Hugging Face's logo

MMR1
/
MMR1-32B-SFT

Image-Text-to-Text
Transformers
Safetensors
qwen2_5_vl
image-to-text
multimodal-llm
reasoning
math
conversational
text-generation-inference
Model card Files
xet
Community
1
Sicong commited on
Commit
8b92ec0
·
verified ·
1 Parent(s): 8a14928

Add files using upload-large-folder tool

Browse files
Files changed (32) hide show
  1. .gitattributes +1 -0
  2. added_tokens.json +24 -0
  3. chat_template.jinja +7 -0
  4. config.json +171 -0
  5. generation_config.json +12 -0
  6. latest +1 -0
  7. merges.txt +0 -0
  8. model-00001-of-00014.safetensors +3 -0
  9. model-00002-of-00014.safetensors +3 -0
  10. model-00003-of-00014.safetensors +3 -0
  11. model-00004-of-00014.safetensors +3 -0
  12. model-00005-of-00014.safetensors +3 -0
  13. model-00006-of-00014.safetensors +3 -0
  14. model-00007-of-00014.safetensors +3 -0
  15. model-00008-of-00014.safetensors +3 -0
  16. model-00009-of-00014.safetensors +3 -0
  17. model-00010-of-00014.safetensors +3 -0
  18. model-00011-of-00014.safetensors +3 -0
  19. model-00012-of-00014.safetensors +3 -0
  20. model-00013-of-00014.safetensors +3 -0
  21. model-00014-of-00014.safetensors +3 -0
  22. model.safetensors.index.json +1169 -0
  23. preprocessor_config.json +37 -0
  24. scheduler.pt +3 -0
  25. special_tokens_map.json +31 -0
  26. tokenizer.json +3 -0
  27. tokenizer_config.json +209 -0
  28. trainer_state.json +2274 -0
  29. training_args.bin +3 -0
  30. video_preprocessor_config.json +43 -0
  31. vocab.json +0 -0
  32. 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
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 %}
config.json ADDED
@@ -0,0 +1,171 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen2_5_VLForConditionalGeneration"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "eos_token_id": 151645,
7
+ "hidden_act": "silu",
8
+ "hidden_size": 5120,
9
+ "image_token_id": 151655,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 27648,
12
+ "max_position_embeddings": 128000,
13
+ "max_window_layers": 64,
14
+ "model_type": "qwen2_5_vl",
15
+ "num_attention_heads": 40,
16
+ "num_hidden_layers": 64,
17
+ "num_key_value_heads": 8,
18
+ "pad_token_id": 151643,
19
+ "rms_norm_eps": 1e-06,
20
+ "rope_scaling": {
21
+ "mrope_section": [
22
+ 16,
23
+ 24,
24
+ 24
25
+ ],
26
+ "rope_type": "default",
27
+ "type": "default"
28
+ },
29
+ "rope_theta": 1000000.0,
30
+ "sliding_window": 32768,
31
+ "text_config": {
32
+ "architectures": [
33
+ "Qwen2_5_VLForConditionalGeneration"
34
+ ],
35
+ "attention_dropout": 0.0,
36
+ "eos_token_id": 151645,
37
+ "hidden_act": "silu",
38
+ "hidden_size": 5120,
39
+ "image_token_id": null,
40
+ "initializer_range": 0.02,
41
+ "intermediate_size": 27648,
42
+ "layer_types": [
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention",
53
+ "full_attention",
54
+ "full_attention",
55
+ "full_attention",
56
+ "full_attention",
57
+ "full_attention",
58
+ "full_attention",
59
+ "full_attention",
60
+ "full_attention",
61
+ "full_attention",
62
+ "full_attention",
63
+ "full_attention",
64
+ "full_attention",
65
+ "full_attention",
66
+ "full_attention",
67
+ "full_attention",
68
+ "full_attention",
69
+ "full_attention",
70
+ "full_attention",
71
+ "full_attention",
72
+ "full_attention",
73
+ "full_attention",
74
+ "full_attention",
75
+ "full_attention",
76
+ "full_attention",
77
+ "full_attention",
78
+ "full_attention",
79
+ "full_attention",
80
+ "full_attention",
81
+ "full_attention",
82
+ "full_attention",
83
+ "full_attention",
84
+ "full_attention",
85
+ "full_attention",
86
+ "full_attention",
87
+ "full_attention",
88
+ "full_attention",
89
+ "full_attention",
90
+ "full_attention",
91
+ "full_attention",
92
+ "full_attention",
93
+ "full_attention",
94
+ "full_attention",
95
+ "full_attention",
96
+ "full_attention",
97
+ "full_attention",
98
+ "full_attention",
99
+ "full_attention",
100
+ "full_attention",
101
+ "full_attention",
102
+ "full_attention",
103
+ "full_attention",
104
+ "full_attention",
105
+ "full_attention",
106
+ "full_attention"
107
+ ],
108
+ "max_position_embeddings": 128000,
109
+ "max_window_layers": 64,
110
+ "model_type": "qwen2_5_vl_text",
111
+ "num_attention_heads": 40,
112
+ "num_hidden_layers": 64,
113
+ "num_key_value_heads": 8,
114
+ "pad_token_id": 151643,
115
+ "rms_norm_eps": 1e-06,
116
+ "rope_scaling": {
117
+ "mrope_section": [
118
+ 16,
119
+ 24,
120
+ 24
121
+ ],
122
+ "rope_type": "default",
123
+ "type": "default"
124
+ },
125
+ "rope_theta": 1000000.0,
126
+ "sliding_window": null,
127
+ "torch_dtype": "float32",
128
+ "use_cache": false,
129
+ "use_sliding_window": false,
130
+ "video_token_id": null,
131
+ "vision_end_token_id": 151653,
132
+ "vision_start_token_id": 151652,
133
+ "vision_token_id": 151654,
134
+ "vocab_size": 152064
135
+ },
136
+ "tie_word_embeddings": false,
137
+ "torch_dtype": "bfloat16",
138
+ "transformers_version": "4.55.0",
139
+ "use_cache": false,
140
+ "use_sliding_window": false,
141
+ "video_token_id": 151656,
142
+ "vision_config": {
143
+ "depth": 32,
144
+ "fullatt_block_indexes": [
145
+ 7,
146
+ 15,
147
+ 23,
148
+ 31
149
+ ],
150
+ "hidden_act": "silu",
151
+ "hidden_size": 1280,
152
+ "in_channels": 3,
153
+ "in_chans": 3,
154
+ "initializer_range": 0.02,
155
+ "intermediate_size": 3456,
156
+ "model_type": "qwen2_5_vl",
157
+ "num_heads": 16,
158
+ "out_hidden_size": 5120,
159
+ "patch_size": 14,
160
+ "spatial_merge_size": 2,
161
+ "spatial_patch_size": 14,
162
+ "temporal_patch_size": 2,
163
+ "tokens_per_second": 2,
164
+ "torch_dtype": "float32",
165
+ "window_size": 112
166
+ },
167
+ "vision_end_token_id": 151653,
168
+ "vision_start_token_id": 151652,
169
+ "vision_token_id": 151654,
170
+ "vocab_size": 152064
171
+ }
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": "4.55.0"
12
+ }
latest ADDED
@@ -0,0 +1 @@
 
 
1
+ global_step3200
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model-00001-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c80ccaccbc028fca6ea6ae27742069b3bd1ed27eaddbf31574d5fe6f1dc6170
3
+ size 4958700928
model-00002-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:047e7bf224566c9d68c983a49fed2006871b1fdff8812bf07e5ff20ae0733ca0
3
+ size 4928488272
model-00003-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3238f89dad1e5c7f6c2ca6c240fc022af6f5d6398ae08ba7b1363ac5cf808e4a
3
+ size 4876059368
model-00004-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c83a7f2174834d1b29f2d2955329a412c7ea6cd9cd6f506d605b94b975c97a38
3
+ size 4876059416
model-00005-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4559e7e8b36d1725e007b38043789c747bfbe1064a4492ecba8c4dd3112df120
3
+ size 4876059416
model-00006-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6414c7f064583da361975ecdc845c7712b292b9cb5fbf8b2eac96c08566ec5d5
3
+ size 4876059416
model-00007-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7124088eb7e629f626ab40ac80d92ade25d74a768f051fed1fe8678453f38de
3
+ size 4876059416
model-00008-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f52eafa2b009d33f0bb5f4448494dcc40d835593609c4c7dd4301b6488a55a9e
3
+ size 4876059416
model-00009-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed6ba6b0dc5740d83bec4da11ff717ca5b61460aa1a5fb90be84078681f9b345
3
+ size 4876059416
model-00010-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae5c76a6f2585d8a070c45caed6ccae1101f77291c4c8cd80ae54e3abb390c11
3
+ size 4876059416
model-00011-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65eb0e667f9df9a81ee95a67ffacb539dd592761d8b79d885dd22f48706798de
3
+ size 4876059416
model-00012-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f69106f8dc9e0202510709c7cbfafe763035bad3210d68144ef0d3c2ef8b3d5
3
+ size 4876059416
model-00013-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e5d5248c193743ba83f2a98d23137971e3aa77345741201b17cfa70957e50e3
3
+ size 4876059416
model-00014-of-00014.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:967b6994de5726afaafd7f6114b2aeead862499f441b563493c2a45937c68fca
3
+ size 3381725320
model.safetensors.index.json ADDED
@@ -0,0 +1,1169 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_parameters": 1638656,
4
+ "total_size": 66905436672
5
+ },
6
+ "weight_map": {
7
+ "lm_head.weight": "model-00014-of-00014.safetensors",
8
+ "model.embed_tokens.weight": "model-00001-of-00014.safetensors",
9
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00014.safetensors",
10
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
11
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
12
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
13
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00014.safetensors",
14
+ "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00014.safetensors",
15
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00014.safetensors",
16
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00014.safetensors",
17
+ "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00014.safetensors",
18
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00014.safetensors",
19
+ "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00014.safetensors",
20
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00014.safetensors",
21
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00014.safetensors",
22
+ "model.layers.1.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
23
+ "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
24
+ "model.layers.1.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
25
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00014.safetensors",
26
+ "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00014.safetensors",
27
+ "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00014.safetensors",
28
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00014.safetensors",
29
+ "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00014.safetensors",
30
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00014.safetensors",
31
+ "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00014.safetensors",
32
+ "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00014.safetensors",
33
+ "model.layers.10.input_layernorm.weight": "model-00003-of-00014.safetensors",
34
+ "model.layers.10.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
35
+ "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
36
+ "model.layers.10.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
37
+ "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
38
+ "model.layers.10.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
39
+ "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
40
+ "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
41
+ "model.layers.10.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
42
+ "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
43
+ "model.layers.10.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
44
+ "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
45
+ "model.layers.11.input_layernorm.weight": "model-00003-of-00014.safetensors",
46
+ "model.layers.11.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
47
+ "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
48
+ "model.layers.11.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
49
+ "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
50
+ "model.layers.11.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
51
+ "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
52
+ "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
53
+ "model.layers.11.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
54
+ "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
55
+ "model.layers.11.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
56
+ "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
57
+ "model.layers.12.input_layernorm.weight": "model-00004-of-00014.safetensors",
58
+ "model.layers.12.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
59
+ "model.layers.12.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
60
+ "model.layers.12.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
61
+ "model.layers.12.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
62
+ "model.layers.12.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
63
+ "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
64
+ "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
65
+ "model.layers.12.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
66
+ "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
67
+ "model.layers.12.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
68
+ "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
69
+ "model.layers.13.input_layernorm.weight": "model-00004-of-00014.safetensors",
70
+ "model.layers.13.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
71
+ "model.layers.13.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
72
+ "model.layers.13.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
73
+ "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
74
+ "model.layers.13.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
75
+ "model.layers.13.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
76
+ "model.layers.13.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
77
+ "model.layers.13.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
78
+ "model.layers.13.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
79
+ "model.layers.13.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
80
+ "model.layers.13.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
81
+ "model.layers.14.input_layernorm.weight": "model-00004-of-00014.safetensors",
82
+ "model.layers.14.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
83
+ "model.layers.14.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
84
+ "model.layers.14.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
85
+ "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
86
+ "model.layers.14.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
87
+ "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
88
+ "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
89
+ "model.layers.14.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
90
+ "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
91
+ "model.layers.14.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
92
+ "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
93
+ "model.layers.15.input_layernorm.weight": "model-00004-of-00014.safetensors",
94
+ "model.layers.15.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
95
+ "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
96
+ "model.layers.15.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
97
+ "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
98
+ "model.layers.15.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
99
+ "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
100
+ "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
101
+ "model.layers.15.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
102
+ "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
103
+ "model.layers.15.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
104
+ "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
105
+ "model.layers.16.input_layernorm.weight": "model-00004-of-00014.safetensors",
106
+ "model.layers.16.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
107
+ "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
108
+ "model.layers.16.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
109
+ "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
110
+ "model.layers.16.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
111
+ "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
112
+ "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
113
+ "model.layers.16.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
114
+ "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
115
+ "model.layers.16.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
116
+ "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
117
+ "model.layers.17.input_layernorm.weight": "model-00005-of-00014.safetensors",
118
+ "model.layers.17.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
119
+ "model.layers.17.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
120
+ "model.layers.17.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
121
+ "model.layers.17.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
122
+ "model.layers.17.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
123
+ "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
124
+ "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
125
+ "model.layers.17.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
126
+ "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
127
+ "model.layers.17.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
128
+ "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
129
+ "model.layers.18.input_layernorm.weight": "model-00005-of-00014.safetensors",
130
+ "model.layers.18.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
131
+ "model.layers.18.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
132
+ "model.layers.18.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
133
+ "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
134
+ "model.layers.18.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
135
+ "model.layers.18.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
136
+ "model.layers.18.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
137
+ "model.layers.18.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
138
+ "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
139
+ "model.layers.18.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
140
+ "model.layers.18.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
141
+ "model.layers.19.input_layernorm.weight": "model-00005-of-00014.safetensors",
142
+ "model.layers.19.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
143
+ "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
144
+ "model.layers.19.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
145
+ "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
146
+ "model.layers.19.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
147
+ "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
148
+ "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
149
+ "model.layers.19.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
150
+ "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
151
+ "model.layers.19.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
152
+ "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
153
+ "model.layers.2.input_layernorm.weight": "model-00002-of-00014.safetensors",
154
+ "model.layers.2.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
155
+ "model.layers.2.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
156
+ "model.layers.2.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
157
+ "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
158
+ "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00014.safetensors",
159
+ "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00014.safetensors",
160
+ "model.layers.2.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
161
+ "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00014.safetensors",
162
+ "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00014.safetensors",
163
+ "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00014.safetensors",
164
+ "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00014.safetensors",
165
+ "model.layers.20.input_layernorm.weight": "model-00005-of-00014.safetensors",
166
+ "model.layers.20.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
167
+ "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
168
+ "model.layers.20.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
169
+ "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
170
+ "model.layers.20.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
171
+ "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
172
+ "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
173
+ "model.layers.20.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
174
+ "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
175
+ "model.layers.20.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
176
+ "model.layers.20.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
177
+ "model.layers.21.input_layernorm.weight": "model-00005-of-00014.safetensors",
178
+ "model.layers.21.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
179
+ "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
180
+ "model.layers.21.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
181
+ "model.layers.21.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
182
+ "model.layers.21.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
183
+ "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
184
+ "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
185
+ "model.layers.21.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
186
+ "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
187
+ "model.layers.21.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
188
+ "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
189
+ "model.layers.22.input_layernorm.weight": "model-00006-of-00014.safetensors",
190
+ "model.layers.22.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
191
+ "model.layers.22.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
192
+ "model.layers.22.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
193
+ "model.layers.22.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
194
+ "model.layers.22.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
195
+ "model.layers.22.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
196
+ "model.layers.22.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
197
+ "model.layers.22.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
198
+ "model.layers.22.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
199
+ "model.layers.22.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
200
+ "model.layers.22.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
201
+ "model.layers.23.input_layernorm.weight": "model-00006-of-00014.safetensors",
202
+ "model.layers.23.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
203
+ "model.layers.23.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
204
+ "model.layers.23.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
205
+ "model.layers.23.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
206
+ "model.layers.23.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
207
+ "model.layers.23.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
208
+ "model.layers.23.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
209
+ "model.layers.23.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
210
+ "model.layers.23.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
211
+ "model.layers.23.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
212
+ "model.layers.23.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
213
+ "model.layers.24.input_layernorm.weight": "model-00006-of-00014.safetensors",
214
+ "model.layers.24.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
215
+ "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
216
+ "model.layers.24.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
217
+ "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
218
+ "model.layers.24.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
219
+ "model.layers.24.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
220
+ "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
221
+ "model.layers.24.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
222
+ "model.layers.24.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
223
+ "model.layers.24.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
224
+ "model.layers.24.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
225
+ "model.layers.25.input_layernorm.weight": "model-00006-of-00014.safetensors",
226
+ "model.layers.25.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
227
+ "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
228
+ "model.layers.25.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
229
+ "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
230
+ "model.layers.25.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
231
+ "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
232
+ "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
233
+ "model.layers.25.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
234
+ "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
235
+ "model.layers.25.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
236
+ "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
237
+ "model.layers.26.input_layernorm.weight": "model-00006-of-00014.safetensors",
238
+ "model.layers.26.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
239
+ "model.layers.26.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
240
+ "model.layers.26.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
241
+ "model.layers.26.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
242
+ "model.layers.26.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
243
+ "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
244
+ "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
245
+ "model.layers.26.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
246
+ "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
247
+ "model.layers.26.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
248
+ "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
249
+ "model.layers.27.input_layernorm.weight": "model-00007-of-00014.safetensors",
250
+ "model.layers.27.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
251
+ "model.layers.27.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
252
+ "model.layers.27.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
253
+ "model.layers.27.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
254
+ "model.layers.27.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
255
+ "model.layers.27.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
256
+ "model.layers.27.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
257
+ "model.layers.27.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
258
+ "model.layers.27.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
259
+ "model.layers.27.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
260
+ "model.layers.27.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
261
+ "model.layers.28.input_layernorm.weight": "model-00007-of-00014.safetensors",
262
+ "model.layers.28.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
263
+ "model.layers.28.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
264
+ "model.layers.28.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
265
+ "model.layers.28.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
266
+ "model.layers.28.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
267
+ "model.layers.28.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
268
+ "model.layers.28.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
269
+ "model.layers.28.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
270
+ "model.layers.28.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
271
+ "model.layers.28.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
272
+ "model.layers.28.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
273
+ "model.layers.29.input_layernorm.weight": "model-00007-of-00014.safetensors",
274
+ "model.layers.29.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
275
+ "model.layers.29.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
276
+ "model.layers.29.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
277
+ "model.layers.29.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
278
+ "model.layers.29.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
279
+ "model.layers.29.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
280
+ "model.layers.29.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
281
+ "model.layers.29.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
282
+ "model.layers.29.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
283
+ "model.layers.29.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
284
+ "model.layers.29.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
285
+ "model.layers.3.input_layernorm.weight": "model-00002-of-00014.safetensors",
286
+ "model.layers.3.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
287
+ "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
288
+ "model.layers.3.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
289
+ "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
290
+ "model.layers.3.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
291
+ "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
292
+ "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
293
+ "model.layers.3.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
294
+ "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
295
+ "model.layers.3.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
296
+ "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
297
+ "model.layers.30.input_layernorm.weight": "model-00007-of-00014.safetensors",
298
+ "model.layers.30.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
299
+ "model.layers.30.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
300
+ "model.layers.30.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
301
+ "model.layers.30.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
302
+ "model.layers.30.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
303
+ "model.layers.30.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
304
+ "model.layers.30.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
305
+ "model.layers.30.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
306
+ "model.layers.30.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
307
+ "model.layers.30.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
308
+ "model.layers.30.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
309
+ "model.layers.31.input_layernorm.weight": "model-00007-of-00014.safetensors",
310
+ "model.layers.31.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
311
+ "model.layers.31.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
312
+ "model.layers.31.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
313
+ "model.layers.31.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
314
+ "model.layers.31.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
315
+ "model.layers.31.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
316
+ "model.layers.31.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
317
+ "model.layers.31.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
318
+ "model.layers.31.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
319
+ "model.layers.31.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
320
+ "model.layers.31.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
321
+ "model.layers.32.input_layernorm.weight": "model-00008-of-00014.safetensors",
322
+ "model.layers.32.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
323
+ "model.layers.32.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
324
+ "model.layers.32.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
325
+ "model.layers.32.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
326
+ "model.layers.32.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
327
+ "model.layers.32.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
328
+ "model.layers.32.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
329
+ "model.layers.32.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
330
+ "model.layers.32.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
331
+ "model.layers.32.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
332
+ "model.layers.32.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
333
+ "model.layers.33.input_layernorm.weight": "model-00008-of-00014.safetensors",
334
+ "model.layers.33.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
335
+ "model.layers.33.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
336
+ "model.layers.33.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
337
+ "model.layers.33.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
338
+ "model.layers.33.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
339
+ "model.layers.33.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
340
+ "model.layers.33.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
341
+ "model.layers.33.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
342
+ "model.layers.33.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
343
+ "model.layers.33.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
344
+ "model.layers.33.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
345
+ "model.layers.34.input_layernorm.weight": "model-00008-of-00014.safetensors",
346
+ "model.layers.34.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
347
+ "model.layers.34.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
348
+ "model.layers.34.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
349
+ "model.layers.34.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
350
+ "model.layers.34.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
351
+ "model.layers.34.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
352
+ "model.layers.34.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
353
+ "model.layers.34.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
354
+ "model.layers.34.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
355
+ "model.layers.34.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
356
+ "model.layers.34.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
357
+ "model.layers.35.input_layernorm.weight": "model-00008-of-00014.safetensors",
358
+ "model.layers.35.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
359
+ "model.layers.35.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
360
+ "model.layers.35.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
361
+ "model.layers.35.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
362
+ "model.layers.35.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
363
+ "model.layers.35.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
364
+ "model.layers.35.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
365
+ "model.layers.35.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
366
+ "model.layers.35.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
367
+ "model.layers.35.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
368
+ "model.layers.35.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
369
+ "model.layers.36.input_layernorm.weight": "model-00008-of-00014.safetensors",
370
+ "model.layers.36.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
371
+ "model.layers.36.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
372
+ "model.layers.36.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
373
+ "model.layers.36.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
374
+ "model.layers.36.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
375
+ "model.layers.36.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
376
+ "model.layers.36.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
377
+ "model.layers.36.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
378
+ "model.layers.36.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
379
+ "model.layers.36.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
380
+ "model.layers.36.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
381
+ "model.layers.37.input_layernorm.weight": "model-00009-of-00014.safetensors",
382
+ "model.layers.37.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
383
+ "model.layers.37.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
384
+ "model.layers.37.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
385
+ "model.layers.37.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
386
+ "model.layers.37.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
387
+ "model.layers.37.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
388
+ "model.layers.37.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
389
+ "model.layers.37.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
390
+ "model.layers.37.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
391
+ "model.layers.37.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
392
+ "model.layers.37.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
393
+ "model.layers.38.input_layernorm.weight": "model-00009-of-00014.safetensors",
394
+ "model.layers.38.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
395
+ "model.layers.38.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
396
+ "model.layers.38.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
397
+ "model.layers.38.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
398
+ "model.layers.38.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
399
+ "model.layers.38.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
400
+ "model.layers.38.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
401
+ "model.layers.38.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
402
+ "model.layers.38.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
403
+ "model.layers.38.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
404
+ "model.layers.38.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
405
+ "model.layers.39.input_layernorm.weight": "model-00009-of-00014.safetensors",
406
+ "model.layers.39.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
407
+ "model.layers.39.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
408
+ "model.layers.39.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
409
+ "model.layers.39.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
410
+ "model.layers.39.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
411
+ "model.layers.39.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
412
+ "model.layers.39.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
413
+ "model.layers.39.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
414
+ "model.layers.39.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
415
+ "model.layers.39.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
416
+ "model.layers.39.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
417
+ "model.layers.4.input_layernorm.weight": "model-00002-of-00014.safetensors",
418
+ "model.layers.4.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
419
+ "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
420
+ "model.layers.4.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
421
+ "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
422
+ "model.layers.4.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
423
+ "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
424
+ "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
425
+ "model.layers.4.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
426
+ "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
427
+ "model.layers.4.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
428
+ "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
429
+ "model.layers.40.input_layernorm.weight": "model-00009-of-00014.safetensors",
430
+ "model.layers.40.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
431
+ "model.layers.40.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
432
+ "model.layers.40.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
433
+ "model.layers.40.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
434
+ "model.layers.40.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
435
+ "model.layers.40.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
436
+ "model.layers.40.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
437
+ "model.layers.40.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
438
+ "model.layers.40.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
439
+ "model.layers.40.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
440
+ "model.layers.40.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
441
+ "model.layers.41.input_layernorm.weight": "model-00009-of-00014.safetensors",
442
+ "model.layers.41.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
443
+ "model.layers.41.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
444
+ "model.layers.41.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
445
+ "model.layers.41.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
446
+ "model.layers.41.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
447
+ "model.layers.41.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
448
+ "model.layers.41.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
449
+ "model.layers.41.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
450
+ "model.layers.41.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
451
+ "model.layers.41.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
452
+ "model.layers.41.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
453
+ "model.layers.42.input_layernorm.weight": "model-00010-of-00014.safetensors",
454
+ "model.layers.42.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
455
+ "model.layers.42.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
456
+ "model.layers.42.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
457
+ "model.layers.42.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
458
+ "model.layers.42.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
459
+ "model.layers.42.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
460
+ "model.layers.42.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
461
+ "model.layers.42.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
462
+ "model.layers.42.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
463
+ "model.layers.42.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
464
+ "model.layers.42.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
465
+ "model.layers.43.input_layernorm.weight": "model-00010-of-00014.safetensors",
466
+ "model.layers.43.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
467
+ "model.layers.43.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
468
+ "model.layers.43.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
469
+ "model.layers.43.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
470
+ "model.layers.43.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
471
+ "model.layers.43.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
472
+ "model.layers.43.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
473
+ "model.layers.43.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
474
+ "model.layers.43.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
475
+ "model.layers.43.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
476
+ "model.layers.43.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
477
+ "model.layers.44.input_layernorm.weight": "model-00010-of-00014.safetensors",
478
+ "model.layers.44.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
479
+ "model.layers.44.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
480
+ "model.layers.44.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
481
+ "model.layers.44.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
482
+ "model.layers.44.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
483
+ "model.layers.44.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
484
+ "model.layers.44.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
485
+ "model.layers.44.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
486
+ "model.layers.44.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
487
+ "model.layers.44.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
488
+ "model.layers.44.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
489
+ "model.layers.45.input_layernorm.weight": "model-00010-of-00014.safetensors",
490
+ "model.layers.45.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
491
+ "model.layers.45.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
492
+ "model.layers.45.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
493
+ "model.layers.45.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
494
+ "model.layers.45.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
495
+ "model.layers.45.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
496
+ "model.layers.45.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
497
+ "model.layers.45.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
498
+ "model.layers.45.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
499
+ "model.layers.45.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
500
+ "model.layers.45.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
501
+ "model.layers.46.input_layernorm.weight": "model-00010-of-00014.safetensors",
502
+ "model.layers.46.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
503
+ "model.layers.46.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
504
+ "model.layers.46.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
505
+ "model.layers.46.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
506
+ "model.layers.46.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
507
+ "model.layers.46.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
508
+ "model.layers.46.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
509
+ "model.layers.46.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
510
+ "model.layers.46.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
511
+ "model.layers.46.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
512
+ "model.layers.46.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
513
+ "model.layers.47.input_layernorm.weight": "model-00011-of-00014.safetensors",
514
+ "model.layers.47.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
515
+ "model.layers.47.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
516
+ "model.layers.47.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
517
+ "model.layers.47.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
518
+ "model.layers.47.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
519
+ "model.layers.47.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
520
+ "model.layers.47.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
521
+ "model.layers.47.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
522
+ "model.layers.47.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
523
+ "model.layers.47.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
524
+ "model.layers.47.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
525
+ "model.layers.48.input_layernorm.weight": "model-00011-of-00014.safetensors",
526
+ "model.layers.48.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
527
+ "model.layers.48.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
528
+ "model.layers.48.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
529
+ "model.layers.48.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
530
+ "model.layers.48.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
531
+ "model.layers.48.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
532
+ "model.layers.48.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
533
+ "model.layers.48.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
534
+ "model.layers.48.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
535
+ "model.layers.48.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
536
+ "model.layers.48.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
537
+ "model.layers.49.input_layernorm.weight": "model-00011-of-00014.safetensors",
538
+ "model.layers.49.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
539
+ "model.layers.49.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
540
+ "model.layers.49.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
541
+ "model.layers.49.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
542
+ "model.layers.49.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
543
+ "model.layers.49.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
544
+ "model.layers.49.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
545
+ "model.layers.49.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
546
+ "model.layers.49.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
547
+ "model.layers.49.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
548
+ "model.layers.49.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
549
+ "model.layers.5.input_layernorm.weight": "model-00002-of-00014.safetensors",
550
+ "model.layers.5.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
551
+ "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
552
+ "model.layers.5.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
553
+ "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
554
+ "model.layers.5.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
555
+ "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
556
+ "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
557
+ "model.layers.5.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
558
+ "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
559
+ "model.layers.5.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
560
+ "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
561
+ "model.layers.50.input_layernorm.weight": "model-00011-of-00014.safetensors",
562
+ "model.layers.50.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
563
+ "model.layers.50.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
564
+ "model.layers.50.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
565
+ "model.layers.50.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
566
+ "model.layers.50.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
567
+ "model.layers.50.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
568
+ "model.layers.50.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
569
+ "model.layers.50.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
570
+ "model.layers.50.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
571
+ "model.layers.50.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
572
+ "model.layers.50.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
573
+ "model.layers.51.input_layernorm.weight": "model-00011-of-00014.safetensors",
574
+ "model.layers.51.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
575
+ "model.layers.51.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
576
+ "model.layers.51.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
577
+ "model.layers.51.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
578
+ "model.layers.51.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
579
+ "model.layers.51.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
580
+ "model.layers.51.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
581
+ "model.layers.51.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
582
+ "model.layers.51.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
583
+ "model.layers.51.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
584
+ "model.layers.51.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
585
+ "model.layers.52.input_layernorm.weight": "model-00012-of-00014.safetensors",
586
+ "model.layers.52.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
587
+ "model.layers.52.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
588
+ "model.layers.52.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
589
+ "model.layers.52.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
590
+ "model.layers.52.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
591
+ "model.layers.52.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
592
+ "model.layers.52.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
593
+ "model.layers.52.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
594
+ "model.layers.52.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
595
+ "model.layers.52.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
596
+ "model.layers.52.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
597
+ "model.layers.53.input_layernorm.weight": "model-00012-of-00014.safetensors",
598
+ "model.layers.53.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
599
+ "model.layers.53.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
600
+ "model.layers.53.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
601
+ "model.layers.53.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
602
+ "model.layers.53.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
603
+ "model.layers.53.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
604
+ "model.layers.53.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
605
+ "model.layers.53.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
606
+ "model.layers.53.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
607
+ "model.layers.53.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
608
+ "model.layers.53.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
609
+ "model.layers.54.input_layernorm.weight": "model-00012-of-00014.safetensors",
610
+ "model.layers.54.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
611
+ "model.layers.54.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
612
+ "model.layers.54.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
613
+ "model.layers.54.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
614
+ "model.layers.54.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
615
+ "model.layers.54.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
616
+ "model.layers.54.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
617
+ "model.layers.54.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
618
+ "model.layers.54.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
619
+ "model.layers.54.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
620
+ "model.layers.54.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
621
+ "model.layers.55.input_layernorm.weight": "model-00012-of-00014.safetensors",
622
+ "model.layers.55.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
623
+ "model.layers.55.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
624
+ "model.layers.55.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
625
+ "model.layers.55.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
626
+ "model.layers.55.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
627
+ "model.layers.55.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
628
+ "model.layers.55.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
629
+ "model.layers.55.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
630
+ "model.layers.55.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
631
+ "model.layers.55.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
632
+ "model.layers.55.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
633
+ "model.layers.56.input_layernorm.weight": "model-00012-of-00014.safetensors",
634
+ "model.layers.56.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
635
+ "model.layers.56.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
636
+ "model.layers.56.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
637
+ "model.layers.56.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
638
+ "model.layers.56.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
639
+ "model.layers.56.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
640
+ "model.layers.56.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
641
+ "model.layers.56.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
642
+ "model.layers.56.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
643
+ "model.layers.56.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
644
+ "model.layers.56.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
645
+ "model.layers.57.input_layernorm.weight": "model-00013-of-00014.safetensors",
646
+ "model.layers.57.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
647
+ "model.layers.57.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
648
+ "model.layers.57.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
649
+ "model.layers.57.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
650
+ "model.layers.57.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
651
+ "model.layers.57.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
652
+ "model.layers.57.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
653
+ "model.layers.57.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
654
+ "model.layers.57.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
655
+ "model.layers.57.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
656
+ "model.layers.57.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
657
+ "model.layers.58.input_layernorm.weight": "model-00013-of-00014.safetensors",
658
+ "model.layers.58.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
659
+ "model.layers.58.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
660
+ "model.layers.58.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
661
+ "model.layers.58.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
662
+ "model.layers.58.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
663
+ "model.layers.58.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
664
+ "model.layers.58.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
665
+ "model.layers.58.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
666
+ "model.layers.58.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
667
+ "model.layers.58.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
668
+ "model.layers.58.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
669
+ "model.layers.59.input_layernorm.weight": "model-00013-of-00014.safetensors",
670
+ "model.layers.59.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
671
+ "model.layers.59.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
672
+ "model.layers.59.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
673
+ "model.layers.59.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
674
+ "model.layers.59.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
675
+ "model.layers.59.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
676
+ "model.layers.59.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
677
+ "model.layers.59.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
678
+ "model.layers.59.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
679
+ "model.layers.59.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
680
+ "model.layers.59.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
681
+ "model.layers.6.input_layernorm.weight": "model-00002-of-00014.safetensors",
682
+ "model.layers.6.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
683
+ "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
684
+ "model.layers.6.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
685
+ "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
686
+ "model.layers.6.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
687
+ "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
688
+ "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
689
+ "model.layers.6.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
690
+ "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
691
+ "model.layers.6.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
692
+ "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
693
+ "model.layers.60.input_layernorm.weight": "model-00013-of-00014.safetensors",
694
+ "model.layers.60.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
695
+ "model.layers.60.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
696
+ "model.layers.60.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
697
+ "model.layers.60.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
698
+ "model.layers.60.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
699
+ "model.layers.60.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
700
+ "model.layers.60.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
701
+ "model.layers.60.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
702
+ "model.layers.60.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
703
+ "model.layers.60.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
704
+ "model.layers.60.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
705
+ "model.layers.61.input_layernorm.weight": "model-00013-of-00014.safetensors",
706
+ "model.layers.61.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
707
+ "model.layers.61.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
708
+ "model.layers.61.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
709
+ "model.layers.61.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
710
+ "model.layers.61.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
711
+ "model.layers.61.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
712
+ "model.layers.61.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
713
+ "model.layers.61.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
714
+ "model.layers.61.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
715
+ "model.layers.61.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
716
+ "model.layers.61.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
717
+ "model.layers.62.input_layernorm.weight": "model-00014-of-00014.safetensors",
718
+ "model.layers.62.mlp.down_proj.weight": "model-00014-of-00014.safetensors",
719
+ "model.layers.62.mlp.gate_proj.weight": "model-00014-of-00014.safetensors",
720
+ "model.layers.62.mlp.up_proj.weight": "model-00014-of-00014.safetensors",
721
+ "model.layers.62.post_attention_layernorm.weight": "model-00014-of-00014.safetensors",
722
+ "model.layers.62.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
723
+ "model.layers.62.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
724
+ "model.layers.62.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
725
+ "model.layers.62.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
726
+ "model.layers.62.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
727
+ "model.layers.62.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
728
+ "model.layers.62.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
729
+ "model.layers.63.input_layernorm.weight": "model-00014-of-00014.safetensors",
730
+ "model.layers.63.mlp.down_proj.weight": "model-00014-of-00014.safetensors",
731
+ "model.layers.63.mlp.gate_proj.weight": "model-00014-of-00014.safetensors",
732
+ "model.layers.63.mlp.up_proj.weight": "model-00014-of-00014.safetensors",
733
+ "model.layers.63.post_attention_layernorm.weight": "model-00014-of-00014.safetensors",
734
+ "model.layers.63.self_attn.k_proj.bias": "model-00014-of-00014.safetensors",
735
+ "model.layers.63.self_attn.k_proj.weight": "model-00014-of-00014.safetensors",
736
+ "model.layers.63.self_attn.o_proj.weight": "model-00014-of-00014.safetensors",
737
+ "model.layers.63.self_attn.q_proj.bias": "model-00014-of-00014.safetensors",
738
+ "model.layers.63.self_attn.q_proj.weight": "model-00014-of-00014.safetensors",
739
+ "model.layers.63.self_attn.v_proj.bias": "model-00014-of-00014.safetensors",
740
+ "model.layers.63.self_attn.v_proj.weight": "model-00014-of-00014.safetensors",
741
+ "model.layers.7.input_layernorm.weight": "model-00003-of-00014.safetensors",
742
+ "model.layers.7.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
743
+ "model.layers.7.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
744
+ "model.layers.7.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
745
+ "model.layers.7.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
746
+ "model.layers.7.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
747
+ "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
748
+ "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
749
+ "model.layers.7.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
750
+ "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
751
+ "model.layers.7.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
752
+ "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
753
+ "model.layers.8.input_layernorm.weight": "model-00003-of-00014.safetensors",
754
+ "model.layers.8.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
755
+ "model.layers.8.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
756
+ "model.layers.8.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
757
+ "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
758
+ "model.layers.8.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
759
+ "model.layers.8.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
760
+ "model.layers.8.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
761
+ "model.layers.8.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
762
+ "model.layers.8.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
763
+ "model.layers.8.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
764
+ "model.layers.8.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
765
+ "model.layers.9.input_layernorm.weight": "model-00003-of-00014.safetensors",
766
+ "model.layers.9.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
767
+ "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
768
+ "model.layers.9.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
769
+ "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
770
+ "model.layers.9.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
771
+ "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
772
+ "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
773
+ "model.layers.9.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
774
+ "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
775
+ "model.layers.9.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
776
+ "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
777
+ "model.norm.weight": "model-00014-of-00014.safetensors",
778
+ "visual.blocks.0.attn.proj.bias": "model-00001-of-00014.safetensors",
779
+ "visual.blocks.0.attn.proj.weight": "model-00001-of-00014.safetensors",
780
+ "visual.blocks.0.attn.qkv.bias": "model-00001-of-00014.safetensors",
781
+ "visual.blocks.0.attn.qkv.weight": "model-00001-of-00014.safetensors",
782
+ "visual.blocks.0.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
783
+ "visual.blocks.0.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
784
+ "visual.blocks.0.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
785
+ "visual.blocks.0.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
786
+ "visual.blocks.0.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
787
+ "visual.blocks.0.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
788
+ "visual.blocks.0.norm1.weight": "model-00001-of-00014.safetensors",
789
+ "visual.blocks.0.norm2.weight": "model-00001-of-00014.safetensors",
790
+ "visual.blocks.1.attn.proj.bias": "model-00001-of-00014.safetensors",
791
+ "visual.blocks.1.attn.proj.weight": "model-00001-of-00014.safetensors",
792
+ "visual.blocks.1.attn.qkv.bias": "model-00001-of-00014.safetensors",
793
+ "visual.blocks.1.attn.qkv.weight": "model-00001-of-00014.safetensors",
794
+ "visual.blocks.1.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
795
+ "visual.blocks.1.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
796
+ "visual.blocks.1.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
797
+ "visual.blocks.1.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
798
+ "visual.blocks.1.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
799
+ "visual.blocks.1.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
800
+ "visual.blocks.1.norm1.weight": "model-00001-of-00014.safetensors",
801
+ "visual.blocks.1.norm2.weight": "model-00001-of-00014.safetensors",
802
+ "visual.blocks.10.attn.proj.bias": "model-00001-of-00014.safetensors",
803
+ "visual.blocks.10.attn.proj.weight": "model-00001-of-00014.safetensors",
804
+ "visual.blocks.10.attn.qkv.bias": "model-00001-of-00014.safetensors",
805
+ "visual.blocks.10.attn.qkv.weight": "model-00001-of-00014.safetensors",
806
+ "visual.blocks.10.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
807
+ "visual.blocks.10.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
808
+ "visual.blocks.10.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
809
+ "visual.blocks.10.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
810
+ "visual.blocks.10.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
811
+ "visual.blocks.10.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
812
+ "visual.blocks.10.norm1.weight": "model-00001-of-00014.safetensors",
813
+ "visual.blocks.10.norm2.weight": "model-00001-of-00014.safetensors",
814
+ "visual.blocks.11.attn.proj.bias": "model-00001-of-00014.safetensors",
815
+ "visual.blocks.11.attn.proj.weight": "model-00001-of-00014.safetensors",
816
+ "visual.blocks.11.attn.qkv.bias": "model-00001-of-00014.safetensors",
817
+ "visual.blocks.11.attn.qkv.weight": "model-00001-of-00014.safetensors",
818
+ "visual.blocks.11.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
819
+ "visual.blocks.11.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
820
+ "visual.blocks.11.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
821
+ "visual.blocks.11.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
822
+ "visual.blocks.11.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
823
+ "visual.blocks.11.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
824
+ "visual.blocks.11.norm1.weight": "model-00001-of-00014.safetensors",
825
+ "visual.blocks.11.norm2.weight": "model-00001-of-00014.safetensors",
826
+ "visual.blocks.12.attn.proj.bias": "model-00001-of-00014.safetensors",
827
+ "visual.blocks.12.attn.proj.weight": "model-00001-of-00014.safetensors",
828
+ "visual.blocks.12.attn.qkv.bias": "model-00001-of-00014.safetensors",
829
+ "visual.blocks.12.attn.qkv.weight": "model-00001-of-00014.safetensors",
830
+ "visual.blocks.12.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
831
+ "visual.blocks.12.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
832
+ "visual.blocks.12.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
833
+ "visual.blocks.12.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
834
+ "visual.blocks.12.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
835
+ "visual.blocks.12.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
836
+ "visual.blocks.12.norm1.weight": "model-00001-of-00014.safetensors",
837
+ "visual.blocks.12.norm2.weight": "model-00001-of-00014.safetensors",
838
+ "visual.blocks.13.attn.proj.bias": "model-00001-of-00014.safetensors",
839
+ "visual.blocks.13.attn.proj.weight": "model-00001-of-00014.safetensors",
840
+ "visual.blocks.13.attn.qkv.bias": "model-00001-of-00014.safetensors",
841
+ "visual.blocks.13.attn.qkv.weight": "model-00001-of-00014.safetensors",
842
+ "visual.blocks.13.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
843
+ "visual.blocks.13.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
844
+ "visual.blocks.13.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
845
+ "visual.blocks.13.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
846
+ "visual.blocks.13.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
847
+ "visual.blocks.13.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
848
+ "visual.blocks.13.norm1.weight": "model-00001-of-00014.safetensors",
849
+ "visual.blocks.13.norm2.weight": "model-00001-of-00014.safetensors",
850
+ "visual.blocks.14.attn.proj.bias": "model-00001-of-00014.safetensors",
851
+ "visual.blocks.14.attn.proj.weight": "model-00001-of-00014.safetensors",
852
+ "visual.blocks.14.attn.qkv.bias": "model-00001-of-00014.safetensors",
853
+ "visual.blocks.14.attn.qkv.weight": "model-00001-of-00014.safetensors",
854
+ "visual.blocks.14.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
855
+ "visual.blocks.14.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
856
+ "visual.blocks.14.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
857
+ "visual.blocks.14.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
858
+ "visual.blocks.14.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
859
+ "visual.blocks.14.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
860
+ "visual.blocks.14.norm1.weight": "model-00001-of-00014.safetensors",
861
+ "visual.blocks.14.norm2.weight": "model-00001-of-00014.safetensors",
862
+ "visual.blocks.15.attn.proj.bias": "model-00001-of-00014.safetensors",
863
+ "visual.blocks.15.attn.proj.weight": "model-00001-of-00014.safetensors",
864
+ "visual.blocks.15.attn.qkv.bias": "model-00001-of-00014.safetensors",
865
+ "visual.blocks.15.attn.qkv.weight": "model-00001-of-00014.safetensors",
866
+ "visual.blocks.15.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
867
+ "visual.blocks.15.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
868
+ "visual.blocks.15.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
869
+ "visual.blocks.15.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
870
+ "visual.blocks.15.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
871
+ "visual.blocks.15.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
872
+ "visual.blocks.15.norm1.weight": "model-00001-of-00014.safetensors",
873
+ "visual.blocks.15.norm2.weight": "model-00001-of-00014.safetensors",
874
+ "visual.blocks.16.attn.proj.bias": "model-00001-of-00014.safetensors",
875
+ "visual.blocks.16.attn.proj.weight": "model-00001-of-00014.safetensors",
876
+ "visual.blocks.16.attn.qkv.bias": "model-00001-of-00014.safetensors",
877
+ "visual.blocks.16.attn.qkv.weight": "model-00001-of-00014.safetensors",
878
+ "visual.blocks.16.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
879
+ "visual.blocks.16.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
880
+ "visual.blocks.16.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
881
+ "visual.blocks.16.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
882
+ "visual.blocks.16.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
883
+ "visual.blocks.16.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
884
+ "visual.blocks.16.norm1.weight": "model-00001-of-00014.safetensors",
885
+ "visual.blocks.16.norm2.weight": "model-00001-of-00014.safetensors",
886
+ "visual.blocks.17.attn.proj.bias": "model-00001-of-00014.safetensors",
887
+ "visual.blocks.17.attn.proj.weight": "model-00001-of-00014.safetensors",
888
+ "visual.blocks.17.attn.qkv.bias": "model-00001-of-00014.safetensors",
889
+ "visual.blocks.17.attn.qkv.weight": "model-00001-of-00014.safetensors",
890
+ "visual.blocks.17.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
891
+ "visual.blocks.17.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
892
+ "visual.blocks.17.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
893
+ "visual.blocks.17.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
894
+ "visual.blocks.17.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
895
+ "visual.blocks.17.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
896
+ "visual.blocks.17.norm1.weight": "model-00001-of-00014.safetensors",
897
+ "visual.blocks.17.norm2.weight": "model-00001-of-00014.safetensors",
898
+ "visual.blocks.18.attn.proj.bias": "model-00001-of-00014.safetensors",
899
+ "visual.blocks.18.attn.proj.weight": "model-00001-of-00014.safetensors",
900
+ "visual.blocks.18.attn.qkv.bias": "model-00001-of-00014.safetensors",
901
+ "visual.blocks.18.attn.qkv.weight": "model-00001-of-00014.safetensors",
902
+ "visual.blocks.18.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
903
+ "visual.blocks.18.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
904
+ "visual.blocks.18.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
905
+ "visual.blocks.18.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
906
+ "visual.blocks.18.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
907
+ "visual.blocks.18.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
908
+ "visual.blocks.18.norm1.weight": "model-00001-of-00014.safetensors",
909
+ "visual.blocks.18.norm2.weight": "model-00001-of-00014.safetensors",
910
+ "visual.blocks.19.attn.proj.bias": "model-00001-of-00014.safetensors",
911
+ "visual.blocks.19.attn.proj.weight": "model-00001-of-00014.safetensors",
912
+ "visual.blocks.19.attn.qkv.bias": "model-00001-of-00014.safetensors",
913
+ "visual.blocks.19.attn.qkv.weight": "model-00001-of-00014.safetensors",
914
+ "visual.blocks.19.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
915
+ "visual.blocks.19.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
916
+ "visual.blocks.19.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
917
+ "visual.blocks.19.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
918
+ "visual.blocks.19.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
919
+ "visual.blocks.19.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
920
+ "visual.blocks.19.norm1.weight": "model-00001-of-00014.safetensors",
921
+ "visual.blocks.19.norm2.weight": "model-00001-of-00014.safetensors",
922
+ "visual.blocks.2.attn.proj.bias": "model-00001-of-00014.safetensors",
923
+ "visual.blocks.2.attn.proj.weight": "model-00001-of-00014.safetensors",
924
+ "visual.blocks.2.attn.qkv.bias": "model-00001-of-00014.safetensors",
925
+ "visual.blocks.2.attn.qkv.weight": "model-00001-of-00014.safetensors",
926
+ "visual.blocks.2.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
927
+ "visual.blocks.2.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
928
+ "visual.blocks.2.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
929
+ "visual.blocks.2.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
930
+ "visual.blocks.2.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
931
+ "visual.blocks.2.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
932
+ "visual.blocks.2.norm1.weight": "model-00001-of-00014.safetensors",
933
+ "visual.blocks.2.norm2.weight": "model-00001-of-00014.safetensors",
934
+ "visual.blocks.20.attn.proj.bias": "model-00001-of-00014.safetensors",
935
+ "visual.blocks.20.attn.proj.weight": "model-00001-of-00014.safetensors",
936
+ "visual.blocks.20.attn.qkv.bias": "model-00001-of-00014.safetensors",
937
+ "visual.blocks.20.attn.qkv.weight": "model-00001-of-00014.safetensors",
938
+ "visual.blocks.20.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
939
+ "visual.blocks.20.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
940
+ "visual.blocks.20.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
941
+ "visual.blocks.20.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
942
+ "visual.blocks.20.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
943
+ "visual.blocks.20.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
944
+ "visual.blocks.20.norm1.weight": "model-00001-of-00014.safetensors",
945
+ "visual.blocks.20.norm2.weight": "model-00001-of-00014.safetensors",
946
+ "visual.blocks.21.attn.proj.bias": "model-00001-of-00014.safetensors",
947
+ "visual.blocks.21.attn.proj.weight": "model-00001-of-00014.safetensors",
948
+ "visual.blocks.21.attn.qkv.bias": "model-00001-of-00014.safetensors",
949
+ "visual.blocks.21.attn.qkv.weight": "model-00001-of-00014.safetensors",
950
+ "visual.blocks.21.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
951
+ "visual.blocks.21.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
952
+ "visual.blocks.21.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
953
+ "visual.blocks.21.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
954
+ "visual.blocks.21.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
955
+ "visual.blocks.21.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
956
+ "visual.blocks.21.norm1.weight": "model-00001-of-00014.safetensors",
957
+ "visual.blocks.21.norm2.weight": "model-00001-of-00014.safetensors",
958
+ "visual.blocks.22.attn.proj.bias": "model-00001-of-00014.safetensors",
959
+ "visual.blocks.22.attn.proj.weight": "model-00001-of-00014.safetensors",
960
+ "visual.blocks.22.attn.qkv.bias": "model-00001-of-00014.safetensors",
961
+ "visual.blocks.22.attn.qkv.weight": "model-00001-of-00014.safetensors",
962
+ "visual.blocks.22.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
963
+ "visual.blocks.22.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
964
+ "visual.blocks.22.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
965
+ "visual.blocks.22.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
966
+ "visual.blocks.22.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
967
+ "visual.blocks.22.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
968
+ "visual.blocks.22.norm1.weight": "model-00001-of-00014.safetensors",
969
+ "visual.blocks.22.norm2.weight": "model-00001-of-00014.safetensors",
970
+ "visual.blocks.23.attn.proj.bias": "model-00001-of-00014.safetensors",
971
+ "visual.blocks.23.attn.proj.weight": "model-00001-of-00014.safetensors",
972
+ "visual.blocks.23.attn.qkv.bias": "model-00001-of-00014.safetensors",
973
+ "visual.blocks.23.attn.qkv.weight": "model-00001-of-00014.safetensors",
974
+ "visual.blocks.23.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
975
+ "visual.blocks.23.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
976
+ "visual.blocks.23.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
977
+ "visual.blocks.23.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
978
+ "visual.blocks.23.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
979
+ "visual.blocks.23.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
980
+ "visual.blocks.23.norm1.weight": "model-00001-of-00014.safetensors",
981
+ "visual.blocks.23.norm2.weight": "model-00001-of-00014.safetensors",
982
+ "visual.blocks.24.attn.proj.bias": "model-00001-of-00014.safetensors",
983
+ "visual.blocks.24.attn.proj.weight": "model-00001-of-00014.safetensors",
984
+ "visual.blocks.24.attn.qkv.bias": "model-00001-of-00014.safetensors",
985
+ "visual.blocks.24.attn.qkv.weight": "model-00001-of-00014.safetensors",
986
+ "visual.blocks.24.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
987
+ "visual.blocks.24.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
988
+ "visual.blocks.24.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
989
+ "visual.blocks.24.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
990
+ "visual.blocks.24.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
991
+ "visual.blocks.24.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
992
+ "visual.blocks.24.norm1.weight": "model-00001-of-00014.safetensors",
993
+ "visual.blocks.24.norm2.weight": "model-00001-of-00014.safetensors",
994
+ "visual.blocks.25.attn.proj.bias": "model-00001-of-00014.safetensors",
995
+ "visual.blocks.25.attn.proj.weight": "model-00001-of-00014.safetensors",
996
+ "visual.blocks.25.attn.qkv.bias": "model-00001-of-00014.safetensors",
997
+ "visual.blocks.25.attn.qkv.weight": "model-00001-of-00014.safetensors",
998
+ "visual.blocks.25.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
999
+ "visual.blocks.25.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1000
+ "visual.blocks.25.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1001
+ "visual.blocks.25.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1002
+ "visual.blocks.25.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1003
+ "visual.blocks.25.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1004
+ "visual.blocks.25.norm1.weight": "model-00001-of-00014.safetensors",
1005
+ "visual.blocks.25.norm2.weight": "model-00001-of-00014.safetensors",
1006
+ "visual.blocks.26.attn.proj.bias": "model-00001-of-00014.safetensors",
1007
+ "visual.blocks.26.attn.proj.weight": "model-00001-of-00014.safetensors",
1008
+ "visual.blocks.26.attn.qkv.bias": "model-00001-of-00014.safetensors",
1009
+ "visual.blocks.26.attn.qkv.weight": "model-00001-of-00014.safetensors",
1010
+ "visual.blocks.26.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1011
+ "visual.blocks.26.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1012
+ "visual.blocks.26.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1013
+ "visual.blocks.26.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1014
+ "visual.blocks.26.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1015
+ "visual.blocks.26.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1016
+ "visual.blocks.26.norm1.weight": "model-00001-of-00014.safetensors",
1017
+ "visual.blocks.26.norm2.weight": "model-00001-of-00014.safetensors",
1018
+ "visual.blocks.27.attn.proj.bias": "model-00001-of-00014.safetensors",
1019
+ "visual.blocks.27.attn.proj.weight": "model-00001-of-00014.safetensors",
1020
+ "visual.blocks.27.attn.qkv.bias": "model-00001-of-00014.safetensors",
1021
+ "visual.blocks.27.attn.qkv.weight": "model-00001-of-00014.safetensors",
1022
+ "visual.blocks.27.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1023
+ "visual.blocks.27.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1024
+ "visual.blocks.27.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1025
+ "visual.blocks.27.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1026
+ "visual.blocks.27.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1027
+ "visual.blocks.27.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1028
+ "visual.blocks.27.norm1.weight": "model-00001-of-00014.safetensors",
1029
+ "visual.blocks.27.norm2.weight": "model-00001-of-00014.safetensors",
1030
+ "visual.blocks.28.attn.proj.bias": "model-00001-of-00014.safetensors",
1031
+ "visual.blocks.28.attn.proj.weight": "model-00001-of-00014.safetensors",
1032
+ "visual.blocks.28.attn.qkv.bias": "model-00001-of-00014.safetensors",
1033
+ "visual.blocks.28.attn.qkv.weight": "model-00001-of-00014.safetensors",
1034
+ "visual.blocks.28.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1035
+ "visual.blocks.28.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1036
+ "visual.blocks.28.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1037
+ "visual.blocks.28.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1038
+ "visual.blocks.28.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1039
+ "visual.blocks.28.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1040
+ "visual.blocks.28.norm1.weight": "model-00001-of-00014.safetensors",
1041
+ "visual.blocks.28.norm2.weight": "model-00001-of-00014.safetensors",
1042
+ "visual.blocks.29.attn.proj.bias": "model-00001-of-00014.safetensors",
1043
+ "visual.blocks.29.attn.proj.weight": "model-00001-of-00014.safetensors",
1044
+ "visual.blocks.29.attn.qkv.bias": "model-00001-of-00014.safetensors",
1045
+ "visual.blocks.29.attn.qkv.weight": "model-00001-of-00014.safetensors",
1046
+ "visual.blocks.29.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1047
+ "visual.blocks.29.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1048
+ "visual.blocks.29.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1049
+ "visual.blocks.29.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1050
+ "visual.blocks.29.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1051
+ "visual.blocks.29.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1052
+ "visual.blocks.29.norm1.weight": "model-00001-of-00014.safetensors",
1053
+ "visual.blocks.29.norm2.weight": "model-00001-of-00014.safetensors",
1054
+ "visual.blocks.3.attn.proj.bias": "model-00001-of-00014.safetensors",
1055
+ "visual.blocks.3.attn.proj.weight": "model-00001-of-00014.safetensors",
1056
+ "visual.blocks.3.attn.qkv.bias": "model-00001-of-00014.safetensors",
1057
+ "visual.blocks.3.attn.qkv.weight": "model-00001-of-00014.safetensors",
1058
+ "visual.blocks.3.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1059
+ "visual.blocks.3.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1060
+ "visual.blocks.3.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1061
+ "visual.blocks.3.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1062
+ "visual.blocks.3.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1063
+ "visual.blocks.3.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1064
+ "visual.blocks.3.norm1.weight": "model-00001-of-00014.safetensors",
1065
+ "visual.blocks.3.norm2.weight": "model-00001-of-00014.safetensors",
1066
+ "visual.blocks.30.attn.proj.bias": "model-00001-of-00014.safetensors",
1067
+ "visual.blocks.30.attn.proj.weight": "model-00001-of-00014.safetensors",
1068
+ "visual.blocks.30.attn.qkv.bias": "model-00001-of-00014.safetensors",
1069
+ "visual.blocks.30.attn.qkv.weight": "model-00001-of-00014.safetensors",
1070
+ "visual.blocks.30.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1071
+ "visual.blocks.30.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1072
+ "visual.blocks.30.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1073
+ "visual.blocks.30.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1074
+ "visual.blocks.30.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1075
+ "visual.blocks.30.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1076
+ "visual.blocks.30.norm1.weight": "model-00001-of-00014.safetensors",
1077
+ "visual.blocks.30.norm2.weight": "model-00001-of-00014.safetensors",
1078
+ "visual.blocks.31.attn.proj.bias": "model-00001-of-00014.safetensors",
1079
+ "visual.blocks.31.attn.proj.weight": "model-00001-of-00014.safetensors",
1080
+ "visual.blocks.31.attn.qkv.bias": "model-00001-of-00014.safetensors",
1081
+ "visual.blocks.31.attn.qkv.weight": "model-00001-of-00014.safetensors",
1082
+ "visual.blocks.31.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1083
+ "visual.blocks.31.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1084
+ "visual.blocks.31.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1085
+ "visual.blocks.31.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1086
+ "visual.blocks.31.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1087
+ "visual.blocks.31.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1088
+ "visual.blocks.31.norm1.weight": "model-00001-of-00014.safetensors",
1089
+ "visual.blocks.31.norm2.weight": "model-00001-of-00014.safetensors",
1090
+ "visual.blocks.4.attn.proj.bias": "model-00001-of-00014.safetensors",
1091
+ "visual.blocks.4.attn.proj.weight": "model-00001-of-00014.safetensors",
1092
+ "visual.blocks.4.attn.qkv.bias": "model-00001-of-00014.safetensors",
1093
+ "visual.blocks.4.attn.qkv.weight": "model-00001-of-00014.safetensors",
1094
+ "visual.blocks.4.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1095
+ "visual.blocks.4.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1096
+ "visual.blocks.4.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1097
+ "visual.blocks.4.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1098
+ "visual.blocks.4.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1099
+ "visual.blocks.4.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1100
+ "visual.blocks.4.norm1.weight": "model-00001-of-00014.safetensors",
1101
+ "visual.blocks.4.norm2.weight": "model-00001-of-00014.safetensors",
1102
+ "visual.blocks.5.attn.proj.bias": "model-00001-of-00014.safetensors",
1103
+ "visual.blocks.5.attn.proj.weight": "model-00001-of-00014.safetensors",
1104
+ "visual.blocks.5.attn.qkv.bias": "model-00001-of-00014.safetensors",
1105
+ "visual.blocks.5.attn.qkv.weight": "model-00001-of-00014.safetensors",
1106
+ "visual.blocks.5.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1107
+ "visual.blocks.5.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1108
+ "visual.blocks.5.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1109
+ "visual.blocks.5.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1110
+ "visual.blocks.5.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1111
+ "visual.blocks.5.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1112
+ "visual.blocks.5.norm1.weight": "model-00001-of-00014.safetensors",
1113
+ "visual.blocks.5.norm2.weight": "model-00001-of-00014.safetensors",
1114
+ "visual.blocks.6.attn.proj.bias": "model-00001-of-00014.safetensors",
1115
+ "visual.blocks.6.attn.proj.weight": "model-00001-of-00014.safetensors",
1116
+ "visual.blocks.6.attn.qkv.bias": "model-00001-of-00014.safetensors",
1117
+ "visual.blocks.6.attn.qkv.weight": "model-00001-of-00014.safetensors",
1118
+ "visual.blocks.6.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1119
+ "visual.blocks.6.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1120
+ "visual.blocks.6.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1121
+ "visual.blocks.6.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1122
+ "visual.blocks.6.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1123
+ "visual.blocks.6.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1124
+ "visual.blocks.6.norm1.weight": "model-00001-of-00014.safetensors",
1125
+ "visual.blocks.6.norm2.weight": "model-00001-of-00014.safetensors",
1126
+ "visual.blocks.7.attn.proj.bias": "model-00001-of-00014.safetensors",
1127
+ "visual.blocks.7.attn.proj.weight": "model-00001-of-00014.safetensors",
1128
+ "visual.blocks.7.attn.qkv.bias": "model-00001-of-00014.safetensors",
1129
+ "visual.blocks.7.attn.qkv.weight": "model-00001-of-00014.safetensors",
1130
+ "visual.blocks.7.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1131
+ "visual.blocks.7.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1132
+ "visual.blocks.7.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1133
+ "visual.blocks.7.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1134
+ "visual.blocks.7.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1135
+ "visual.blocks.7.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1136
+ "visual.blocks.7.norm1.weight": "model-00001-of-00014.safetensors",
1137
+ "visual.blocks.7.norm2.weight": "model-00001-of-00014.safetensors",
1138
+ "visual.blocks.8.attn.proj.bias": "model-00001-of-00014.safetensors",
1139
+ "visual.blocks.8.attn.proj.weight": "model-00001-of-00014.safetensors",
1140
+ "visual.blocks.8.attn.qkv.bias": "model-00001-of-00014.safetensors",
1141
+ "visual.blocks.8.attn.qkv.weight": "model-00001-of-00014.safetensors",
1142
+ "visual.blocks.8.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1143
+ "visual.blocks.8.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1144
+ "visual.blocks.8.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1145
+ "visual.blocks.8.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1146
+ "visual.blocks.8.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1147
+ "visual.blocks.8.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1148
+ "visual.blocks.8.norm1.weight": "model-00001-of-00014.safetensors",
1149
+ "visual.blocks.8.norm2.weight": "model-00001-of-00014.safetensors",
1150
+ "visual.blocks.9.attn.proj.bias": "model-00001-of-00014.safetensors",
1151
+ "visual.blocks.9.attn.proj.weight": "model-00001-of-00014.safetensors",
1152
+ "visual.blocks.9.attn.qkv.bias": "model-00001-of-00014.safetensors",
1153
+ "visual.blocks.9.attn.qkv.weight": "model-00001-of-00014.safetensors",
1154
+ "visual.blocks.9.mlp.down_proj.bias": "model-00001-of-00014.safetensors",
1155
+ "visual.blocks.9.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
1156
+ "visual.blocks.9.mlp.gate_proj.bias": "model-00001-of-00014.safetensors",
1157
+ "visual.blocks.9.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
1158
+ "visual.blocks.9.mlp.up_proj.bias": "model-00001-of-00014.safetensors",
1159
+ "visual.blocks.9.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
1160
+ "visual.blocks.9.norm1.weight": "model-00001-of-00014.safetensors",
1161
+ "visual.blocks.9.norm2.weight": "model-00001-of-00014.safetensors",
1162
+ "visual.merger.ln_q.weight": "model-00001-of-00014.safetensors",
1163
+ "visual.merger.mlp.0.bias": "model-00001-of-00014.safetensors",
1164
+ "visual.merger.mlp.0.weight": "model-00001-of-00014.safetensors",
1165
+ "visual.merger.mlp.2.bias": "model-00001-of-00014.safetensors",
1166
+ "visual.merger.mlp.2.weight": "model-00001-of-00014.safetensors",
1167
+ "visual.patch_embed.proj.weight": "model-00001-of-00014.safetensors"
1168
+ }
1169
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": null,
3
+ "data_format": "channels_first",
4
+ "default_to_square": true,
5
+ "device": null,
6
+ "disable_grouping": null,
7
+ "do_center_crop": null,
8
+ "do_convert_rgb": true,
9
+ "do_normalize": true,
10
+ "do_rescale": true,
11
+ "do_resize": true,
12
+ "image_mean": [
13
+ 0.48145466,
14
+ 0.4578275,
15
+ 0.40821073
16
+ ],
17
+ "image_processor_type": "Qwen2VLImageProcessorFast",
18
+ "image_std": [
19
+ 0.26862954,
20
+ 0.26130258,
21
+ 0.27577711
22
+ ],
23
+ "input_data_format": null,
24
+ "max_pixels": 12845056,
25
+ "merge_size": 2,
26
+ "min_pixels": 3136,
27
+ "patch_size": 14,
28
+ "processor_class": "Qwen2_5_VLProcessor",
29
+ "resample": 3,
30
+ "rescale_factor": 0.00392156862745098,
31
+ "return_tensors": null,
32
+ "size": {
33
+ "longest_edge": 12845056,
34
+ "shortest_edge": 3136
35
+ },
36
+ "temporal_patch_size": 2
37
+ }
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75881bc0e709c509b10f83b89a79347ce9120d3eec42882604a1ac7baa572563
3
+ size 1465
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
+ }
trainer_state.json ADDED
@@ -0,0 +1,2274 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0110567051018797,
6
+ "eval_steps": 500,
7
+ "global_step": 3200,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.00315905860053704,
14
+ "grad_norm": 4.247729183572734,
15
+ "learning_rate": 5.685407454200885e-08,
16
+ "loss": 0.6057,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.00631811720107408,
21
+ "grad_norm": 4.531449453000221,
22
+ "learning_rate": 1.2002526847757423e-07,
23
+ "loss": 0.6101,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.00947717580161112,
28
+ "grad_norm": 4.104292927990717,
29
+ "learning_rate": 1.831964624131396e-07,
30
+ "loss": 0.5947,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.01263623440214816,
35
+ "grad_norm": 3.1655981231206747,
36
+ "learning_rate": 2.46367656348705e-07,
37
+ "loss": 0.5609,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.0157952930026852,
42
+ "grad_norm": 1.6081249047797948,
43
+ "learning_rate": 3.095388502842704e-07,
44
+ "loss": 0.5128,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.01895435160322224,
49
+ "grad_norm": 1.0896969313030245,
50
+ "learning_rate": 3.727100442198358e-07,
51
+ "loss": 0.4618,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.02211341020375928,
56
+ "grad_norm": 0.720186068205468,
57
+ "learning_rate": 4.3588123815540116e-07,
58
+ "loss": 0.437,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.02527246880429632,
63
+ "grad_norm": 0.5712683718673494,
64
+ "learning_rate": 4.990524320909665e-07,
65
+ "loss": 0.4145,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.02843152740483336,
70
+ "grad_norm": 0.41839125700043645,
71
+ "learning_rate": 5.62223626026532e-07,
72
+ "loss": 0.3887,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.0315905860053704,
77
+ "grad_norm": 0.33539615628708846,
78
+ "learning_rate": 6.253948199620974e-07,
79
+ "loss": 0.3661,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.03474964460590744,
84
+ "grad_norm": 0.25866894040378646,
85
+ "learning_rate": 6.885660138976627e-07,
86
+ "loss": 0.352,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.03790870320644448,
91
+ "grad_norm": 0.24295153980385867,
92
+ "learning_rate": 7.517372078332281e-07,
93
+ "loss": 0.3457,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.04106776180698152,
98
+ "grad_norm": 0.28237511864322024,
99
+ "learning_rate": 8.149084017687935e-07,
100
+ "loss": 0.3345,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.04422682040751856,
105
+ "grad_norm": 0.21156218948831834,
106
+ "learning_rate": 8.780795957043589e-07,
107
+ "loss": 0.3281,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.0473858790080556,
112
+ "grad_norm": 0.20112450248918143,
113
+ "learning_rate": 9.412507896399242e-07,
114
+ "loss": 0.3267,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.05054493760859264,
119
+ "grad_norm": 0.19763198530161658,
120
+ "learning_rate": 1.0044219835754897e-06,
121
+ "loss": 0.3206,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.05370399620912968,
126
+ "grad_norm": 0.21467283595984932,
127
+ "learning_rate": 1.067593177511055e-06,
128
+ "loss": 0.3136,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.05686305480966672,
133
+ "grad_norm": 0.18660741459520747,
134
+ "learning_rate": 1.1307643714466204e-06,
135
+ "loss": 0.3087,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.06002211341020376,
140
+ "grad_norm": 0.21253735556196968,
141
+ "learning_rate": 1.1939355653821858e-06,
142
+ "loss": 0.3049,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.0631811720107408,
147
+ "grad_norm": 0.237093634634235,
148
+ "learning_rate": 1.2571067593177513e-06,
149
+ "loss": 0.3005,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.06634023061127783,
154
+ "grad_norm": 0.2231876630708564,
155
+ "learning_rate": 1.3202779532533167e-06,
156
+ "loss": 0.2972,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.06949928921181488,
161
+ "grad_norm": 0.2017070855177483,
162
+ "learning_rate": 1.383449147188882e-06,
163
+ "loss": 0.2935,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.07265834781235192,
168
+ "grad_norm": 0.21789091247744216,
169
+ "learning_rate": 1.4466203411244472e-06,
170
+ "loss": 0.29,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.07581740641288896,
175
+ "grad_norm": 0.24210256044037484,
176
+ "learning_rate": 1.509791535060013e-06,
177
+ "loss": 0.289,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.07897646501342599,
182
+ "grad_norm": 0.21139319424520517,
183
+ "learning_rate": 1.572962728995578e-06,
184
+ "loss": 0.2895,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.08213552361396304,
189
+ "grad_norm": 0.20081643445857786,
190
+ "learning_rate": 1.6361339229311434e-06,
191
+ "loss": 0.2848,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.08529458221450008,
196
+ "grad_norm": 0.21867950206393189,
197
+ "learning_rate": 1.6993051168667088e-06,
198
+ "loss": 0.2842,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.08845364081503712,
203
+ "grad_norm": 0.1996646570779461,
204
+ "learning_rate": 1.7624763108022743e-06,
205
+ "loss": 0.2785,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.09161269941557416,
210
+ "grad_norm": 0.1991696950635834,
211
+ "learning_rate": 1.8256475047378397e-06,
212
+ "loss": 0.2747,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.0947717580161112,
217
+ "grad_norm": 0.21382346420276455,
218
+ "learning_rate": 1.888818698673405e-06,
219
+ "loss": 0.2778,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.09793081661664824,
224
+ "grad_norm": 0.22163335945192605,
225
+ "learning_rate": 1.9519898926089704e-06,
226
+ "loss": 0.2747,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.10108987521718528,
231
+ "grad_norm": 0.2338176849443747,
232
+ "learning_rate": 2.0151610865445357e-06,
233
+ "loss": 0.2719,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.10424893381772232,
238
+ "grad_norm": 0.18002603770625134,
239
+ "learning_rate": 2.078332280480101e-06,
240
+ "loss": 0.2713,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.10740799241825937,
245
+ "grad_norm": 0.19849866049815784,
246
+ "learning_rate": 2.1415034744156664e-06,
247
+ "loss": 0.2671,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.1105670510187964,
252
+ "grad_norm": 0.18726127817800872,
253
+ "learning_rate": 2.2046746683512322e-06,
254
+ "loss": 0.2679,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.11372610961933344,
259
+ "grad_norm": 0.18164666420060355,
260
+ "learning_rate": 2.2678458622867976e-06,
261
+ "loss": 0.2696,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.11688516821987048,
266
+ "grad_norm": 0.188419554987382,
267
+ "learning_rate": 2.331017056222363e-06,
268
+ "loss": 0.2657,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.12004422682040752,
273
+ "grad_norm": 0.21408765943095384,
274
+ "learning_rate": 2.394188250157928e-06,
275
+ "loss": 0.2631,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.12320328542094455,
280
+ "grad_norm": 0.1728133517187042,
281
+ "learning_rate": 2.4573594440934936e-06,
282
+ "loss": 0.2627,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.1263623440214816,
287
+ "grad_norm": 0.20446817453058305,
288
+ "learning_rate": 2.520530638029059e-06,
289
+ "loss": 0.2633,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.12952140262201864,
294
+ "grad_norm": 0.2867541071966156,
295
+ "learning_rate": 2.5837018319646247e-06,
296
+ "loss": 0.2606,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.13268046122255567,
301
+ "grad_norm": 0.19615186721360886,
302
+ "learning_rate": 2.6468730259001897e-06,
303
+ "loss": 0.2589,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.13583951982309272,
308
+ "grad_norm": 0.17250843271916932,
309
+ "learning_rate": 2.710044219835755e-06,
310
+ "loss": 0.2578,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.13899857842362975,
315
+ "grad_norm": 0.21061094083012397,
316
+ "learning_rate": 2.7732154137713208e-06,
317
+ "loss": 0.256,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.1421576370241668,
322
+ "grad_norm": 0.16796761706106636,
323
+ "learning_rate": 2.8363866077068857e-06,
324
+ "loss": 0.254,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.14531669562470384,
329
+ "grad_norm": 0.20087509437202522,
330
+ "learning_rate": 2.8995578016424515e-06,
331
+ "loss": 0.2554,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.14847575422524087,
336
+ "grad_norm": 0.20431208429215614,
337
+ "learning_rate": 2.9627289955780164e-06,
338
+ "loss": 0.2524,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.15163481282577793,
343
+ "grad_norm": 0.20436652455880414,
344
+ "learning_rate": 3.025900189513582e-06,
345
+ "loss": 0.2557,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.15479387142631496,
350
+ "grad_norm": 0.19923295760265683,
351
+ "learning_rate": 3.0890713834491475e-06,
352
+ "loss": 0.252,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.15795293002685198,
357
+ "grad_norm": 0.1925405706241213,
358
+ "learning_rate": 3.1522425773847125e-06,
359
+ "loss": 0.2505,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.16111198862738904,
364
+ "grad_norm": 0.20206673704458114,
365
+ "learning_rate": 3.2154137713202782e-06,
366
+ "loss": 0.2501,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.16427104722792607,
371
+ "grad_norm": 0.24333700245600473,
372
+ "learning_rate": 3.278584965255844e-06,
373
+ "loss": 0.2464,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.16743010582846313,
378
+ "grad_norm": 0.18205244482695793,
379
+ "learning_rate": 3.341756159191409e-06,
380
+ "loss": 0.2485,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.17058916442900016,
385
+ "grad_norm": 0.18264545298219137,
386
+ "learning_rate": 3.4049273531269743e-06,
387
+ "loss": 0.2456,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.17374822302953719,
392
+ "grad_norm": 0.17873652388985004,
393
+ "learning_rate": 3.46809854706254e-06,
394
+ "loss": 0.248,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.17690728163007424,
399
+ "grad_norm": 0.17062239190578055,
400
+ "learning_rate": 3.531269740998105e-06,
401
+ "loss": 0.2449,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.18006634023061127,
406
+ "grad_norm": 0.19143037375033525,
407
+ "learning_rate": 3.5944409349336708e-06,
408
+ "loss": 0.2464,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.18322539883114833,
413
+ "grad_norm": 0.1861722136519532,
414
+ "learning_rate": 3.6576121288692357e-06,
415
+ "loss": 0.2489,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.18638445743168536,
420
+ "grad_norm": 0.1680077751454427,
421
+ "learning_rate": 3.7207833228048014e-06,
422
+ "loss": 0.2433,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.1895435160322224,
427
+ "grad_norm": 0.1827874982767101,
428
+ "learning_rate": 3.783954516740367e-06,
429
+ "loss": 0.2414,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.19270257463275944,
434
+ "grad_norm": 0.17739754432609572,
435
+ "learning_rate": 3.847125710675932e-06,
436
+ "loss": 0.2439,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.19586163323329647,
441
+ "grad_norm": 0.18166974193276042,
442
+ "learning_rate": 3.910296904611497e-06,
443
+ "loss": 0.2427,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.19902069183383353,
448
+ "grad_norm": 0.23535305540149423,
449
+ "learning_rate": 3.973468098547063e-06,
450
+ "loss": 0.2397,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.20217975043437056,
455
+ "grad_norm": 0.17950832267537836,
456
+ "learning_rate": 4.036639292482628e-06,
457
+ "loss": 0.2413,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.2053388090349076,
462
+ "grad_norm": 0.1871436271310335,
463
+ "learning_rate": 4.099810486418194e-06,
464
+ "loss": 0.2392,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.20849786763544464,
469
+ "grad_norm": 0.18940758631882895,
470
+ "learning_rate": 4.162981680353759e-06,
471
+ "loss": 0.239,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.21165692623598167,
476
+ "grad_norm": 0.18679899786034626,
477
+ "learning_rate": 4.226152874289325e-06,
478
+ "loss": 0.2394,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.21481598483651873,
483
+ "grad_norm": 0.2005980889421409,
484
+ "learning_rate": 4.28932406822489e-06,
485
+ "loss": 0.2369,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.21797504343705576,
490
+ "grad_norm": 0.18654971719873092,
491
+ "learning_rate": 4.3524952621604545e-06,
492
+ "loss": 0.2372,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.2211341020375928,
497
+ "grad_norm": 0.19896341390367112,
498
+ "learning_rate": 4.415666456096021e-06,
499
+ "loss": 0.2384,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.22429316063812985,
504
+ "grad_norm": 0.1824518487303919,
505
+ "learning_rate": 4.478837650031586e-06,
506
+ "loss": 0.2322,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.22745221923866688,
511
+ "grad_norm": 0.19030275242513275,
512
+ "learning_rate": 4.542008843967151e-06,
513
+ "loss": 0.2373,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.2306112778392039,
518
+ "grad_norm": 0.18760298890701887,
519
+ "learning_rate": 4.605180037902717e-06,
520
+ "loss": 0.2345,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.23377033643974096,
525
+ "grad_norm": 0.15845613302953646,
526
+ "learning_rate": 4.668351231838282e-06,
527
+ "loss": 0.2343,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.236929395040278,
532
+ "grad_norm": 0.1710181463737659,
533
+ "learning_rate": 4.7315224257738475e-06,
534
+ "loss": 0.2318,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.24008845364081505,
539
+ "grad_norm": 0.23880301915969937,
540
+ "learning_rate": 4.794693619709413e-06,
541
+ "loss": 0.2339,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.24324751224135208,
546
+ "grad_norm": 0.1678386308387805,
547
+ "learning_rate": 4.857864813644978e-06,
548
+ "loss": 0.2306,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.2464065708418891,
553
+ "grad_norm": 0.1671346172529239,
554
+ "learning_rate": 4.9210360075805435e-06,
555
+ "loss": 0.2333,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.24956562944242616,
560
+ "grad_norm": 0.19944596405981949,
561
+ "learning_rate": 4.984207201516109e-06,
562
+ "loss": 0.2313,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.2527246880429632,
567
+ "grad_norm": 0.18321786541799437,
568
+ "learning_rate": 5.047378395451674e-06,
569
+ "loss": 0.2321,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.2558837466435002,
574
+ "grad_norm": 0.16781254221044775,
575
+ "learning_rate": 5.11054958938724e-06,
576
+ "loss": 0.2313,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.2590428052440373,
581
+ "grad_norm": 0.17492747555934146,
582
+ "learning_rate": 5.173720783322806e-06,
583
+ "loss": 0.2312,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.26220186384457433,
588
+ "grad_norm": 0.17965819326676355,
589
+ "learning_rate": 5.23689197725837e-06,
590
+ "loss": 0.2285,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.26536092244511134,
595
+ "grad_norm": 0.16804749386309387,
596
+ "learning_rate": 5.3000631711939365e-06,
597
+ "loss": 0.2282,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.2685199810456484,
602
+ "grad_norm": 0.1723473524041935,
603
+ "learning_rate": 5.363234365129502e-06,
604
+ "loss": 0.2296,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.27167903964618545,
609
+ "grad_norm": 0.20773827836779976,
610
+ "learning_rate": 5.426405559065066e-06,
611
+ "loss": 0.2273,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.27483809824672245,
616
+ "grad_norm": 0.1641196545454829,
617
+ "learning_rate": 5.489576753000632e-06,
618
+ "loss": 0.2301,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.2779971568472595,
623
+ "grad_norm": 0.19323437206655858,
624
+ "learning_rate": 5.552747946936198e-06,
625
+ "loss": 0.2249,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.28115621544779656,
630
+ "grad_norm": 0.18267311407058162,
631
+ "learning_rate": 5.615919140871763e-06,
632
+ "loss": 0.227,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.2843152740483336,
637
+ "grad_norm": 0.19801209807316134,
638
+ "learning_rate": 5.679090334807328e-06,
639
+ "loss": 0.2273,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.2874743326488706,
644
+ "grad_norm": 0.17413734456766244,
645
+ "learning_rate": 5.742261528742894e-06,
646
+ "loss": 0.2267,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.2906333912494077,
651
+ "grad_norm": 0.21136423644297928,
652
+ "learning_rate": 5.805432722678459e-06,
653
+ "loss": 0.2263,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.29379244984994474,
658
+ "grad_norm": 0.20233679461777548,
659
+ "learning_rate": 5.868603916614024e-06,
660
+ "loss": 0.2263,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.29695150845048174,
665
+ "grad_norm": 0.19521629273254087,
666
+ "learning_rate": 5.93177511054959e-06,
667
+ "loss": 0.2269,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.3001105670510188,
672
+ "grad_norm": 0.1696883052949813,
673
+ "learning_rate": 5.994946304485155e-06,
674
+ "loss": 0.2265,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.30326962565155585,
679
+ "grad_norm": 0.18758477411326932,
680
+ "learning_rate": 6.058117498420721e-06,
681
+ "loss": 0.2247,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.30642868425209285,
686
+ "grad_norm": 0.1897879092670083,
687
+ "learning_rate": 6.121288692356287e-06,
688
+ "loss": 0.2282,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.3095877428526299,
693
+ "grad_norm": 0.18304573332433055,
694
+ "learning_rate": 6.184459886291851e-06,
695
+ "loss": 0.2223,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.31274680145316697,
700
+ "grad_norm": 0.1782561802917822,
701
+ "learning_rate": 6.247631080227417e-06,
702
+ "loss": 0.224,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.31590586005370397,
707
+ "grad_norm": 0.20741862373031386,
708
+ "learning_rate": 6.310802274162983e-06,
709
+ "loss": 0.2219,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.319064918654241,
714
+ "grad_norm": 0.1539256973132976,
715
+ "learning_rate": 6.373973468098547e-06,
716
+ "loss": 0.2207,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.3222239772547781,
721
+ "grad_norm": 0.16875500034138524,
722
+ "learning_rate": 6.437144662034113e-06,
723
+ "loss": 0.2225,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.32538303585531514,
728
+ "grad_norm": 0.17168781652376466,
729
+ "learning_rate": 6.500315855969679e-06,
730
+ "loss": 0.22,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.32854209445585214,
735
+ "grad_norm": 0.18011584935641983,
736
+ "learning_rate": 6.563487049905244e-06,
737
+ "loss": 0.2216,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.3317011530563892,
742
+ "grad_norm": 0.19414711454837585,
743
+ "learning_rate": 6.626658243840809e-06,
744
+ "loss": 0.2249,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.33486021165692625,
749
+ "grad_norm": 0.2103163638452648,
750
+ "learning_rate": 6.689829437776375e-06,
751
+ "loss": 0.2242,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.33801927025746326,
756
+ "grad_norm": 0.17555158548112104,
757
+ "learning_rate": 6.75300063171194e-06,
758
+ "loss": 0.2206,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.3411783288580003,
763
+ "grad_norm": 0.16163792223112092,
764
+ "learning_rate": 6.816171825647505e-06,
765
+ "loss": 0.221,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.34433738745853737,
770
+ "grad_norm": 0.17047037019662084,
771
+ "learning_rate": 6.87934301958307e-06,
772
+ "loss": 0.2222,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.34749644605907437,
777
+ "grad_norm": 0.18118959670707843,
778
+ "learning_rate": 6.942514213518636e-06,
779
+ "loss": 0.226,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.35065550465961143,
784
+ "grad_norm": 0.18596851555170507,
785
+ "learning_rate": 7.005685407454202e-06,
786
+ "loss": 0.2208,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.3538145632601485,
791
+ "grad_norm": 0.17540680145260182,
792
+ "learning_rate": 7.068856601389766e-06,
793
+ "loss": 0.2228,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.35697362186068554,
798
+ "grad_norm": 0.16362413537591447,
799
+ "learning_rate": 7.1320277953253324e-06,
800
+ "loss": 0.2199,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.36013268046122254,
805
+ "grad_norm": 0.16388409560788866,
806
+ "learning_rate": 7.195198989260898e-06,
807
+ "loss": 0.2183,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.3632917390617596,
812
+ "grad_norm": 0.16780920971825364,
813
+ "learning_rate": 7.258370183196462e-06,
814
+ "loss": 0.2223,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.36645079766229666,
819
+ "grad_norm": 0.16629698944531449,
820
+ "learning_rate": 7.3215413771320285e-06,
821
+ "loss": 0.2184,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.36960985626283366,
826
+ "grad_norm": 0.1760306249090938,
827
+ "learning_rate": 7.384712571067594e-06,
828
+ "loss": 0.2219,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.3727689148633707,
833
+ "grad_norm": 0.1790776173967007,
834
+ "learning_rate": 7.447883765003159e-06,
835
+ "loss": 0.2198,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.37592797346390777,
840
+ "grad_norm": 0.1857393797937426,
841
+ "learning_rate": 7.5110549589387245e-06,
842
+ "loss": 0.2183,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.3790870320644448,
847
+ "grad_norm": 0.18072716423697788,
848
+ "learning_rate": 7.57422615287429e-06,
849
+ "loss": 0.2198,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.38224609066498183,
854
+ "grad_norm": 0.1627749965589205,
855
+ "learning_rate": 7.637397346809855e-06,
856
+ "loss": 0.2204,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.3854051492655189,
861
+ "grad_norm": 0.23081430880637033,
862
+ "learning_rate": 7.700568540745421e-06,
863
+ "loss": 0.2179,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.3885642078660559,
868
+ "grad_norm": 0.14668683659878062,
869
+ "learning_rate": 7.763739734680986e-06,
870
+ "loss": 0.218,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.39172326646659295,
875
+ "grad_norm": 0.17314703270798587,
876
+ "learning_rate": 7.82691092861655e-06,
877
+ "loss": 0.2172,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.39488232506713,
882
+ "grad_norm": 0.1587830451358659,
883
+ "learning_rate": 7.890082122552117e-06,
884
+ "loss": 0.2183,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.39804138366766706,
889
+ "grad_norm": 0.1581230238900689,
890
+ "learning_rate": 7.953253316487683e-06,
891
+ "loss": 0.2157,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.40120044226820406,
896
+ "grad_norm": 0.15808321097279437,
897
+ "learning_rate": 8.016424510423247e-06,
898
+ "loss": 0.2152,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.4043595008687411,
903
+ "grad_norm": 0.17727435096583632,
904
+ "learning_rate": 8.079595704358814e-06,
905
+ "loss": 0.2169,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.4075185594692782,
910
+ "grad_norm": 0.16825167940141256,
911
+ "learning_rate": 8.142766898294378e-06,
912
+ "loss": 0.2143,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.4106776180698152,
917
+ "grad_norm": 0.1623055799684783,
918
+ "learning_rate": 8.205938092229944e-06,
919
+ "loss": 0.214,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.41383667667035223,
924
+ "grad_norm": 0.16866092597564897,
925
+ "learning_rate": 8.269109286165509e-06,
926
+ "loss": 0.2189,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.4169957352708893,
931
+ "grad_norm": 0.15469945071292018,
932
+ "learning_rate": 8.332280480101075e-06,
933
+ "loss": 0.2147,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.4201547938714263,
938
+ "grad_norm": 0.15496524048072485,
939
+ "learning_rate": 8.39545167403664e-06,
940
+ "loss": 0.2155,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.42331385247196335,
945
+ "grad_norm": 0.1607478129687131,
946
+ "learning_rate": 8.458622867972206e-06,
947
+ "loss": 0.2142,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.4264729110725004,
952
+ "grad_norm": 0.16297866728767108,
953
+ "learning_rate": 8.521794061907772e-06,
954
+ "loss": 0.2148,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.42963196967303746,
959
+ "grad_norm": 0.1549484893694436,
960
+ "learning_rate": 8.584965255843336e-06,
961
+ "loss": 0.2162,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.43279102827357446,
966
+ "grad_norm": 0.15790917490616427,
967
+ "learning_rate": 8.6481364497789e-06,
968
+ "loss": 0.2142,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.4359500868741115,
973
+ "grad_norm": 0.17471621371832,
974
+ "learning_rate": 8.711307643714467e-06,
975
+ "loss": 0.215,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.4391091454746486,
980
+ "grad_norm": 0.1704061630987402,
981
+ "learning_rate": 8.774478837650032e-06,
982
+ "loss": 0.2145,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.4422682040751856,
987
+ "grad_norm": 0.17423241802858616,
988
+ "learning_rate": 8.837650031585598e-06,
989
+ "loss": 0.2132,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.44542726267572263,
994
+ "grad_norm": 0.16758619433784536,
995
+ "learning_rate": 8.900821225521164e-06,
996
+ "loss": 0.2138,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.4485863212762597,
1001
+ "grad_norm": 0.17999186900204928,
1002
+ "learning_rate": 8.963992419456728e-06,
1003
+ "loss": 0.2127,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.4517453798767967,
1008
+ "grad_norm": 0.17065420980005516,
1009
+ "learning_rate": 9.027163613392293e-06,
1010
+ "loss": 0.2138,
1011
+ "step": 1430
1012
+ },
1013
+ {
1014
+ "epoch": 0.45490443847733375,
1015
+ "grad_norm": 0.1974703018692422,
1016
+ "learning_rate": 9.090334807327859e-06,
1017
+ "loss": 0.2127,
1018
+ "step": 1440
1019
+ },
1020
+ {
1021
+ "epoch": 0.4580634970778708,
1022
+ "grad_norm": 0.20057492546176425,
1023
+ "learning_rate": 9.153506001263425e-06,
1024
+ "loss": 0.2141,
1025
+ "step": 1450
1026
+ },
1027
+ {
1028
+ "epoch": 0.4612225556784078,
1029
+ "grad_norm": 0.17543621309019505,
1030
+ "learning_rate": 9.21667719519899e-06,
1031
+ "loss": 0.2109,
1032
+ "step": 1460
1033
+ },
1034
+ {
1035
+ "epoch": 0.46438161427894487,
1036
+ "grad_norm": 0.1693436152342155,
1037
+ "learning_rate": 9.279848389134556e-06,
1038
+ "loss": 0.2138,
1039
+ "step": 1470
1040
+ },
1041
+ {
1042
+ "epoch": 0.4675406728794819,
1043
+ "grad_norm": 0.18371499893258605,
1044
+ "learning_rate": 9.34301958307012e-06,
1045
+ "loss": 0.2121,
1046
+ "step": 1480
1047
+ },
1048
+ {
1049
+ "epoch": 0.470699731480019,
1050
+ "grad_norm": 0.18302132069209362,
1051
+ "learning_rate": 9.406190777005687e-06,
1052
+ "loss": 0.2125,
1053
+ "step": 1490
1054
+ },
1055
+ {
1056
+ "epoch": 0.473858790080556,
1057
+ "grad_norm": 0.16878973943654269,
1058
+ "learning_rate": 9.469361970941253e-06,
1059
+ "loss": 0.2096,
1060
+ "step": 1500
1061
+ },
1062
+ {
1063
+ "epoch": 0.47701784868109304,
1064
+ "grad_norm": 0.15672142571721928,
1065
+ "learning_rate": 9.532533164876817e-06,
1066
+ "loss": 0.2157,
1067
+ "step": 1510
1068
+ },
1069
+ {
1070
+ "epoch": 0.4801769072816301,
1071
+ "grad_norm": 0.17793569853299288,
1072
+ "learning_rate": 9.595704358812382e-06,
1073
+ "loss": 0.2116,
1074
+ "step": 1520
1075
+ },
1076
+ {
1077
+ "epoch": 0.4833359658821671,
1078
+ "grad_norm": 0.1711221799529836,
1079
+ "learning_rate": 9.658875552747946e-06,
1080
+ "loss": 0.2086,
1081
+ "step": 1530
1082
+ },
1083
+ {
1084
+ "epoch": 0.48649502448270415,
1085
+ "grad_norm": 0.15221905413795137,
1086
+ "learning_rate": 9.722046746683513e-06,
1087
+ "loss": 0.2079,
1088
+ "step": 1540
1089
+ },
1090
+ {
1091
+ "epoch": 0.4896540830832412,
1092
+ "grad_norm": 0.1706269772815951,
1093
+ "learning_rate": 9.785217940619079e-06,
1094
+ "loss": 0.2124,
1095
+ "step": 1550
1096
+ },
1097
+ {
1098
+ "epoch": 0.4928131416837782,
1099
+ "grad_norm": 0.17041271545684786,
1100
+ "learning_rate": 9.848389134554643e-06,
1101
+ "loss": 0.2105,
1102
+ "step": 1560
1103
+ },
1104
+ {
1105
+ "epoch": 0.49597220028431527,
1106
+ "grad_norm": 0.1707130122866249,
1107
+ "learning_rate": 9.91156032849021e-06,
1108
+ "loss": 0.2094,
1109
+ "step": 1570
1110
+ },
1111
+ {
1112
+ "epoch": 0.4991312588848523,
1113
+ "grad_norm": 0.1533585694871482,
1114
+ "learning_rate": 9.974731522425774e-06,
1115
+ "loss": 0.2125,
1116
+ "step": 1580
1117
+ },
1118
+ {
1119
+ "epoch": 0.5022903174853893,
1120
+ "grad_norm": 0.1445021677961463,
1121
+ "learning_rate": 9.99999562381833e-06,
1122
+ "loss": 0.2104,
1123
+ "step": 1590
1124
+ },
1125
+ {
1126
+ "epoch": 0.5054493760859264,
1127
+ "grad_norm": 0.14451068806922954,
1128
+ "learning_rate": 9.999968880513634e-06,
1129
+ "loss": 0.2115,
1130
+ "step": 1600
1131
+ },
1132
+ {
1133
+ "epoch": 0.5086084346864634,
1134
+ "grad_norm": 0.1711007683006565,
1135
+ "learning_rate": 9.99991782524616e-06,
1136
+ "loss": 0.2119,
1137
+ "step": 1610
1138
+ },
1139
+ {
1140
+ "epoch": 0.5117674932870004,
1141
+ "grad_norm": 0.14862996999525604,
1142
+ "learning_rate": 9.999842458264166e-06,
1143
+ "loss": 0.2091,
1144
+ "step": 1620
1145
+ },
1146
+ {
1147
+ "epoch": 0.5149265518875376,
1148
+ "grad_norm": 0.16833296434447828,
1149
+ "learning_rate": 9.999742779934113e-06,
1150
+ "loss": 0.2089,
1151
+ "step": 1630
1152
+ },
1153
+ {
1154
+ "epoch": 0.5180856104880746,
1155
+ "grad_norm": 0.16624207360408486,
1156
+ "learning_rate": 9.999618790740677e-06,
1157
+ "loss": 0.2076,
1158
+ "step": 1640
1159
+ },
1160
+ {
1161
+ "epoch": 0.5212446690886116,
1162
+ "grad_norm": 0.18734837267014448,
1163
+ "learning_rate": 9.99947049128675e-06,
1164
+ "loss": 0.2093,
1165
+ "step": 1650
1166
+ },
1167
+ {
1168
+ "epoch": 0.5244037276891487,
1169
+ "grad_norm": 0.17593769602265188,
1170
+ "learning_rate": 9.999297882293429e-06,
1171
+ "loss": 0.2104,
1172
+ "step": 1660
1173
+ },
1174
+ {
1175
+ "epoch": 0.5275627862896857,
1176
+ "grad_norm": 0.1534253401122401,
1177
+ "learning_rate": 9.999100964600006e-06,
1178
+ "loss": 0.2094,
1179
+ "step": 1670
1180
+ },
1181
+ {
1182
+ "epoch": 0.5307218448902227,
1183
+ "grad_norm": 0.1488174937468484,
1184
+ "learning_rate": 9.998879739163982e-06,
1185
+ "loss": 0.2087,
1186
+ "step": 1680
1187
+ },
1188
+ {
1189
+ "epoch": 0.5338809034907598,
1190
+ "grad_norm": 0.13829240161020598,
1191
+ "learning_rate": 9.998634207061047e-06,
1192
+ "loss": 0.2083,
1193
+ "step": 1690
1194
+ },
1195
+ {
1196
+ "epoch": 0.5370399620912968,
1197
+ "grad_norm": 0.15837901482606578,
1198
+ "learning_rate": 9.998364369485083e-06,
1199
+ "loss": 0.2065,
1200
+ "step": 1700
1201
+ },
1202
+ {
1203
+ "epoch": 0.5401990206918338,
1204
+ "grad_norm": 0.14373686188939075,
1205
+ "learning_rate": 9.998070227748153e-06,
1206
+ "loss": 0.2077,
1207
+ "step": 1710
1208
+ },
1209
+ {
1210
+ "epoch": 0.5433580792923709,
1211
+ "grad_norm": 0.13856399191761934,
1212
+ "learning_rate": 9.9977517832805e-06,
1213
+ "loss": 0.2074,
1214
+ "step": 1720
1215
+ },
1216
+ {
1217
+ "epoch": 0.5465171378929079,
1218
+ "grad_norm": 0.16428912206666343,
1219
+ "learning_rate": 9.997409037630533e-06,
1220
+ "loss": 0.2072,
1221
+ "step": 1730
1222
+ },
1223
+ {
1224
+ "epoch": 0.5496761964934449,
1225
+ "grad_norm": 0.14598794955578479,
1226
+ "learning_rate": 9.997041992464828e-06,
1227
+ "loss": 0.207,
1228
+ "step": 1740
1229
+ },
1230
+ {
1231
+ "epoch": 0.552835255093982,
1232
+ "grad_norm": 0.1544681862373563,
1233
+ "learning_rate": 9.996650649568116e-06,
1234
+ "loss": 0.2067,
1235
+ "step": 1750
1236
+ },
1237
+ {
1238
+ "epoch": 0.555994313694519,
1239
+ "grad_norm": 0.16063147300592975,
1240
+ "learning_rate": 9.996235010843269e-06,
1241
+ "loss": 0.2091,
1242
+ "step": 1760
1243
+ },
1244
+ {
1245
+ "epoch": 0.5591533722950561,
1246
+ "grad_norm": 0.16009334025881428,
1247
+ "learning_rate": 9.9957950783113e-06,
1248
+ "loss": 0.2068,
1249
+ "step": 1770
1250
+ },
1251
+ {
1252
+ "epoch": 0.5623124308955931,
1253
+ "grad_norm": 0.1473832672273453,
1254
+ "learning_rate": 9.995330854111342e-06,
1255
+ "loss": 0.2072,
1256
+ "step": 1780
1257
+ },
1258
+ {
1259
+ "epoch": 0.5654714894961301,
1260
+ "grad_norm": 0.1758179524267874,
1261
+ "learning_rate": 9.994842340500654e-06,
1262
+ "loss": 0.2051,
1263
+ "step": 1790
1264
+ },
1265
+ {
1266
+ "epoch": 0.5686305480966672,
1267
+ "grad_norm": 0.160062543193855,
1268
+ "learning_rate": 9.994329539854597e-06,
1269
+ "loss": 0.2023,
1270
+ "step": 1800
1271
+ },
1272
+ {
1273
+ "epoch": 0.5717896066972042,
1274
+ "grad_norm": 0.14481223083195094,
1275
+ "learning_rate": 9.993792454666622e-06,
1276
+ "loss": 0.2049,
1277
+ "step": 1810
1278
+ },
1279
+ {
1280
+ "epoch": 0.5749486652977412,
1281
+ "grad_norm": 0.17516638701269122,
1282
+ "learning_rate": 9.993231087548263e-06,
1283
+ "loss": 0.2056,
1284
+ "step": 1820
1285
+ },
1286
+ {
1287
+ "epoch": 0.5781077238982784,
1288
+ "grad_norm": 0.15577929587795278,
1289
+ "learning_rate": 9.992645441229128e-06,
1290
+ "loss": 0.2053,
1291
+ "step": 1830
1292
+ },
1293
+ {
1294
+ "epoch": 0.5812667824988154,
1295
+ "grad_norm": 0.16063695498724154,
1296
+ "learning_rate": 9.992035518556873e-06,
1297
+ "loss": 0.2032,
1298
+ "step": 1840
1299
+ },
1300
+ {
1301
+ "epoch": 0.5844258410993524,
1302
+ "grad_norm": 0.13465469823943357,
1303
+ "learning_rate": 9.991401322497202e-06,
1304
+ "loss": 0.2078,
1305
+ "step": 1850
1306
+ },
1307
+ {
1308
+ "epoch": 0.5875848996998895,
1309
+ "grad_norm": 0.14458781318924407,
1310
+ "learning_rate": 9.990742856133844e-06,
1311
+ "loss": 0.2075,
1312
+ "step": 1860
1313
+ },
1314
+ {
1315
+ "epoch": 0.5907439583004265,
1316
+ "grad_norm": 0.15458252884662305,
1317
+ "learning_rate": 9.990060122668543e-06,
1318
+ "loss": 0.2058,
1319
+ "step": 1870
1320
+ },
1321
+ {
1322
+ "epoch": 0.5939030169009635,
1323
+ "grad_norm": 0.13378310324370862,
1324
+ "learning_rate": 9.989353125421034e-06,
1325
+ "loss": 0.2077,
1326
+ "step": 1880
1327
+ },
1328
+ {
1329
+ "epoch": 0.5970620755015006,
1330
+ "grad_norm": 0.14092827320728965,
1331
+ "learning_rate": 9.98862186782904e-06,
1332
+ "loss": 0.205,
1333
+ "step": 1890
1334
+ },
1335
+ {
1336
+ "epoch": 0.6002211341020376,
1337
+ "grad_norm": 0.14528693075899177,
1338
+ "learning_rate": 9.987866353448241e-06,
1339
+ "loss": 0.2056,
1340
+ "step": 1900
1341
+ },
1342
+ {
1343
+ "epoch": 0.6033801927025746,
1344
+ "grad_norm": 0.16835849378700468,
1345
+ "learning_rate": 9.987086585952271e-06,
1346
+ "loss": 0.202,
1347
+ "step": 1910
1348
+ },
1349
+ {
1350
+ "epoch": 0.6065392513031117,
1351
+ "grad_norm": 0.14702108474072842,
1352
+ "learning_rate": 9.986282569132688e-06,
1353
+ "loss": 0.2046,
1354
+ "step": 1920
1355
+ },
1356
+ {
1357
+ "epoch": 0.6096983099036487,
1358
+ "grad_norm": 0.15598954366138304,
1359
+ "learning_rate": 9.98545430689896e-06,
1360
+ "loss": 0.2037,
1361
+ "step": 1930
1362
+ },
1363
+ {
1364
+ "epoch": 0.6128573685041857,
1365
+ "grad_norm": 0.15363339652231678,
1366
+ "learning_rate": 9.984601803278451e-06,
1367
+ "loss": 0.2065,
1368
+ "step": 1940
1369
+ },
1370
+ {
1371
+ "epoch": 0.6160164271047228,
1372
+ "grad_norm": 0.15079303342445485,
1373
+ "learning_rate": 9.983725062416392e-06,
1374
+ "loss": 0.2046,
1375
+ "step": 1950
1376
+ },
1377
+ {
1378
+ "epoch": 0.6191754857052598,
1379
+ "grad_norm": 0.13780206734265157,
1380
+ "learning_rate": 9.98282408857587e-06,
1381
+ "loss": 0.2054,
1382
+ "step": 1960
1383
+ },
1384
+ {
1385
+ "epoch": 0.6223345443057968,
1386
+ "grad_norm": 0.1355920930705493,
1387
+ "learning_rate": 9.981898886137795e-06,
1388
+ "loss": 0.2039,
1389
+ "step": 1970
1390
+ },
1391
+ {
1392
+ "epoch": 0.6254936029063339,
1393
+ "grad_norm": 0.14054778625440462,
1394
+ "learning_rate": 9.980949459600899e-06,
1395
+ "loss": 0.2045,
1396
+ "step": 1980
1397
+ },
1398
+ {
1399
+ "epoch": 0.6286526615068709,
1400
+ "grad_norm": 0.16259862482128506,
1401
+ "learning_rate": 9.979975813581694e-06,
1402
+ "loss": 0.2033,
1403
+ "step": 1990
1404
+ },
1405
+ {
1406
+ "epoch": 0.6318117201074079,
1407
+ "grad_norm": 0.16320641790440754,
1408
+ "learning_rate": 9.978977952814456e-06,
1409
+ "loss": 0.2053,
1410
+ "step": 2000
1411
+ },
1412
+ {
1413
+ "epoch": 0.634970778707945,
1414
+ "grad_norm": 0.14504338218528204,
1415
+ "learning_rate": 9.97795588215121e-06,
1416
+ "loss": 0.2041,
1417
+ "step": 2010
1418
+ },
1419
+ {
1420
+ "epoch": 0.638129837308482,
1421
+ "grad_norm": 0.14088062087038478,
1422
+ "learning_rate": 9.97690960656169e-06,
1423
+ "loss": 0.2034,
1424
+ "step": 2020
1425
+ },
1426
+ {
1427
+ "epoch": 0.6412888959090192,
1428
+ "grad_norm": 0.16368648316215648,
1429
+ "learning_rate": 9.975839131133335e-06,
1430
+ "loss": 0.2004,
1431
+ "step": 2030
1432
+ },
1433
+ {
1434
+ "epoch": 0.6444479545095562,
1435
+ "grad_norm": 0.14137074446596326,
1436
+ "learning_rate": 9.974744461071246e-06,
1437
+ "loss": 0.2039,
1438
+ "step": 2040
1439
+ },
1440
+ {
1441
+ "epoch": 0.6476070131100932,
1442
+ "grad_norm": 0.16743385196077595,
1443
+ "learning_rate": 9.973625601698176e-06,
1444
+ "loss": 0.2024,
1445
+ "step": 2050
1446
+ },
1447
+ {
1448
+ "epoch": 0.6507660717106303,
1449
+ "grad_norm": 0.1420212347341941,
1450
+ "learning_rate": 9.972482558454488e-06,
1451
+ "loss": 0.2006,
1452
+ "step": 2060
1453
+ },
1454
+ {
1455
+ "epoch": 0.6539251303111673,
1456
+ "grad_norm": 0.1541147190019739,
1457
+ "learning_rate": 9.971315336898144e-06,
1458
+ "loss": 0.2031,
1459
+ "step": 2070
1460
+ },
1461
+ {
1462
+ "epoch": 0.6570841889117043,
1463
+ "grad_norm": 0.14954970584415023,
1464
+ "learning_rate": 9.970123942704667e-06,
1465
+ "loss": 0.2022,
1466
+ "step": 2080
1467
+ },
1468
+ {
1469
+ "epoch": 0.6602432475122414,
1470
+ "grad_norm": 0.13835942774621643,
1471
+ "learning_rate": 9.968908381667122e-06,
1472
+ "loss": 0.205,
1473
+ "step": 2090
1474
+ },
1475
+ {
1476
+ "epoch": 0.6634023061127784,
1477
+ "grad_norm": 0.1444063211993615,
1478
+ "learning_rate": 9.967668659696077e-06,
1479
+ "loss": 0.2003,
1480
+ "step": 2100
1481
+ },
1482
+ {
1483
+ "epoch": 0.6665613647133154,
1484
+ "grad_norm": 0.15181588903023388,
1485
+ "learning_rate": 9.966404782819587e-06,
1486
+ "loss": 0.2041,
1487
+ "step": 2110
1488
+ },
1489
+ {
1490
+ "epoch": 0.6697204233138525,
1491
+ "grad_norm": 0.14453828048169265,
1492
+ "learning_rate": 9.965116757183156e-06,
1493
+ "loss": 0.2008,
1494
+ "step": 2120
1495
+ },
1496
+ {
1497
+ "epoch": 0.6728794819143895,
1498
+ "grad_norm": 0.16433983585515474,
1499
+ "learning_rate": 9.963804589049709e-06,
1500
+ "loss": 0.2045,
1501
+ "step": 2130
1502
+ },
1503
+ {
1504
+ "epoch": 0.6760385405149265,
1505
+ "grad_norm": 0.14280102041208004,
1506
+ "learning_rate": 9.962468284799559e-06,
1507
+ "loss": 0.2021,
1508
+ "step": 2140
1509
+ },
1510
+ {
1511
+ "epoch": 0.6791975991154636,
1512
+ "grad_norm": 0.14204139100462726,
1513
+ "learning_rate": 9.961107850930386e-06,
1514
+ "loss": 0.201,
1515
+ "step": 2150
1516
+ },
1517
+ {
1518
+ "epoch": 0.6823566577160006,
1519
+ "grad_norm": 0.1324076473779632,
1520
+ "learning_rate": 9.959723294057195e-06,
1521
+ "loss": 0.1991,
1522
+ "step": 2160
1523
+ },
1524
+ {
1525
+ "epoch": 0.6855157163165376,
1526
+ "grad_norm": 0.14295162932415698,
1527
+ "learning_rate": 9.958314620912283e-06,
1528
+ "loss": 0.2025,
1529
+ "step": 2170
1530
+ },
1531
+ {
1532
+ "epoch": 0.6886747749170747,
1533
+ "grad_norm": 0.15726554545849142,
1534
+ "learning_rate": 9.956881838345221e-06,
1535
+ "loss": 0.2033,
1536
+ "step": 2180
1537
+ },
1538
+ {
1539
+ "epoch": 0.6918338335176117,
1540
+ "grad_norm": 0.16250545787100992,
1541
+ "learning_rate": 9.955424953322797e-06,
1542
+ "loss": 0.2015,
1543
+ "step": 2190
1544
+ },
1545
+ {
1546
+ "epoch": 0.6949928921181487,
1547
+ "grad_norm": 0.1614266022365173,
1548
+ "learning_rate": 9.953943972929003e-06,
1549
+ "loss": 0.1996,
1550
+ "step": 2200
1551
+ },
1552
+ {
1553
+ "epoch": 0.6981519507186859,
1554
+ "grad_norm": 0.15900062576977386,
1555
+ "learning_rate": 9.952438904364996e-06,
1556
+ "loss": 0.2026,
1557
+ "step": 2210
1558
+ },
1559
+ {
1560
+ "epoch": 0.7013110093192229,
1561
+ "grad_norm": 0.13873539419869083,
1562
+ "learning_rate": 9.950909754949052e-06,
1563
+ "loss": 0.2035,
1564
+ "step": 2220
1565
+ },
1566
+ {
1567
+ "epoch": 0.7044700679197599,
1568
+ "grad_norm": 0.16329226810226954,
1569
+ "learning_rate": 9.949356532116546e-06,
1570
+ "loss": 0.2017,
1571
+ "step": 2230
1572
+ },
1573
+ {
1574
+ "epoch": 0.707629126520297,
1575
+ "grad_norm": 0.130433030569765,
1576
+ "learning_rate": 9.947779243419899e-06,
1577
+ "loss": 0.2017,
1578
+ "step": 2240
1579
+ },
1580
+ {
1581
+ "epoch": 0.710788185120834,
1582
+ "grad_norm": 0.13992897215148528,
1583
+ "learning_rate": 9.946177896528557e-06,
1584
+ "loss": 0.2001,
1585
+ "step": 2250
1586
+ },
1587
+ {
1588
+ "epoch": 0.7139472437213711,
1589
+ "grad_norm": 0.14070910045992718,
1590
+ "learning_rate": 9.944552499228947e-06,
1591
+ "loss": 0.2007,
1592
+ "step": 2260
1593
+ },
1594
+ {
1595
+ "epoch": 0.7171063023219081,
1596
+ "grad_norm": 0.1469150772379642,
1597
+ "learning_rate": 9.942903059424441e-06,
1598
+ "loss": 0.2006,
1599
+ "step": 2270
1600
+ },
1601
+ {
1602
+ "epoch": 0.7202653609224451,
1603
+ "grad_norm": 0.18224682305919618,
1604
+ "learning_rate": 9.941229585135307e-06,
1605
+ "loss": 0.2028,
1606
+ "step": 2280
1607
+ },
1608
+ {
1609
+ "epoch": 0.7234244195229822,
1610
+ "grad_norm": 0.14288906489417755,
1611
+ "learning_rate": 9.939532084498685e-06,
1612
+ "loss": 0.1987,
1613
+ "step": 2290
1614
+ },
1615
+ {
1616
+ "epoch": 0.7265834781235192,
1617
+ "grad_norm": 0.13437271496120856,
1618
+ "learning_rate": 9.937810565768544e-06,
1619
+ "loss": 0.1993,
1620
+ "step": 2300
1621
+ },
1622
+ {
1623
+ "epoch": 0.7297425367240562,
1624
+ "grad_norm": 0.1446887097806904,
1625
+ "learning_rate": 9.936065037315636e-06,
1626
+ "loss": 0.2011,
1627
+ "step": 2310
1628
+ },
1629
+ {
1630
+ "epoch": 0.7329015953245933,
1631
+ "grad_norm": 0.13251427513003186,
1632
+ "learning_rate": 9.934295507627456e-06,
1633
+ "loss": 0.2022,
1634
+ "step": 2320
1635
+ },
1636
+ {
1637
+ "epoch": 0.7360606539251303,
1638
+ "grad_norm": 0.1452288699010562,
1639
+ "learning_rate": 9.932501985308206e-06,
1640
+ "loss": 0.2009,
1641
+ "step": 2330
1642
+ },
1643
+ {
1644
+ "epoch": 0.7392197125256673,
1645
+ "grad_norm": 0.14728963865115374,
1646
+ "learning_rate": 9.93068447907875e-06,
1647
+ "loss": 0.1987,
1648
+ "step": 2340
1649
+ },
1650
+ {
1651
+ "epoch": 0.7423787711262044,
1652
+ "grad_norm": 0.1517245487863863,
1653
+ "learning_rate": 9.928842997776574e-06,
1654
+ "loss": 0.2013,
1655
+ "step": 2350
1656
+ },
1657
+ {
1658
+ "epoch": 0.7455378297267414,
1659
+ "grad_norm": 0.1614866572575232,
1660
+ "learning_rate": 9.926977550355734e-06,
1661
+ "loss": 0.1997,
1662
+ "step": 2360
1663
+ },
1664
+ {
1665
+ "epoch": 0.7486968883272784,
1666
+ "grad_norm": 0.14031629423175507,
1667
+ "learning_rate": 9.92508814588683e-06,
1668
+ "loss": 0.199,
1669
+ "step": 2370
1670
+ },
1671
+ {
1672
+ "epoch": 0.7518559469278155,
1673
+ "grad_norm": 0.14118485538481557,
1674
+ "learning_rate": 9.92317479355694e-06,
1675
+ "loss": 0.1976,
1676
+ "step": 2380
1677
+ },
1678
+ {
1679
+ "epoch": 0.7550150055283525,
1680
+ "grad_norm": 0.13181048592845238,
1681
+ "learning_rate": 9.921237502669595e-06,
1682
+ "loss": 0.198,
1683
+ "step": 2390
1684
+ },
1685
+ {
1686
+ "epoch": 0.7581740641288895,
1687
+ "grad_norm": 0.1345265368505879,
1688
+ "learning_rate": 9.919276282644723e-06,
1689
+ "loss": 0.201,
1690
+ "step": 2400
1691
+ },
1692
+ {
1693
+ "epoch": 0.7613331227294267,
1694
+ "grad_norm": 0.13720890023292417,
1695
+ "learning_rate": 9.917291143018604e-06,
1696
+ "loss": 0.2009,
1697
+ "step": 2410
1698
+ },
1699
+ {
1700
+ "epoch": 0.7644921813299637,
1701
+ "grad_norm": 0.13845111986552353,
1702
+ "learning_rate": 9.915282093443825e-06,
1703
+ "loss": 0.2008,
1704
+ "step": 2420
1705
+ },
1706
+ {
1707
+ "epoch": 0.7676512399305007,
1708
+ "grad_norm": 0.14617335075904797,
1709
+ "learning_rate": 9.913249143689234e-06,
1710
+ "loss": 0.1991,
1711
+ "step": 2430
1712
+ },
1713
+ {
1714
+ "epoch": 0.7708102985310378,
1715
+ "grad_norm": 0.1320877727948845,
1716
+ "learning_rate": 9.911192303639896e-06,
1717
+ "loss": 0.1999,
1718
+ "step": 2440
1719
+ },
1720
+ {
1721
+ "epoch": 0.7739693571315748,
1722
+ "grad_norm": 0.13482883240500468,
1723
+ "learning_rate": 9.909111583297035e-06,
1724
+ "loss": 0.1997,
1725
+ "step": 2450
1726
+ },
1727
+ {
1728
+ "epoch": 0.7771284157321118,
1729
+ "grad_norm": 0.14200097798675781,
1730
+ "learning_rate": 9.907006992777991e-06,
1731
+ "loss": 0.2008,
1732
+ "step": 2460
1733
+ },
1734
+ {
1735
+ "epoch": 0.7802874743326489,
1736
+ "grad_norm": 0.13733098129824253,
1737
+ "learning_rate": 9.904878542316177e-06,
1738
+ "loss": 0.1988,
1739
+ "step": 2470
1740
+ },
1741
+ {
1742
+ "epoch": 0.7834465329331859,
1743
+ "grad_norm": 0.14967333958732693,
1744
+ "learning_rate": 9.902726242261015e-06,
1745
+ "loss": 0.2,
1746
+ "step": 2480
1747
+ },
1748
+ {
1749
+ "epoch": 0.786605591533723,
1750
+ "grad_norm": 0.18469961665919096,
1751
+ "learning_rate": 9.9005501030779e-06,
1752
+ "loss": 0.1998,
1753
+ "step": 2490
1754
+ },
1755
+ {
1756
+ "epoch": 0.78976465013426,
1757
+ "grad_norm": 0.1291065536177641,
1758
+ "learning_rate": 9.898350135348143e-06,
1759
+ "loss": 0.1994,
1760
+ "step": 2500
1761
+ },
1762
+ {
1763
+ "epoch": 0.792923708734797,
1764
+ "grad_norm": 0.1455654343221393,
1765
+ "learning_rate": 9.896126349768913e-06,
1766
+ "loss": 0.1961,
1767
+ "step": 2510
1768
+ },
1769
+ {
1770
+ "epoch": 0.7960827673353341,
1771
+ "grad_norm": 0.1305825664747534,
1772
+ "learning_rate": 9.893878757153197e-06,
1773
+ "loss": 0.1997,
1774
+ "step": 2520
1775
+ },
1776
+ {
1777
+ "epoch": 0.7992418259358711,
1778
+ "grad_norm": 0.1355469021976556,
1779
+ "learning_rate": 9.891607368429741e-06,
1780
+ "loss": 0.2009,
1781
+ "step": 2530
1782
+ },
1783
+ {
1784
+ "epoch": 0.8024008845364081,
1785
+ "grad_norm": 0.1393422807545934,
1786
+ "learning_rate": 9.889312194642999e-06,
1787
+ "loss": 0.1996,
1788
+ "step": 2540
1789
+ },
1790
+ {
1791
+ "epoch": 0.8055599431369452,
1792
+ "grad_norm": 0.14137833832556562,
1793
+ "learning_rate": 9.886993246953075e-06,
1794
+ "loss": 0.1984,
1795
+ "step": 2550
1796
+ },
1797
+ {
1798
+ "epoch": 0.8087190017374822,
1799
+ "grad_norm": 0.1288518244334966,
1800
+ "learning_rate": 9.884650536635674e-06,
1801
+ "loss": 0.1998,
1802
+ "step": 2560
1803
+ },
1804
+ {
1805
+ "epoch": 0.8118780603380192,
1806
+ "grad_norm": 0.12540981604153706,
1807
+ "learning_rate": 9.882284075082042e-06,
1808
+ "loss": 0.1953,
1809
+ "step": 2570
1810
+ },
1811
+ {
1812
+ "epoch": 0.8150371189385563,
1813
+ "grad_norm": 0.1298044947287737,
1814
+ "learning_rate": 9.879893873798918e-06,
1815
+ "loss": 0.1998,
1816
+ "step": 2580
1817
+ },
1818
+ {
1819
+ "epoch": 0.8181961775390934,
1820
+ "grad_norm": 0.13593942142698026,
1821
+ "learning_rate": 9.877479944408469e-06,
1822
+ "loss": 0.1994,
1823
+ "step": 2590
1824
+ },
1825
+ {
1826
+ "epoch": 0.8213552361396304,
1827
+ "grad_norm": 0.14577596422732375,
1828
+ "learning_rate": 9.875042298648241e-06,
1829
+ "loss": 0.1968,
1830
+ "step": 2600
1831
+ },
1832
+ {
1833
+ "epoch": 0.8245142947401675,
1834
+ "grad_norm": 0.13651649878117303,
1835
+ "learning_rate": 9.872580948371101e-06,
1836
+ "loss": 0.1959,
1837
+ "step": 2610
1838
+ },
1839
+ {
1840
+ "epoch": 0.8276733533407045,
1841
+ "grad_norm": 0.13458568002303536,
1842
+ "learning_rate": 9.870095905545172e-06,
1843
+ "loss": 0.1975,
1844
+ "step": 2620
1845
+ },
1846
+ {
1847
+ "epoch": 0.8308324119412415,
1848
+ "grad_norm": 0.13467497395688513,
1849
+ "learning_rate": 9.867587182253783e-06,
1850
+ "loss": 0.198,
1851
+ "step": 2630
1852
+ },
1853
+ {
1854
+ "epoch": 0.8339914705417786,
1855
+ "grad_norm": 0.1334106181890542,
1856
+ "learning_rate": 9.86505479069541e-06,
1857
+ "loss": 0.1975,
1858
+ "step": 2640
1859
+ },
1860
+ {
1861
+ "epoch": 0.8371505291423156,
1862
+ "grad_norm": 0.14726119516550862,
1863
+ "learning_rate": 9.862498743183606e-06,
1864
+ "loss": 0.1962,
1865
+ "step": 2650
1866
+ },
1867
+ {
1868
+ "epoch": 0.8403095877428526,
1869
+ "grad_norm": 0.13818332053028007,
1870
+ "learning_rate": 9.85991905214696e-06,
1871
+ "loss": 0.1998,
1872
+ "step": 2660
1873
+ },
1874
+ {
1875
+ "epoch": 0.8434686463433897,
1876
+ "grad_norm": 0.14412901343016873,
1877
+ "learning_rate": 9.85731573012902e-06,
1878
+ "loss": 0.2,
1879
+ "step": 2670
1880
+ },
1881
+ {
1882
+ "epoch": 0.8466277049439267,
1883
+ "grad_norm": 0.1255413662933083,
1884
+ "learning_rate": 9.854688789788236e-06,
1885
+ "loss": 0.198,
1886
+ "step": 2680
1887
+ },
1888
+ {
1889
+ "epoch": 0.8497867635444637,
1890
+ "grad_norm": 0.13187344960522424,
1891
+ "learning_rate": 9.852038243897903e-06,
1892
+ "loss": 0.1972,
1893
+ "step": 2690
1894
+ },
1895
+ {
1896
+ "epoch": 0.8529458221450008,
1897
+ "grad_norm": 0.13247624619465903,
1898
+ "learning_rate": 9.849364105346098e-06,
1899
+ "loss": 0.1982,
1900
+ "step": 2700
1901
+ },
1902
+ {
1903
+ "epoch": 0.8561048807455378,
1904
+ "grad_norm": 0.12352050553226067,
1905
+ "learning_rate": 9.846666387135613e-06,
1906
+ "loss": 0.1954,
1907
+ "step": 2710
1908
+ },
1909
+ {
1910
+ "epoch": 0.8592639393460749,
1911
+ "grad_norm": 0.13384183862518867,
1912
+ "learning_rate": 9.843945102383892e-06,
1913
+ "loss": 0.197,
1914
+ "step": 2720
1915
+ },
1916
+ {
1917
+ "epoch": 0.8624229979466119,
1918
+ "grad_norm": 0.12329786824129346,
1919
+ "learning_rate": 9.841200264322974e-06,
1920
+ "loss": 0.1977,
1921
+ "step": 2730
1922
+ },
1923
+ {
1924
+ "epoch": 0.8655820565471489,
1925
+ "grad_norm": 0.13477039032719176,
1926
+ "learning_rate": 9.838431886299421e-06,
1927
+ "loss": 0.1961,
1928
+ "step": 2740
1929
+ },
1930
+ {
1931
+ "epoch": 0.868741115147686,
1932
+ "grad_norm": 0.13828179616792935,
1933
+ "learning_rate": 9.83563998177426e-06,
1934
+ "loss": 0.1967,
1935
+ "step": 2750
1936
+ },
1937
+ {
1938
+ "epoch": 0.871900173748223,
1939
+ "grad_norm": 0.14564045162827766,
1940
+ "learning_rate": 9.83282456432291e-06,
1941
+ "loss": 0.1965,
1942
+ "step": 2760
1943
+ },
1944
+ {
1945
+ "epoch": 0.87505923234876,
1946
+ "grad_norm": 0.13685777011127429,
1947
+ "learning_rate": 9.829985647635118e-06,
1948
+ "loss": 0.1981,
1949
+ "step": 2770
1950
+ },
1951
+ {
1952
+ "epoch": 0.8782182909492972,
1953
+ "grad_norm": 0.13617569439938054,
1954
+ "learning_rate": 9.827123245514901e-06,
1955
+ "loss": 0.1951,
1956
+ "step": 2780
1957
+ },
1958
+ {
1959
+ "epoch": 0.8813773495498342,
1960
+ "grad_norm": 0.12868163128280088,
1961
+ "learning_rate": 9.824237371880469e-06,
1962
+ "loss": 0.195,
1963
+ "step": 2790
1964
+ },
1965
+ {
1966
+ "epoch": 0.8845364081503712,
1967
+ "grad_norm": 0.13965590647075304,
1968
+ "learning_rate": 9.821328040764157e-06,
1969
+ "loss": 0.1984,
1970
+ "step": 2800
1971
+ },
1972
+ {
1973
+ "epoch": 0.8876954667509083,
1974
+ "grad_norm": 0.1191526599411457,
1975
+ "learning_rate": 9.818395266312363e-06,
1976
+ "loss": 0.1925,
1977
+ "step": 2810
1978
+ },
1979
+ {
1980
+ "epoch": 0.8908545253514453,
1981
+ "grad_norm": 0.1209195098683138,
1982
+ "learning_rate": 9.81543906278548e-06,
1983
+ "loss": 0.1995,
1984
+ "step": 2820
1985
+ },
1986
+ {
1987
+ "epoch": 0.8940135839519823,
1988
+ "grad_norm": 0.12466523038010362,
1989
+ "learning_rate": 9.812459444557815e-06,
1990
+ "loss": 0.1937,
1991
+ "step": 2830
1992
+ },
1993
+ {
1994
+ "epoch": 0.8971726425525194,
1995
+ "grad_norm": 0.12952393163312614,
1996
+ "learning_rate": 9.809456426117533e-06,
1997
+ "loss": 0.1932,
1998
+ "step": 2840
1999
+ },
2000
+ {
2001
+ "epoch": 0.9003317011530564,
2002
+ "grad_norm": 0.13392155972728179,
2003
+ "learning_rate": 9.806430022066582e-06,
2004
+ "loss": 0.1978,
2005
+ "step": 2850
2006
+ },
2007
+ {
2008
+ "epoch": 0.9034907597535934,
2009
+ "grad_norm": 0.13186718584966667,
2010
+ "learning_rate": 9.803380247120616e-06,
2011
+ "loss": 0.1953,
2012
+ "step": 2860
2013
+ },
2014
+ {
2015
+ "epoch": 0.9066498183541305,
2016
+ "grad_norm": 0.1283857878998356,
2017
+ "learning_rate": 9.800307116108931e-06,
2018
+ "loss": 0.1962,
2019
+ "step": 2870
2020
+ },
2021
+ {
2022
+ "epoch": 0.9098088769546675,
2023
+ "grad_norm": 0.12302487764368193,
2024
+ "learning_rate": 9.797210643974388e-06,
2025
+ "loss": 0.1954,
2026
+ "step": 2880
2027
+ },
2028
+ {
2029
+ "epoch": 0.9129679355552045,
2030
+ "grad_norm": 0.1279449953769118,
2031
+ "learning_rate": 9.794090845773346e-06,
2032
+ "loss": 0.1936,
2033
+ "step": 2890
2034
+ },
2035
+ {
2036
+ "epoch": 0.9161269941557416,
2037
+ "grad_norm": 0.13723185562370793,
2038
+ "learning_rate": 9.79094773667558e-06,
2039
+ "loss": 0.1948,
2040
+ "step": 2900
2041
+ },
2042
+ {
2043
+ "epoch": 0.9192860527562786,
2044
+ "grad_norm": 0.1382694059956154,
2045
+ "learning_rate": 9.787781331964217e-06,
2046
+ "loss": 0.1961,
2047
+ "step": 2910
2048
+ },
2049
+ {
2050
+ "epoch": 0.9224451113568156,
2051
+ "grad_norm": 0.13522487929855218,
2052
+ "learning_rate": 9.784591647035654e-06,
2053
+ "loss": 0.1944,
2054
+ "step": 2920
2055
+ },
2056
+ {
2057
+ "epoch": 0.9256041699573527,
2058
+ "grad_norm": 0.13619232160862846,
2059
+ "learning_rate": 9.781378697399492e-06,
2060
+ "loss": 0.1939,
2061
+ "step": 2930
2062
+ },
2063
+ {
2064
+ "epoch": 0.9287632285578897,
2065
+ "grad_norm": 0.13966605421891545,
2066
+ "learning_rate": 9.778142498678447e-06,
2067
+ "loss": 0.1936,
2068
+ "step": 2940
2069
+ },
2070
+ {
2071
+ "epoch": 0.9319222871584268,
2072
+ "grad_norm": 0.13610146242659704,
2073
+ "learning_rate": 9.774883066608288e-06,
2074
+ "loss": 0.1955,
2075
+ "step": 2950
2076
+ },
2077
+ {
2078
+ "epoch": 0.9350813457589638,
2079
+ "grad_norm": 0.1272898066516385,
2080
+ "learning_rate": 9.771600417037747e-06,
2081
+ "loss": 0.1951,
2082
+ "step": 2960
2083
+ },
2084
+ {
2085
+ "epoch": 0.9382404043595008,
2086
+ "grad_norm": 0.12577223515891656,
2087
+ "learning_rate": 9.76829456592846e-06,
2088
+ "loss": 0.1941,
2089
+ "step": 2970
2090
+ },
2091
+ {
2092
+ "epoch": 0.941399462960038,
2093
+ "grad_norm": 0.13229230817335338,
2094
+ "learning_rate": 9.76496552935487e-06,
2095
+ "loss": 0.1948,
2096
+ "step": 2980
2097
+ },
2098
+ {
2099
+ "epoch": 0.944558521560575,
2100
+ "grad_norm": 0.13057771902599097,
2101
+ "learning_rate": 9.76161332350416e-06,
2102
+ "loss": 0.1945,
2103
+ "step": 2990
2104
+ },
2105
+ {
2106
+ "epoch": 0.947717580161112,
2107
+ "grad_norm": 0.1375087406497119,
2108
+ "learning_rate": 9.758237964676175e-06,
2109
+ "loss": 0.1946,
2110
+ "step": 3000
2111
+ },
2112
+ {
2113
+ "epoch": 0.9508766387616491,
2114
+ "grad_norm": 0.13144411031384784,
2115
+ "learning_rate": 9.754839469283333e-06,
2116
+ "loss": 0.1916,
2117
+ "step": 3010
2118
+ },
2119
+ {
2120
+ "epoch": 0.9540356973621861,
2121
+ "grad_norm": 0.1307858984474674,
2122
+ "learning_rate": 9.751417853850557e-06,
2123
+ "loss": 0.1961,
2124
+ "step": 3020
2125
+ },
2126
+ {
2127
+ "epoch": 0.9571947559627231,
2128
+ "grad_norm": 0.12400375737914372,
2129
+ "learning_rate": 9.747973135015187e-06,
2130
+ "loss": 0.1948,
2131
+ "step": 3030
2132
+ },
2133
+ {
2134
+ "epoch": 0.9603538145632602,
2135
+ "grad_norm": 0.14800804219149807,
2136
+ "learning_rate": 9.744505329526906e-06,
2137
+ "loss": 0.1951,
2138
+ "step": 3040
2139
+ },
2140
+ {
2141
+ "epoch": 0.9635128731637972,
2142
+ "grad_norm": 0.13528680108076863,
2143
+ "learning_rate": 9.741014454247648e-06,
2144
+ "loss": 0.1946,
2145
+ "step": 3050
2146
+ },
2147
+ {
2148
+ "epoch": 0.9666719317643342,
2149
+ "grad_norm": 0.15276170947974638,
2150
+ "learning_rate": 9.737500526151525e-06,
2151
+ "loss": 0.1935,
2152
+ "step": 3060
2153
+ },
2154
+ {
2155
+ "epoch": 0.9698309903648713,
2156
+ "grad_norm": 0.1280505905081845,
2157
+ "learning_rate": 9.733963562324739e-06,
2158
+ "loss": 0.193,
2159
+ "step": 3070
2160
+ },
2161
+ {
2162
+ "epoch": 0.9729900489654083,
2163
+ "grad_norm": 0.13045657850862527,
2164
+ "learning_rate": 9.730403579965508e-06,
2165
+ "loss": 0.1953,
2166
+ "step": 3080
2167
+ },
2168
+ {
2169
+ "epoch": 0.9761491075659453,
2170
+ "grad_norm": 0.12920108483379814,
2171
+ "learning_rate": 9.726820596383968e-06,
2172
+ "loss": 0.194,
2173
+ "step": 3090
2174
+ },
2175
+ {
2176
+ "epoch": 0.9793081661664824,
2177
+ "grad_norm": 0.1275121663048079,
2178
+ "learning_rate": 9.723214629002103e-06,
2179
+ "loss": 0.1937,
2180
+ "step": 3100
2181
+ },
2182
+ {
2183
+ "epoch": 0.9824672247670194,
2184
+ "grad_norm": 0.13153937715884076,
2185
+ "learning_rate": 9.719585695353648e-06,
2186
+ "loss": 0.1927,
2187
+ "step": 3110
2188
+ },
2189
+ {
2190
+ "epoch": 0.9856262833675564,
2191
+ "grad_norm": 0.13540077808963083,
2192
+ "learning_rate": 9.715933813084012e-06,
2193
+ "loss": 0.1948,
2194
+ "step": 3120
2195
+ },
2196
+ {
2197
+ "epoch": 0.9887853419680935,
2198
+ "grad_norm": 0.13634113133020404,
2199
+ "learning_rate": 9.712258999950196e-06,
2200
+ "loss": 0.192,
2201
+ "step": 3130
2202
+ },
2203
+ {
2204
+ "epoch": 0.9919444005686305,
2205
+ "grad_norm": 0.11908229587933031,
2206
+ "learning_rate": 9.70856127382069e-06,
2207
+ "loss": 0.1937,
2208
+ "step": 3140
2209
+ },
2210
+ {
2211
+ "epoch": 0.9951034591691675,
2212
+ "grad_norm": 0.11928897371871516,
2213
+ "learning_rate": 9.704840652675405e-06,
2214
+ "loss": 0.1939,
2215
+ "step": 3150
2216
+ },
2217
+ {
2218
+ "epoch": 0.9982625177697046,
2219
+ "grad_norm": 0.12957379845580613,
2220
+ "learning_rate": 9.701097154605572e-06,
2221
+ "loss": 0.1934,
2222
+ "step": 3160
2223
+ },
2224
+ {
2225
+ "epoch": 1.0015795293002685,
2226
+ "grad_norm": 0.12555682464472692,
2227
+ "learning_rate": 9.697330797813665e-06,
2228
+ "loss": 0.2078,
2229
+ "step": 3170
2230
+ },
2231
+ {
2232
+ "epoch": 1.0047385879008055,
2233
+ "grad_norm": 0.12366696963941619,
2234
+ "learning_rate": 9.693541600613297e-06,
2235
+ "loss": 0.1833,
2236
+ "step": 3180
2237
+ },
2238
+ {
2239
+ "epoch": 1.0078976465013425,
2240
+ "grad_norm": 0.12387115361971955,
2241
+ "learning_rate": 9.689729581429154e-06,
2242
+ "loss": 0.184,
2243
+ "step": 3190
2244
+ },
2245
+ {
2246
+ "epoch": 1.0110567051018797,
2247
+ "grad_norm": 0.12394939667356165,
2248
+ "learning_rate": 9.68589475879688e-06,
2249
+ "loss": 0.182,
2250
+ "step": 3200
2251
+ }
2252
+ ],
2253
+ "logging_steps": 10,
2254
+ "max_steps": 15830,
2255
+ "num_input_tokens_seen": 0,
2256
+ "num_train_epochs": 5,
2257
+ "save_steps": 200,
2258
+ "stateful_callbacks": {
2259
+ "TrainerControl": {
2260
+ "args": {
2261
+ "should_epoch_stop": false,
2262
+ "should_evaluate": false,
2263
+ "should_log": false,
2264
+ "should_save": true,
2265
+ "should_training_stop": false
2266
+ },
2267
+ "attributes": {}
2268
+ }
2269
+ },
2270
+ "total_flos": 4.177750985487155e+16,
2271
+ "train_batch_size": 2,
2272
+ "trial_name": null,
2273
+ "trial_params": null
2274
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:49ec8656b2f5b25fb0f58d99a8d027f2fd4ab4b6554e85d3b125d1665b4a4843
3
+ size 8401
video_preprocessor_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": null,
3
+ "data_format": "channels_first",
4
+ "default_to_square": true,
5
+ "device": null,
6
+ "do_center_crop": null,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_pad": null,
10
+ "do_rescale": true,
11
+ "do_resize": true,
12
+ "do_sample_frames": false,
13
+ "fps": null,
14
+ "image_mean": [
15
+ 0.48145466,
16
+ 0.4578275,
17
+ 0.40821073
18
+ ],
19
+ "image_std": [
20
+ 0.26862954,
21
+ 0.26130258,
22
+ 0.27577711
23
+ ],
24
+ "input_data_format": null,
25
+ "max_frames": 768,
26
+ "max_pixels": 12845056,
27
+ "merge_size": 2,
28
+ "min_frames": 4,
29
+ "min_pixels": 3136,
30
+ "num_frames": null,
31
+ "patch_size": 14,
32
+ "processor_class": "Qwen2_5_VLProcessor",
33
+ "resample": 3,
34
+ "rescale_factor": 0.00392156862745098,
35
+ "size": {
36
+ "longest_edge": 12845056,
37
+ "shortest_edge": 3136
38
+ },
39
+ "size_divisor": null,
40
+ "temporal_patch_size": 2,
41
+ "video_metadata": null,
42
+ "video_processor_type": "Qwen2VLVideoProcessor"
43
+ }
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)