
513
+ with open(latest_path, 'r') as fd:
514
+ tag = fd.read().strip()
515
+ else:
516
+ raise ValueError(f"Unable to find 'latest' file at {latest_path}")
517
+
518
+ ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
519
+
520
+ if not os.path.isdir(ds_checkpoint_dir):
521
+ raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
522
+
523
+ return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
524
+
525
+
526
+ def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
527
+ output_dir,
528
+ max_shard_size="5GB",
529
+ safe_serialization=False,
530
+ tag=None,
531
+ exclude_frozen_parameters=False):
532
+ """
533
+ Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
534
+ loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
535
+
536
+ Args:
537
+ - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
538
+ - ``output_dir``: directory to the pytorch fp32 state_dict output files
539
+ - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
540
+ - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
541
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
542
+ - ``exclude_frozen_parameters``: exclude frozen parameters
543
+ """
544
+ # Dependency pre-check
545
+ if safe_serialization:
546
+ try:
547
+ from safetensors.torch import save_file
548
+ except ImportError:
549
+ print('If you want to use `safe_serialization`, please `pip install safetensors`')
550
+ raise
551
+ if max_shard_size is not None:
552
+ try:
553
+ from huggingface_hub import split_torch_state_dict_into_shards
554
+ except ImportError:
555
+ print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
556
+ raise
557
+
558
+ # Convert zero checkpoint to state_dict
559
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
560
+
561
+ # Shard the model if it is too big.
562
+ weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
563
+ if max_shard_size is not None:
564
+ filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
565
+ state_dict_split = split_torch_state_dict_into_shards(state_dict,
566
+ filename_pattern=filename_pattern,
567
+ max_shard_size=max_shard_size)
568
+ else:
569
+ from collections import namedtuple
570
+ StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
571
+ state_dict_split = StateDictSplit(is_sharded=False,
572
+ filename_to_tensors={weights_name: list(state_dict.keys())})
573
+
574
+ # Save the model
575
+ filename_to_tensors = state_dict_split.filename_to_tensors.items()
576
+ for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
577
+ shard = {tensor: state_dict[tensor].contiguous() for tensor in tensors}
578
+ output_path = os.path.join(output_dir, shard_file)
579
+ if safe_serialization:
580
+ save_file(shard, output_path, metadata={"format": "pt"})
581
+ else:
582
+ torch.save(shard, output_path)
583
+
584
+ # Save index if sharded
585
+ if state_dict_split.is_sharded:
586
+ index = {
587
+ "metadata": state_dict_split.metadata,
588
+ "weight_map": state_dict_split.tensor_to_filename,
589
+ }
590
+ save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
591
+ save_index_file = os.path.join(output_dir, save_index_file)
592
+ with open(save_index_file, "w", encoding="utf-8") as f:
593
+ content = json.dumps(index, indent=2, sort_keys=True) + "\n"
594
+ f.write(content)
595
+
596
+
597
+ def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
598
+ """
599
+ 1. Put the provided model to cpu
600
+ 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
601
+ 3. Load it into the provided model
602
+
603
+ Args:
604
+ - ``model``: the model object to update
605
+ - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
606
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
607
+
608
+ Returns:
609
+ - ``model`: modified model
610
+
611
+ Make sure you have plenty of CPU memory available before you call this function. If you don't
612
+ have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
613
+ conveniently placed for you in the checkpoint folder.
614
+
615
+ A typical usage might be ::
616
+
617
+ from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
618
+ model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
619
+ # submit to model hub or save the model to share with others
620
+
621
+ Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
622
+ of the same application. i.e. you will need to re-initialize the deepspeed engine, since
623
+ ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
624
+
625
+ """
626
+ logger.info(f"Extracting fp32 weights")
627
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
628
+
629
+ logger.info(f"Overwriting model with fp32 weights")
630
+ model = model.cpu()
631
+ model.load_state_dict(state_dict, strict=False)
632
+
633
+ return model
634
+
635
+
636
+ if __name__ == "__main__":
637
+ parser = argparse.ArgumentParser()
638
+ parser.add_argument("checkpoint_dir",
639
+ type=str,
640
+ help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
641
+ parser.add_argument("output_dir",
642
+ type=str,
643
+ help="directory to the pytorch fp32 state_dict output files"
644
+ "(e.g. path/checkpoint-12-output/)")
645
+ parser.add_argument(
646
+ "--max_shard_size",
647
+ type=str,
648
+ default="5GB",
649
+ help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
650
+ "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
651
+ "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
652
+ "without CPU OOM issues.")
653
+ parser.add_argument(
654
+ "--safe_serialization",
655
+ default=False,
656
+ action='store_true',
657
+ help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
658
+ parser.add_argument("-t",
659
+ "--tag",
660
+ type=str,
661
+ default=None,
662
+ help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
663
+ parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
664
+ parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
665
+ args = parser.parse_args()
666
+
667
+ debug = args.debug
668
+
669
+ convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
670
+ args.output_dir,
671
+ max_shard_size=args.max_shard_size,
672
+ safe_serialization=args.safe_serialization,
673
+ tag=args.tag,
674
+ exclude_frozen_parameters=args.exclude_frozen_parameters)