Hugging Face's logo

chenyili818
/
clone

Safetensors
Model card Files
xet
Community
chenyili818 commited on
Commit
1d8cff7
·
verified ·
1 Parent(s): 661c54a

Add files using upload-large-folder tool

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +2 -0
  2. checkpoint-100/rng_state_0.pth +3 -0
  3. checkpoint-100/rng_state_1.pth +3 -0
  4. checkpoint-100/rng_state_2.pth +3 -0
  5. checkpoint-100/rng_state_3.pth +3 -0
  6. checkpoint-100/scheduler.pt +3 -0
  7. checkpoint-100/training_args.bin +3 -0
  8. checkpoint-200/rng_state_0.pth +3 -0
  9. checkpoint-200/rng_state_1.pth +3 -0
  10. checkpoint-200/rng_state_2.pth +3 -0
  11. checkpoint-200/rng_state_3.pth +3 -0
  12. checkpoint-200/scheduler.pt +3 -0
  13. checkpoint-200/tokenizer_config.json +209 -0
  14. checkpoint-200/trainer_state.json +1434 -0
  15. checkpoint-200/vocab.json +0 -0
  16. checkpoint-300/chat_template.jinja +54 -0
  17. checkpoint-300/merges.txt +0 -0
  18. checkpoint-300/rng_state_0.pth +3 -0
  19. checkpoint-300/rng_state_2.pth +3 -0
  20. checkpoint-300/special_tokens_map.json +31 -0
  21. checkpoint-300/tokenizer_config.json +209 -0
  22. checkpoint-400/chat_template.jinja +54 -0
  23. checkpoint-400/generation_config.json +9 -0
  24. checkpoint-400/merges.txt +0 -0
  25. checkpoint-400/model.safetensors.index.json +347 -0
  26. checkpoint-400/rng_state_1.pth +3 -0
  27. checkpoint-400/special_tokens_map.json +31 -0
  28. checkpoint-400/tokenizer.json +3 -0
  29. checkpoint-400/tokenizer_config.json +209 -0
  30. checkpoint-400/trainer_state.json +2834 -0
  31. checkpoint-500/added_tokens.json +24 -0
  32. checkpoint-500/chat_template.jinja +54 -0
  33. checkpoint-500/config.json +59 -0
  34. checkpoint-500/generation_config.json +9 -0
  35. checkpoint-500/merges.txt +0 -0
  36. checkpoint-500/model.safetensors.index.json +347 -0
  37. checkpoint-500/special_tokens_map.json +31 -0
  38. checkpoint-500/tokenizer_config.json +209 -0
  39. checkpoint-500/trainer_state.json +3534 -0
  40. checkpoint-600/added_tokens.json +24 -0
  41. checkpoint-600/chat_template.jinja +54 -0
  42. checkpoint-600/config.json +59 -0
  43. checkpoint-600/generation_config.json +9 -0
  44. checkpoint-600/model.safetensors.index.json +347 -0
  45. checkpoint-700/rng_state_2.pth +3 -0
  46. checkpoint-700/rng_state_3.pth +3 -0
  47. checkpoint-700/tokenizer.json +3 -0
  48. checkpoint-748/rng_state_0.pth +3 -0
  49. checkpoint-748/rng_state_2.pth +3 -0
  50. checkpoint-748/rng_state_3.pth +3 -0
.gitattributes CHANGED
@@ -33,3 +33,5 @@ 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
+ checkpoint-400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ checkpoint-700/tokenizer.json filter=lfs diff=lfs merge=lfs -text
checkpoint-100/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f91f87bd9a92e8739289f46162ef308559319763234c64c5b77db9ec569f87f
3
+ size 15429
checkpoint-100/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7d39b8b541e3fd455fdaac992f8e531a8b6ecbd342d84bd34ce103caf8c4829
3
+ size 15429
checkpoint-100/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ff092a382271ec0416bca1eaebe1589a5c027148f661f515a99be89a41550e9
3
+ size 15429
checkpoint-100/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ee31a8654359133d1c40cfb2da8b7350b44cc7eec9bf923c997643420f7061d
3
+ size 15429
checkpoint-100/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a541d9ad6f33c9f9690a5a1e225ae53c29f7e864e21c60b267d472f9e96ea6d1
3
+ size 1465
checkpoint-100/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17bb87c35a490ad6ee546a7a2cd412da9e702f2f3558c3d94f76ad65329cc535
3
+ size 6225
checkpoint-200/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f91f87bd9a92e8739289f46162ef308559319763234c64c5b77db9ec569f87f
3
+ size 15429
checkpoint-200/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7d39b8b541e3fd455fdaac992f8e531a8b6ecbd342d84bd34ce103caf8c4829
3
+ size 15429
checkpoint-200/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ff092a382271ec0416bca1eaebe1589a5c027148f661f515a99be89a41550e9
3
+ size 15429
checkpoint-200/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ee31a8654359133d1c40cfb2da8b7350b44cc7eec9bf923c997643420f7061d
3
+ size 15429
checkpoint-200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a89a08bb3bf6b2953622f49ee6efada8398e024dafb73fb354435650f5f6727c
3
+ size 1465
checkpoint-200/tokenizer_config.json ADDED
@@ -0,0 +1,209 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": false,
5
+ "added_tokens_decoder": {
6
+ "151643": {
7
+ "content": "<|endoftext|>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "151644": {
15
+ "content": "<|im_start|>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "151645": {
23
+ "content": "<|im_end|>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "151646": {
31
+ "content": "<|object_ref_start|>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "151647": {
39
+ "content": "<|object_ref_end|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "151648": {
47
+ "content": "<|box_start|>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "151649": {
55
+ "content": "<|box_end|>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "151650": {
63
+ "content": "<|quad_start|>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "151651": {
71
+ "content": "<|quad_end|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "151652": {
79
+ "content": "<|vision_start|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "151653": {
87
+ "content": "<|vision_end|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "151654": {
95
+ "content": "<|vision_pad|>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": true
101
+ },
102
+ "151655": {
103
+ "content": "<|image_pad|>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": true
109
+ },
110
+ "151656": {
111
+ "content": "<|video_pad|>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": true
117
+ },
118
+ "151657": {
119
+ "content": "<tool_call>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": false
125
+ },
126
+ "151658": {
127
+ "content": "</tool_call>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": false
133
+ },
134
+ "151659": {
135
+ "content": "<|fim_prefix|>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": false
141
+ },
142
+ "151660": {
143
+ "content": "<|fim_middle|>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": false
149
+ },
150
+ "151661": {
151
+ "content": "<|fim_suffix|>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": false
157
+ },
158
+ "151662": {
159
+ "content": "<|fim_pad|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": false
165
+ },
166
+ "151663": {
167
+ "content": "<|repo_name|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": false
173
+ },
174
+ "151664": {
175
+ "content": "<|file_sep|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": false
181
+ }
182
+ },
183
+ "additional_special_tokens": [
184
+ "<|im_start|>",
185
+ "<|im_end|>",
186
+ "<|object_ref_start|>",
187
+ "<|object_ref_end|>",
188
+ "<|box_start|>",
189
+ "<|box_end|>",
190
+ "<|quad_start|>",
191
+ "<|quad_end|>",
192
+ "<|vision_start|>",
193
+ "<|vision_end|>",
194
+ "<|vision_pad|>",
195
+ "<|image_pad|>",
196
+ "<|video_pad|>"
197
+ ],
198
+ "bos_token": null,
199
+ "clean_up_tokenization_spaces": false,
200
+ "eos_token": "<|im_end|>",
201
+ "errors": "replace",
202
+ "extra_special_tokens": {},
203
+ "model_max_length": 131072,
204
+ "pad_token": "<|endoftext|>",
205
+ "padding_side": "right",
206
+ "split_special_tokens": false,
207
+ "tokenizer_class": "Qwen2Tokenizer",
208
+ "unk_token": null
209
+ }
checkpoint-200/trainer_state.json ADDED
@@ -0,0 +1,1434 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.5352960856473737,
6
+ "eval_steps": 500,
7
+ "global_step": 200,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0026764804282368685,
14
+ "grad_norm": 416.7599792480469,
15
+ "learning_rate": 0.0,
16
+ "loss": 3.2698,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.005352960856473737,
21
+ "grad_norm": 316.08477783203125,
22
+ "learning_rate": 2.0000000000000002e-07,
23
+ "loss": 2.9631,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.008029441284710606,
28
+ "grad_norm": 472.76885986328125,
29
+ "learning_rate": 4.0000000000000003e-07,
30
+ "loss": 3.9863,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.010705921712947474,
35
+ "grad_norm": 251.41015625,
36
+ "learning_rate": 6.000000000000001e-07,
37
+ "loss": 3.2654,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.013382402141184342,
42
+ "grad_norm": 217.40838623046875,
43
+ "learning_rate": 8.000000000000001e-07,
44
+ "loss": 3.4505,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.01605888256942121,
49
+ "grad_norm": 136.2932586669922,
50
+ "learning_rate": 1.0000000000000002e-06,
51
+ "loss": 3.2498,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.01873536299765808,
56
+ "grad_norm": 135.7803192138672,
57
+ "learning_rate": 1.2000000000000002e-06,
58
+ "loss": 3.0983,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.021411843425894948,
63
+ "grad_norm": 93.63102722167969,
64
+ "learning_rate": 1.4000000000000001e-06,
65
+ "loss": 2.965,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.024088323854131816,
70
+ "grad_norm": 82.7247085571289,
71
+ "learning_rate": 1.6000000000000001e-06,
72
+ "loss": 2.6341,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.026764804282368684,
77
+ "grad_norm": 94.11923217773438,
78
+ "learning_rate": 1.8000000000000001e-06,
79
+ "loss": 3.2072,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.029441284710605555,
84
+ "grad_norm": 65.76191711425781,
85
+ "learning_rate": 2.0000000000000003e-06,
86
+ "loss": 2.6866,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.03211776513884242,
91
+ "grad_norm": 89.41738891601562,
92
+ "learning_rate": 2.2e-06,
93
+ "loss": 2.8421,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.03479424556707929,
98
+ "grad_norm": 82.1783676147461,
99
+ "learning_rate": 2.4000000000000003e-06,
100
+ "loss": 2.5489,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.03747072599531616,
105
+ "grad_norm": 72.41712951660156,
106
+ "learning_rate": 2.6e-06,
107
+ "loss": 2.9365,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.04014720642355303,
112
+ "grad_norm": 77.541015625,
113
+ "learning_rate": 2.8000000000000003e-06,
114
+ "loss": 2.0516,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.042823686851789895,
119
+ "grad_norm": 125.65862274169922,
120
+ "learning_rate": 3e-06,
121
+ "loss": 2.7676,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.04550016728002677,
126
+ "grad_norm": 94.85375213623047,
127
+ "learning_rate": 3.2000000000000003e-06,
128
+ "loss": 2.5114,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.04817664770826363,
133
+ "grad_norm": 77.89604187011719,
134
+ "learning_rate": 3.4000000000000005e-06,
135
+ "loss": 2.536,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.0508531281365005,
140
+ "grad_norm": 59.268680572509766,
141
+ "learning_rate": 3.6000000000000003e-06,
142
+ "loss": 2.1877,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.05352960856473737,
147
+ "grad_norm": 57.32849884033203,
148
+ "learning_rate": 3.8000000000000005e-06,
149
+ "loss": 2.2084,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.05620608899297424,
154
+ "grad_norm": 58.27803421020508,
155
+ "learning_rate": 4.000000000000001e-06,
156
+ "loss": 2.191,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.05888256942121111,
161
+ "grad_norm": 32.10304260253906,
162
+ "learning_rate": 4.2000000000000004e-06,
163
+ "loss": 2.0009,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.061559049849447975,
168
+ "grad_norm": 17.106046676635742,
169
+ "learning_rate": 4.4e-06,
170
+ "loss": 1.7496,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.06423553027768485,
175
+ "grad_norm": 16.371326446533203,
176
+ "learning_rate": 4.600000000000001e-06,
177
+ "loss": 2.0558,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.06691201070592172,
182
+ "grad_norm": 14.423291206359863,
183
+ "learning_rate": 4.800000000000001e-06,
184
+ "loss": 2.1511,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.06958849113415858,
189
+ "grad_norm": 16.198841094970703,
190
+ "learning_rate": 5e-06,
191
+ "loss": 1.9571,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.07226497156239545,
196
+ "grad_norm": 14.810019493103027,
197
+ "learning_rate": 5.2e-06,
198
+ "loss": 1.9937,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.07494145199063232,
203
+ "grad_norm": 13.464497566223145,
204
+ "learning_rate": 5.400000000000001e-06,
205
+ "loss": 1.7204,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.07761793241886919,
210
+ "grad_norm": 14.364913940429688,
211
+ "learning_rate": 5.600000000000001e-06,
212
+ "loss": 1.7,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.08029441284710606,
217
+ "grad_norm": 12.647878646850586,
218
+ "learning_rate": 5.8e-06,
219
+ "loss": 1.7034,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.08297089327534292,
224
+ "grad_norm": 12.305310249328613,
225
+ "learning_rate": 6e-06,
226
+ "loss": 1.7873,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.08564737370357979,
231
+ "grad_norm": 9.316176414489746,
232
+ "learning_rate": 6.200000000000001e-06,
233
+ "loss": 1.7204,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.08832385413181666,
238
+ "grad_norm": 10.51429271697998,
239
+ "learning_rate": 6.4000000000000006e-06,
240
+ "loss": 1.401,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.09100033456005353,
245
+ "grad_norm": 10.570704460144043,
246
+ "learning_rate": 6.600000000000001e-06,
247
+ "loss": 1.6091,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.0936768149882904,
252
+ "grad_norm": 12.805994987487793,
253
+ "learning_rate": 6.800000000000001e-06,
254
+ "loss": 1.5669,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.09635329541652726,
259
+ "grad_norm": 10.955977439880371,
260
+ "learning_rate": 7e-06,
261
+ "loss": 1.4644,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.09902977584476413,
266
+ "grad_norm": 15.71002197265625,
267
+ "learning_rate": 7.2000000000000005e-06,
268
+ "loss": 1.5497,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.101706256273001,
273
+ "grad_norm": 11.375880241394043,
274
+ "learning_rate": 7.4e-06,
275
+ "loss": 1.5646,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.10438273670123788,
280
+ "grad_norm": 12.987727165222168,
281
+ "learning_rate": 7.600000000000001e-06,
282
+ "loss": 1.4149,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.10705921712947473,
287
+ "grad_norm": 10.491798400878906,
288
+ "learning_rate": 7.800000000000002e-06,
289
+ "loss": 1.5017,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.1097356975577116,
294
+ "grad_norm": 12.43316650390625,
295
+ "learning_rate": 8.000000000000001e-06,
296
+ "loss": 1.4759,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.11241217798594848,
301
+ "grad_norm": 12.077725410461426,
302
+ "learning_rate": 8.2e-06,
303
+ "loss": 1.4892,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.11508865841418535,
308
+ "grad_norm": 12.642361640930176,
309
+ "learning_rate": 8.400000000000001e-06,
310
+ "loss": 1.458,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.11776513884242222,
315
+ "grad_norm": 15.535161018371582,
316
+ "learning_rate": 8.6e-06,
317
+ "loss": 1.6861,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.12044161927065908,
322
+ "grad_norm": 11.678996086120605,
323
+ "learning_rate": 8.8e-06,
324
+ "loss": 1.6968,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.12311809969889595,
329
+ "grad_norm": 11.906571388244629,
330
+ "learning_rate": 9e-06,
331
+ "loss": 1.44,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.12579458012713282,
336
+ "grad_norm": 11.492127418518066,
337
+ "learning_rate": 9.200000000000002e-06,
338
+ "loss": 1.3714,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.1284710605553697,
343
+ "grad_norm": 15.786943435668945,
344
+ "learning_rate": 9.4e-06,
345
+ "loss": 1.4132,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.13114754098360656,
350
+ "grad_norm": 12.31556224822998,
351
+ "learning_rate": 9.600000000000001e-06,
352
+ "loss": 1.166,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.13382402141184344,
357
+ "grad_norm": 12.496630668640137,
358
+ "learning_rate": 9.800000000000001e-06,
359
+ "loss": 1.2859,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.1365005018400803,
364
+ "grad_norm": 11.003588676452637,
365
+ "learning_rate": 1e-05,
366
+ "loss": 1.2614,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.13917698226831715,
371
+ "grad_norm": 12.946131706237793,
372
+ "learning_rate": 9.99994935597953e-06,
373
+ "loss": 1.4555,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.14185346269655402,
378
+ "grad_norm": 13.676822662353516,
379
+ "learning_rate": 9.999797424944041e-06,
380
+ "loss": 1.2167,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.1445299431247909,
385
+ "grad_norm": 14.827473640441895,
386
+ "learning_rate": 9.999544209971299e-06,
387
+ "loss": 1.3456,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.14720642355302777,
392
+ "grad_norm": 11.921989440917969,
393
+ "learning_rate": 9.99918971619083e-06,
394
+ "loss": 1.2535,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.14988290398126464,
399
+ "grad_norm": 12.726682662963867,
400
+ "learning_rate": 9.99873395078383e-06,
401
+ "loss": 1.2172,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.1525593844095015,
406
+ "grad_norm": 18.52472496032715,
407
+ "learning_rate": 9.998176922983017e-06,
408
+ "loss": 1.4418,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.15523586483773838,
413
+ "grad_norm": 13.32449722290039,
414
+ "learning_rate": 9.99751864407244e-06,
415
+ "loss": 1.18,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.15791234526597525,
420
+ "grad_norm": 10.778595924377441,
421
+ "learning_rate": 9.996759127387259e-06,
422
+ "loss": 1.1879,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.16058882569421212,
427
+ "grad_norm": 10.155744552612305,
428
+ "learning_rate": 9.99589838831346e-06,
429
+ "loss": 1.1274,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.16326530612244897,
434
+ "grad_norm": 13.751497268676758,
435
+ "learning_rate": 9.994936444287565e-06,
436
+ "loss": 1.1499,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.16594178655068584,
441
+ "grad_norm": 17.752635955810547,
442
+ "learning_rate": 9.993873314796253e-06,
443
+ "loss": 1.1425,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.1686182669789227,
448
+ "grad_norm": 13.581279754638672,
449
+ "learning_rate": 9.992709021375987e-06,
450
+ "loss": 1.0341,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.17129474740715958,
455
+ "grad_norm": 11.867856979370117,
456
+ "learning_rate": 9.991443587612568e-06,
457
+ "loss": 1.024,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.17397122783539645,
462
+ "grad_norm": 13.20876693725586,
463
+ "learning_rate": 9.990077039140655e-06,
464
+ "loss": 1.0375,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.17664770826363332,
469
+ "grad_norm": 12.211270332336426,
470
+ "learning_rate": 9.988609403643254e-06,
471
+ "loss": 1.1334,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.1793241886918702,
476
+ "grad_norm": 11.89450454711914,
477
+ "learning_rate": 9.987040710851148e-06,
478
+ "loss": 1.0447,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.18200066912010707,
483
+ "grad_norm": 10.132635116577148,
484
+ "learning_rate": 9.9853709925423e-06,
485
+ "loss": 0.7626,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.18467714954834394,
490
+ "grad_norm": 9.661967277526855,
491
+ "learning_rate": 9.983600282541213e-06,
492
+ "loss": 0.968,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.1873536299765808,
497
+ "grad_norm": 11.19580078125,
498
+ "learning_rate": 9.981728616718234e-06,
499
+ "loss": 0.9161,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.19003011040481765,
504
+ "grad_norm": 9.165005683898926,
505
+ "learning_rate": 9.979756032988837e-06,
506
+ "loss": 0.8821,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.19270659083305453,
511
+ "grad_norm": 9.17795181274414,
512
+ "learning_rate": 9.977682571312847e-06,
513
+ "loss": 1.0452,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.1953830712612914,
518
+ "grad_norm": 7.483065128326416,
519
+ "learning_rate": 9.975508273693643e-06,
520
+ "loss": 1.0135,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.19805955168952827,
525
+ "grad_norm": 7.2489728927612305,
526
+ "learning_rate": 9.97323318417729e-06,
527
+ "loss": 0.9675,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.20073603211776514,
532
+ "grad_norm": 7.151744365692139,
533
+ "learning_rate": 9.970857348851667e-06,
534
+ "loss": 0.829,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.203412512546002,
539
+ "grad_norm": 7.535038948059082,
540
+ "learning_rate": 9.968380815845504e-06,
541
+ "loss": 0.9383,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.20608899297423888,
546
+ "grad_norm": 5.609306812286377,
547
+ "learning_rate": 9.965803635327445e-06,
548
+ "loss": 0.8914,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.20876547340247575,
553
+ "grad_norm": 7.203204154968262,
554
+ "learning_rate": 9.963125859505e-06,
555
+ "loss": 1.0889,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.21144195383071263,
560
+ "grad_norm": 4.907174587249756,
561
+ "learning_rate": 9.960347542623506e-06,
562
+ "loss": 0.7694,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.21411843425894947,
567
+ "grad_norm": 5.539646148681641,
568
+ "learning_rate": 9.95746874096501e-06,
569
+ "loss": 0.9965,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.21679491468718634,
574
+ "grad_norm": 5.0046868324279785,
575
+ "learning_rate": 9.954489512847156e-06,
576
+ "loss": 0.7795,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.2194713951154232,
581
+ "grad_norm": 5.138219833374023,
582
+ "learning_rate": 9.951409918621977e-06,
583
+ "loss": 0.8877,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.22214787554366008,
588
+ "grad_norm": 4.388546466827393,
589
+ "learning_rate": 9.948230020674685e-06,
590
+ "loss": 0.9253,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.22482435597189696,
595
+ "grad_norm": 4.456934928894043,
596
+ "learning_rate": 9.944949883422409e-06,
597
+ "loss": 0.7335,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.22750083640013383,
602
+ "grad_norm": 4.645761013031006,
603
+ "learning_rate": 9.941569573312882e-06,
604
+ "loss": 0.8997,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.2301773168283707,
609
+ "grad_norm": 4.329288482666016,
610
+ "learning_rate": 9.938089158823101e-06,
611
+ "loss": 0.852,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.23285379725660757,
616
+ "grad_norm": 5.044316291809082,
617
+ "learning_rate": 9.934508710457944e-06,
618
+ "loss": 0.8236,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.23553027768484444,
623
+ "grad_norm": 5.356631755828857,
624
+ "learning_rate": 9.930828300748726e-06,
625
+ "loss": 0.9572,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.23820675811308128,
630
+ "grad_norm": 3.933661699295044,
631
+ "learning_rate": 9.927048004251748e-06,
632
+ "loss": 0.8479,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.24088323854131816,
637
+ "grad_norm": 5.228504180908203,
638
+ "learning_rate": 9.923167897546773e-06,
639
+ "loss": 0.741,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.24355971896955503,
644
+ "grad_norm": 4.520639419555664,
645
+ "learning_rate": 9.919188059235483e-06,
646
+ "loss": 0.746,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.2462361993977919,
651
+ "grad_norm": 4.191668510437012,
652
+ "learning_rate": 9.915108569939884e-06,
653
+ "loss": 0.6904,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.24891267982602877,
658
+ "grad_norm": 4.300325393676758,
659
+ "learning_rate": 9.910929512300673e-06,
660
+ "loss": 0.8753,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.25158916025426564,
665
+ "grad_norm": 5.979667663574219,
666
+ "learning_rate": 9.90665097097556e-06,
667
+ "loss": 0.7355,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.2542656406825025,
672
+ "grad_norm": 5.374980449676514,
673
+ "learning_rate": 9.902273032637558e-06,
674
+ "loss": 0.7932,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.2569421211107394,
679
+ "grad_norm": 3.9075119495391846,
680
+ "learning_rate": 9.897795785973227e-06,
681
+ "loss": 0.9069,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.25961860153897626,
686
+ "grad_norm": 4.439886569976807,
687
+ "learning_rate": 9.89321932168088e-06,
688
+ "loss": 0.8408,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.26229508196721313,
693
+ "grad_norm": 4.403755187988281,
694
+ "learning_rate": 9.888543732468732e-06,
695
+ "loss": 0.709,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.26497156239545,
700
+ "grad_norm": 5.023388862609863,
701
+ "learning_rate": 9.883769113053039e-06,
702
+ "loss": 0.8928,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.26764804282368687,
707
+ "grad_norm": 5.586976528167725,
708
+ "learning_rate": 9.878895560156172e-06,
709
+ "loss": 0.9644,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 0.27032452325192374,
714
+ "grad_norm": 15.905163764953613,
715
+ "learning_rate": 9.873923172504653e-06,
716
+ "loss": 0.8362,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 0.2730010036801606,
721
+ "grad_norm": 5.6663432121276855,
722
+ "learning_rate": 9.868852050827167e-06,
723
+ "loss": 0.7585,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 0.27567748410839743,
728
+ "grad_norm": 4.5769124031066895,
729
+ "learning_rate": 9.863682297852506e-06,
730
+ "loss": 0.9748,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 0.2783539645366343,
735
+ "grad_norm": 4.305916786193848,
736
+ "learning_rate": 9.858414018307503e-06,
737
+ "loss": 0.805,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 0.2810304449648712,
742
+ "grad_norm": 4.756903648376465,
743
+ "learning_rate": 9.853047318914898e-06,
744
+ "loss": 0.8221,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 0.28370692539310804,
749
+ "grad_norm": 5.709683418273926,
750
+ "learning_rate": 9.847582308391189e-06,
751
+ "loss": 0.7416,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 0.2863834058213449,
756
+ "grad_norm": 3.933176279067993,
757
+ "learning_rate": 9.842019097444414e-06,
758
+ "loss": 0.7298,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 0.2890598862495818,
763
+ "grad_norm": 4.9454264640808105,
764
+ "learning_rate": 9.836357798771922e-06,
765
+ "loss": 0.7683,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 0.29173636667781866,
770
+ "grad_norm": 4.274379730224609,
771
+ "learning_rate": 9.830598527058083e-06,
772
+ "loss": 0.7788,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 0.29441284710605553,
777
+ "grad_norm": 3.6148972511291504,
778
+ "learning_rate": 9.824741398971966e-06,
779
+ "loss": 0.6941,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 0.2970893275342924,
784
+ "grad_norm": 3.8976197242736816,
785
+ "learning_rate": 9.81878653316498e-06,
786
+ "loss": 0.8783,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 0.2997658079625293,
791
+ "grad_norm": 4.712560176849365,
792
+ "learning_rate": 9.81273405026846e-06,
793
+ "loss": 0.9022,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 0.30244228839076615,
798
+ "grad_norm": 5.358889579772949,
799
+ "learning_rate": 9.806584072891234e-06,
800
+ "loss": 0.8587,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 0.305118768819003,
805
+ "grad_norm": 4.296327114105225,
806
+ "learning_rate": 9.800336725617136e-06,
807
+ "loss": 0.8813,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 0.3077952492472399,
812
+ "grad_norm": 4.895114421844482,
813
+ "learning_rate": 9.793992135002476e-06,
814
+ "loss": 0.8184,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 0.31047172967547676,
819
+ "grad_norm": 4.782822132110596,
820
+ "learning_rate": 9.787550429573487e-06,
821
+ "loss": 0.6049,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 0.31314821010371363,
826
+ "grad_norm": 5.41484260559082,
827
+ "learning_rate": 9.781011739823715e-06,
828
+ "loss": 0.7804,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 0.3158246905319505,
833
+ "grad_norm": 3.639780044555664,
834
+ "learning_rate": 9.77437619821137e-06,
835
+ "loss": 0.7241,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 0.3185011709601874,
840
+ "grad_norm": 4.5406880378723145,
841
+ "learning_rate": 9.767643939156658e-06,
842
+ "loss": 0.8398,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 0.32117765138842425,
847
+ "grad_norm": 4.575490951538086,
848
+ "learning_rate": 9.760815099039045e-06,
849
+ "loss": 0.6951,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 0.3238541318166611,
854
+ "grad_norm": 3.868921995162964,
855
+ "learning_rate": 9.753889816194498e-06,
856
+ "loss": 0.706,
857
+ "step": 121
858
+ },
859
+ {
860
+ "epoch": 0.32653061224489793,
861
+ "grad_norm": 4.064291477203369,
862
+ "learning_rate": 9.746868230912683e-06,
863
+ "loss": 0.7386,
864
+ "step": 122
865
+ },
866
+ {
867
+ "epoch": 0.3292070926731348,
868
+ "grad_norm": 4.059055805206299,
869
+ "learning_rate": 9.739750485434126e-06,
870
+ "loss": 0.7171,
871
+ "step": 123
872
+ },
873
+ {
874
+ "epoch": 0.3318835731013717,
875
+ "grad_norm": 4.522129535675049,
876
+ "learning_rate": 9.73253672394732e-06,
877
+ "loss": 0.6679,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 0.33456005352960855,
882
+ "grad_norm": 3.669271945953369,
883
+ "learning_rate": 9.725227092585824e-06,
884
+ "loss": 0.7409,
885
+ "step": 125
886
+ },
887
+ {
888
+ "epoch": 0.3372365339578454,
889
+ "grad_norm": 3.603746175765991,
890
+ "learning_rate": 9.717821739425286e-06,
891
+ "loss": 0.7286,
892
+ "step": 126
893
+ },
894
+ {
895
+ "epoch": 0.3399130143860823,
896
+ "grad_norm": 4.118206977844238,
897
+ "learning_rate": 9.710320814480448e-06,
898
+ "loss": 0.7164,
899
+ "step": 127
900
+ },
901
+ {
902
+ "epoch": 0.34258949481431916,
903
+ "grad_norm": 3.6243247985839844,
904
+ "learning_rate": 9.702724469702107e-06,
905
+ "loss": 0.7772,
906
+ "step": 128
907
+ },
908
+ {
909
+ "epoch": 0.34526597524255603,
910
+ "grad_norm": 3.8515329360961914,
911
+ "learning_rate": 9.695032858974042e-06,
912
+ "loss": 0.7173,
913
+ "step": 129
914
+ },
915
+ {
916
+ "epoch": 0.3479424556707929,
917
+ "grad_norm": 4.385611534118652,
918
+ "learning_rate": 9.687246138109888e-06,
919
+ "loss": 0.8672,
920
+ "step": 130
921
+ },
922
+ {
923
+ "epoch": 0.3506189360990298,
924
+ "grad_norm": 3.6981892585754395,
925
+ "learning_rate": 9.679364464849983e-06,
926
+ "loss": 0.6571,
927
+ "step": 131
928
+ },
929
+ {
930
+ "epoch": 0.35329541652726665,
931
+ "grad_norm": 4.061744213104248,
932
+ "learning_rate": 9.671387998858178e-06,
933
+ "loss": 0.7114,
934
+ "step": 132
935
+ },
936
+ {
937
+ "epoch": 0.3559718969555035,
938
+ "grad_norm": 4.793642997741699,
939
+ "learning_rate": 9.663316901718599e-06,
940
+ "loss": 0.7537,
941
+ "step": 133
942
+ },
943
+ {
944
+ "epoch": 0.3586483773837404,
945
+ "grad_norm": 3.751180648803711,
946
+ "learning_rate": 9.655151336932362e-06,
947
+ "loss": 0.6826,
948
+ "step": 134
949
+ },
950
+ {
951
+ "epoch": 0.36132485781197726,
952
+ "grad_norm": 3.8843295574188232,
953
+ "learning_rate": 9.646891469914285e-06,
954
+ "loss": 0.6868,
955
+ "step": 135
956
+ },
957
+ {
958
+ "epoch": 0.36400133824021413,
959
+ "grad_norm": 4.41197395324707,
960
+ "learning_rate": 9.638537467989517e-06,
961
+ "loss": 0.7224,
962
+ "step": 136
963
+ },
964
+ {
965
+ "epoch": 0.366677818668451,
966
+ "grad_norm": 3.2272307872772217,
967
+ "learning_rate": 9.630089500390154e-06,
968
+ "loss": 0.5735,
969
+ "step": 137
970
+ },
971
+ {
972
+ "epoch": 0.3693542990966879,
973
+ "grad_norm": 3.5232367515563965,
974
+ "learning_rate": 9.621547738251816e-06,
975
+ "loss": 0.7621,
976
+ "step": 138
977
+ },
978
+ {
979
+ "epoch": 0.37203077952492475,
980
+ "grad_norm": 3.4856607913970947,
981
+ "learning_rate": 9.61291235461017e-06,
982
+ "loss": 0.6517,
983
+ "step": 139
984
+ },
985
+ {
986
+ "epoch": 0.3747072599531616,
987
+ "grad_norm": 3.958531379699707,
988
+ "learning_rate": 9.604183524397439e-06,
989
+ "loss": 0.8994,
990
+ "step": 140
991
+ },
992
+ {
993
+ "epoch": 0.37738374038139844,
994
+ "grad_norm": 3.3670690059661865,
995
+ "learning_rate": 9.595361424438841e-06,
996
+ "loss": 0.5975,
997
+ "step": 141
998
+ },
999
+ {
1000
+ "epoch": 0.3800602208096353,
1001
+ "grad_norm": 4.388239860534668,
1002
+ "learning_rate": 9.586446233449024e-06,
1003
+ "loss": 0.7168,
1004
+ "step": 142
1005
+ },
1006
+ {
1007
+ "epoch": 0.3827367012378722,
1008
+ "grad_norm": 3.513667345046997,
1009
+ "learning_rate": 9.577438132028431e-06,
1010
+ "loss": 0.7645,
1011
+ "step": 143
1012
+ },
1013
+ {
1014
+ "epoch": 0.38541318166610905,
1015
+ "grad_norm": 3.383009195327759,
1016
+ "learning_rate": 9.568337302659652e-06,
1017
+ "loss": 0.7159,
1018
+ "step": 144
1019
+ },
1020
+ {
1021
+ "epoch": 0.3880896620943459,
1022
+ "grad_norm": 3.7690727710723877,
1023
+ "learning_rate": 9.559143929703724e-06,
1024
+ "loss": 0.8527,
1025
+ "step": 145
1026
+ },
1027
+ {
1028
+ "epoch": 0.3907661425225828,
1029
+ "grad_norm": 3.1912662982940674,
1030
+ "learning_rate": 9.549858199396394e-06,
1031
+ "loss": 0.6288,
1032
+ "step": 146
1033
+ },
1034
+ {
1035
+ "epoch": 0.39344262295081966,
1036
+ "grad_norm": 3.616091012954712,
1037
+ "learning_rate": 9.540480299844345e-06,
1038
+ "loss": 0.6928,
1039
+ "step": 147
1040
+ },
1041
+ {
1042
+ "epoch": 0.39611910337905654,
1043
+ "grad_norm": 3.7126312255859375,
1044
+ "learning_rate": 9.531010421021396e-06,
1045
+ "loss": 0.7583,
1046
+ "step": 148
1047
+ },
1048
+ {
1049
+ "epoch": 0.3987955838072934,
1050
+ "grad_norm": 3.61950421333313,
1051
+ "learning_rate": 9.52144875476464e-06,
1052
+ "loss": 0.6796,
1053
+ "step": 149
1054
+ },
1055
+ {
1056
+ "epoch": 0.4014720642355303,
1057
+ "grad_norm": 3.0952789783477783,
1058
+ "learning_rate": 9.511795494770563e-06,
1059
+ "loss": 0.6761,
1060
+ "step": 150
1061
+ },
1062
+ {
1063
+ "epoch": 0.40414854466376715,
1064
+ "grad_norm": 3.4935519695281982,
1065
+ "learning_rate": 9.50205083659113e-06,
1066
+ "loss": 0.622,
1067
+ "step": 151
1068
+ },
1069
+ {
1070
+ "epoch": 0.406825025092004,
1071
+ "grad_norm": 4.367190837860107,
1072
+ "learning_rate": 9.492214977629804e-06,
1073
+ "loss": 0.8146,
1074
+ "step": 152
1075
+ },
1076
+ {
1077
+ "epoch": 0.4095015055202409,
1078
+ "grad_norm": 3.501554489135742,
1079
+ "learning_rate": 9.482288117137561e-06,
1080
+ "loss": 0.6101,
1081
+ "step": 153
1082
+ },
1083
+ {
1084
+ "epoch": 0.41217798594847777,
1085
+ "grad_norm": 3.7483065128326416,
1086
+ "learning_rate": 9.472270456208856e-06,
1087
+ "loss": 0.6095,
1088
+ "step": 154
1089
+ },
1090
+ {
1091
+ "epoch": 0.41485446637671464,
1092
+ "grad_norm": 2.7559807300567627,
1093
+ "learning_rate": 9.462162197777533e-06,
1094
+ "loss": 0.6167,
1095
+ "step": 155
1096
+ },
1097
+ {
1098
+ "epoch": 0.4175309468049515,
1099
+ "grad_norm": 3.520737409591675,
1100
+ "learning_rate": 9.451963546612737e-06,
1101
+ "loss": 0.5388,
1102
+ "step": 156
1103
+ },
1104
+ {
1105
+ "epoch": 0.4202074272331884,
1106
+ "grad_norm": 4.042897701263428,
1107
+ "learning_rate": 9.441674709314743e-06,
1108
+ "loss": 0.7228,
1109
+ "step": 157
1110
+ },
1111
+ {
1112
+ "epoch": 0.42288390766142525,
1113
+ "grad_norm": 4.090766429901123,
1114
+ "learning_rate": 9.431295894310786e-06,
1115
+ "loss": 0.5569,
1116
+ "step": 158
1117
+ },
1118
+ {
1119
+ "epoch": 0.42556038808966207,
1120
+ "grad_norm": 4.141130447387695,
1121
+ "learning_rate": 9.420827311850836e-06,
1122
+ "loss": 0.6297,
1123
+ "step": 159
1124
+ },
1125
+ {
1126
+ "epoch": 0.42823686851789894,
1127
+ "grad_norm": 3.472973346710205,
1128
+ "learning_rate": 9.410269174003333e-06,
1129
+ "loss": 0.6868,
1130
+ "step": 160
1131
+ },
1132
+ {
1133
+ "epoch": 0.4309133489461358,
1134
+ "grad_norm": 3.8000664710998535,
1135
+ "learning_rate": 9.399621694650898e-06,
1136
+ "loss": 0.6547,
1137
+ "step": 161
1138
+ },
1139
+ {
1140
+ "epoch": 0.4335898293743727,
1141
+ "grad_norm": 4.582136154174805,
1142
+ "learning_rate": 9.388885089485995e-06,
1143
+ "loss": 0.7552,
1144
+ "step": 162
1145
+ },
1146
+ {
1147
+ "epoch": 0.43626630980260955,
1148
+ "grad_norm": 3.4165501594543457,
1149
+ "learning_rate": 9.378059576006567e-06,
1150
+ "loss": 0.622,
1151
+ "step": 163
1152
+ },
1153
+ {
1154
+ "epoch": 0.4389427902308464,
1155
+ "grad_norm": 3.642669916152954,
1156
+ "learning_rate": 9.36714537351162e-06,
1157
+ "loss": 0.6302,
1158
+ "step": 164
1159
+ },
1160
+ {
1161
+ "epoch": 0.4416192706590833,
1162
+ "grad_norm": 3.53709077835083,
1163
+ "learning_rate": 9.356142703096793e-06,
1164
+ "loss": 0.5357,
1165
+ "step": 165
1166
+ },
1167
+ {
1168
+ "epoch": 0.44429575108732017,
1169
+ "grad_norm": 3.6043941974639893,
1170
+ "learning_rate": 9.345051787649877e-06,
1171
+ "loss": 0.6978,
1172
+ "step": 166
1173
+ },
1174
+ {
1175
+ "epoch": 0.44697223151555704,
1176
+ "grad_norm": 4.232686519622803,
1177
+ "learning_rate": 9.333872851846285e-06,
1178
+ "loss": 0.6698,
1179
+ "step": 167
1180
+ },
1181
+ {
1182
+ "epoch": 0.4496487119437939,
1183
+ "grad_norm": 3.1254665851593018,
1184
+ "learning_rate": 9.322606122144524e-06,
1185
+ "loss": 0.5173,
1186
+ "step": 168
1187
+ },
1188
+ {
1189
+ "epoch": 0.4523251923720308,
1190
+ "grad_norm": 4.0060930252075195,
1191
+ "learning_rate": 9.311251826781587e-06,
1192
+ "loss": 0.6674,
1193
+ "step": 169
1194
+ },
1195
+ {
1196
+ "epoch": 0.45500167280026765,
1197
+ "grad_norm": 4.720844268798828,
1198
+ "learning_rate": 9.299810195768341e-06,
1199
+ "loss": 0.8349,
1200
+ "step": 170
1201
+ },
1202
+ {
1203
+ "epoch": 0.4576781532285045,
1204
+ "grad_norm": 4.017560005187988,
1205
+ "learning_rate": 9.288281460884864e-06,
1206
+ "loss": 0.8489,
1207
+ "step": 171
1208
+ },
1209
+ {
1210
+ "epoch": 0.4603546336567414,
1211
+ "grad_norm": 4.482756614685059,
1212
+ "learning_rate": 9.276665855675751e-06,
1213
+ "loss": 0.6877,
1214
+ "step": 172
1215
+ },
1216
+ {
1217
+ "epoch": 0.46303111408497827,
1218
+ "grad_norm": 3.9189202785491943,
1219
+ "learning_rate": 9.264963615445378e-06,
1220
+ "loss": 0.58,
1221
+ "step": 173
1222
+ },
1223
+ {
1224
+ "epoch": 0.46570759451321514,
1225
+ "grad_norm": 3.285618543624878,
1226
+ "learning_rate": 9.25317497725315e-06,
1227
+ "loss": 0.585,
1228
+ "step": 174
1229
+ },
1230
+ {
1231
+ "epoch": 0.468384074941452,
1232
+ "grad_norm": 3.4484951496124268,
1233
+ "learning_rate": 9.241300179908672e-06,
1234
+ "loss": 0.5898,
1235
+ "step": 175
1236
+ },
1237
+ {
1238
+ "epoch": 0.4710605553696889,
1239
+ "grad_norm": 3.892336845397949,
1240
+ "learning_rate": 9.229339463966942e-06,
1241
+ "loss": 0.7691,
1242
+ "step": 176
1243
+ },
1244
+ {
1245
+ "epoch": 0.47373703579792575,
1246
+ "grad_norm": 3.3191475868225098,
1247
+ "learning_rate": 9.217293071723455e-06,
1248
+ "loss": 0.6468,
1249
+ "step": 177
1250
+ },
1251
+ {
1252
+ "epoch": 0.47641351622616257,
1253
+ "grad_norm": 3.2376253604888916,
1254
+ "learning_rate": 9.205161247209303e-06,
1255
+ "loss": 0.6233,
1256
+ "step": 178
1257
+ },
1258
+ {
1259
+ "epoch": 0.47908999665439944,
1260
+ "grad_norm": 3.029269218444824,
1261
+ "learning_rate": 9.192944236186237e-06,
1262
+ "loss": 0.5845,
1263
+ "step": 179
1264
+ },
1265
+ {
1266
+ "epoch": 0.4817664770826363,
1267
+ "grad_norm": 3.5170111656188965,
1268
+ "learning_rate": 9.180642286141678e-06,
1269
+ "loss": 0.6815,
1270
+ "step": 180
1271
+ },
1272
+ {
1273
+ "epoch": 0.4844429575108732,
1274
+ "grad_norm": 4.0202155113220215,
1275
+ "learning_rate": 9.16825564628371e-06,
1276
+ "loss": 0.7131,
1277
+ "step": 181
1278
+ },
1279
+ {
1280
+ "epoch": 0.48711943793911006,
1281
+ "grad_norm": 3.566044330596924,
1282
+ "learning_rate": 9.15578456753603e-06,
1283
+ "loss": 0.6628,
1284
+ "step": 182
1285
+ },
1286
+ {
1287
+ "epoch": 0.4897959183673469,
1288
+ "grad_norm": 3.9581563472747803,
1289
+ "learning_rate": 9.143229302532866e-06,
1290
+ "loss": 0.6497,
1291
+ "step": 183
1292
+ },
1293
+ {
1294
+ "epoch": 0.4924723987955838,
1295
+ "grad_norm": 3.2238948345184326,
1296
+ "learning_rate": 9.130590105613854e-06,
1297
+ "loss": 0.5773,
1298
+ "step": 184
1299
+ },
1300
+ {
1301
+ "epoch": 0.49514887922382067,
1302
+ "grad_norm": 4.199437618255615,
1303
+ "learning_rate": 9.117867232818897e-06,
1304
+ "loss": 0.5552,
1305
+ "step": 185
1306
+ },
1307
+ {
1308
+ "epoch": 0.49782535965205754,
1309
+ "grad_norm": 3.223834753036499,
1310
+ "learning_rate": 9.105060941882966e-06,
1311
+ "loss": 0.4985,
1312
+ "step": 186
1313
+ },
1314
+ {
1315
+ "epoch": 0.5005018400802944,
1316
+ "grad_norm": 3.2973475456237793,
1317
+ "learning_rate": 9.092171492230883e-06,
1318
+ "loss": 0.6637,
1319
+ "step": 187
1320
+ },
1321
+ {
1322
+ "epoch": 0.5031783205085313,
1323
+ "grad_norm": 3.9001471996307373,
1324
+ "learning_rate": 9.079199144972072e-06,
1325
+ "loss": 0.7487,
1326
+ "step": 188
1327
+ },
1328
+ {
1329
+ "epoch": 0.5058548009367682,
1330
+ "grad_norm": 3.913593292236328,
1331
+ "learning_rate": 9.066144162895259e-06,
1332
+ "loss": 0.5707,
1333
+ "step": 189
1334
+ },
1335
+ {
1336
+ "epoch": 0.508531281365005,
1337
+ "grad_norm": 2.6204168796539307,
1338
+ "learning_rate": 9.053006810463156e-06,
1339
+ "loss": 0.4685,
1340
+ "step": 190
1341
+ },
1342
+ {
1343
+ "epoch": 0.5112077617932419,
1344
+ "grad_norm": 3.507556200027466,
1345
+ "learning_rate": 9.039787353807101e-06,
1346
+ "loss": 0.5564,
1347
+ "step": 191
1348
+ },
1349
+ {
1350
+ "epoch": 0.5138842422214788,
1351
+ "grad_norm": 3.072106122970581,
1352
+ "learning_rate": 9.026486060721668e-06,
1353
+ "loss": 0.5427,
1354
+ "step": 192
1355
+ },
1356
+ {
1357
+ "epoch": 0.5165607226497156,
1358
+ "grad_norm": 2.578962802886963,
1359
+ "learning_rate": 9.01310320065924e-06,
1360
+ "loss": 0.3969,
1361
+ "step": 193
1362
+ },
1363
+ {
1364
+ "epoch": 0.5192372030779525,
1365
+ "grad_norm": 3.3628005981445312,
1366
+ "learning_rate": 8.999639044724555e-06,
1367
+ "loss": 0.6076,
1368
+ "step": 194
1369
+ },
1370
+ {
1371
+ "epoch": 0.5219136835061894,
1372
+ "grad_norm": 3.64412784576416,
1373
+ "learning_rate": 8.986093865669205e-06,
1374
+ "loss": 0.6826,
1375
+ "step": 195
1376
+ },
1377
+ {
1378
+ "epoch": 0.5245901639344263,
1379
+ "grad_norm": 3.7886106967926025,
1380
+ "learning_rate": 8.972467937886122e-06,
1381
+ "loss": 0.5279,
1382
+ "step": 196
1383
+ },
1384
+ {
1385
+ "epoch": 0.5272666443626631,
1386
+ "grad_norm": 3.543992757797241,
1387
+ "learning_rate": 8.958761537404012e-06,
1388
+ "loss": 0.5748,
1389
+ "step": 197
1390
+ },
1391
+ {
1392
+ "epoch": 0.5299431247909,
1393
+ "grad_norm": 4.291012763977051,
1394
+ "learning_rate": 8.944974941881766e-06,
1395
+ "loss": 0.6919,
1396
+ "step": 198
1397
+ },
1398
+ {
1399
+ "epoch": 0.5326196052191369,
1400
+ "grad_norm": 4.906763553619385,
1401
+ "learning_rate": 8.931108430602834e-06,
1402
+ "loss": 0.5633,
1403
+ "step": 199
1404
+ },
1405
+ {
1406
+ "epoch": 0.5352960856473737,
1407
+ "grad_norm": 3.177494764328003,
1408
+ "learning_rate": 8.917162284469569e-06,
1409
+ "loss": 0.5792,
1410
+ "step": 200
1411
+ }
1412
+ ],
1413
+ "logging_steps": 1,
1414
+ "max_steps": 748,
1415
+ "num_input_tokens_seen": 0,
1416
+ "num_train_epochs": 2,
1417
+ "save_steps": 100,
1418
+ "stateful_callbacks": {
1419
+ "TrainerControl": {
1420
+ "args": {
1421
+ "should_epoch_stop": false,
1422
+ "should_evaluate": false,
1423
+ "should_log": false,
1424
+ "should_save": true,
1425
+ "should_training_stop": false
1426
+ },
1427
+ "attributes": {}
1428
+ }
1429
+ },
1430
+ "total_flos": 2.225232452601774e+17,
1431
+ "train_batch_size": 1,
1432
+ "trial_name": null,
1433
+ "trial_params": null
1434
+ }
checkpoint-200/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-300/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'Please reason step by step, and put your final answer within \\boxed{}.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nPlease reason step by step, and put your final answer within \\boxed{}.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
checkpoint-300/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-300/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f91f87bd9a92e8739289f46162ef308559319763234c64c5b77db9ec569f87f
3
+ size 15429
checkpoint-300/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ff092a382271ec0416bca1eaebe1589a5c027148f661f515a99be89a41550e9
3
+ size 15429
checkpoint-300/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
+ }
checkpoint-300/tokenizer_config.json ADDED
@@ -0,0 +1,209 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": false,
5
+ "added_tokens_decoder": {
6
+ "151643": {
7
+ "content": "<|endoftext|>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "151644": {
15
+ "content": "<|im_start|>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "151645": {
23
+ "content": "<|im_end|>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "151646": {
31
+ "content": "<|object_ref_start|>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "151647": {
39
+ "content": "<|object_ref_end|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "151648": {
47
+ "content": "<|box_start|>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "151649": {
55
+ "content": "<|box_end|>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "151650": {
63
+ "content": "<|quad_start|>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "151651": {
71
+ "content": "<|quad_end|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "151652": {
79
+ "content": "<|vision_start|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "151653": {
87
+ "content": "<|vision_end|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "151654": {
95
+ "content": "<|vision_pad|>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": true
101
+ },
102
+ "151655": {
103
+ "content": "<|image_pad|>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": true
109
+ },
110
+ "151656": {
111
+ "content": "<|video_pad|>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": true
117
+ },
118
+ "151657": {
119
+ "content": "<tool_call>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": false
125
+ },
126
+ "151658": {
127
+ "content": "</tool_call>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": false
133
+ },
134
+ "151659": {
135
+ "content": "<|fim_prefix|>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": false
141
+ },
142
+ "151660": {
143
+ "content": "<|fim_middle|>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": false
149
+ },
150
+ "151661": {
151
+ "content": "<|fim_suffix|>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": false
157
+ },
158
+ "151662": {
159
+ "content": "<|fim_pad|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": false
165
+ },
166
+ "151663": {
167
+ "content": "<|repo_name|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": false
173
+ },
174
+ "151664": {
175
+ "content": "<|file_sep|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": false
181
+ }
182
+ },
183
+ "additional_special_tokens": [
184
+ "<|im_start|>",
185
+ "<|im_end|>",
186
+ "<|object_ref_start|>",
187
+ "<|object_ref_end|>",
188
+ "<|box_start|>",
189
+ "<|box_end|>",
190
+ "<|quad_start|>",
191
+ "<|quad_end|>",
192
+ "<|vision_start|>",
193
+ "<|vision_end|>",
194
+ "<|vision_pad|>",
195
+ "<|image_pad|>",
196
+ "<|video_pad|>"
197
+ ],
198
+ "bos_token": null,
199
+ "clean_up_tokenization_spaces": false,
200
+ "eos_token": "<|im_end|>",
201
+ "errors": "replace",
202
+ "extra_special_tokens": {},
203
+ "model_max_length": 131072,
204
+ "pad_token": "<|endoftext|>",
205
+ "padding_side": "right",
206
+ "split_special_tokens": false,
207
+ "tokenizer_class": "Qwen2Tokenizer",
208
+ "unk_token": null
209
+ }
checkpoint-400/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'Please reason step by step, and put your final answer within \\boxed{}.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nPlease reason step by step, and put your final answer within \\boxed{}.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
checkpoint-400/generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "eos_token_id": [
3
+ 151645,
4
+ 151643
5
+ ],
6
+ "max_new_tokens": 2048,
7
+ "pad_token_id": 151643,
8
+ "transformers_version": "4.57.1"
9
+ }
checkpoint-400/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-400/model.safetensors.index.json ADDED
@@ -0,0 +1,347 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_parameters": 7615616512,
4
+ "total_size": 30462466048
5
+ },
6
+ "weight_map": {
7
+ "lm_head.weight": "model-00007-of-00007.safetensors",
8
+ "model.embed_tokens.weight": "model-00001-of-00007.safetensors",
9
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00007.safetensors",
10
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
11
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
12
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
13
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
14
+ "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
15
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
16
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
17
+ "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
18
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
19
+ "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
20
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
21
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00007.safetensors",
22
+ "model.layers.1.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
23
+ "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
24
+ "model.layers.1.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
25
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
26
+ "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
27
+ "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
28
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
29
+ "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
30
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
31
+ "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
32
+ "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
33
+ "model.layers.10.input_layernorm.weight": "model-00003-of-00007.safetensors",
34
+ "model.layers.10.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
35
+ "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
36
+ "model.layers.10.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
37
+ "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
38
+ "model.layers.10.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
39
+ "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
40
+ "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
41
+ "model.layers.10.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
42
+ "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
43
+ "model.layers.10.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
44
+ "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
45
+ "model.layers.11.input_layernorm.weight": "model-00003-of-00007.safetensors",
46
+ "model.layers.11.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
47
+ "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
48
+ "model.layers.11.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
49
+ "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
50
+ "model.layers.11.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
51
+ "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
52
+ "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
53
+ "model.layers.11.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
54
+ "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
55
+ "model.layers.11.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
56
+ "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
57
+ "model.layers.12.input_layernorm.weight": "model-00003-of-00007.safetensors",
58
+ "model.layers.12.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
59
+ "model.layers.12.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
60
+ "model.layers.12.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
61
+ "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
62
+ "model.layers.12.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
63
+ "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
64
+ "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
65
+ "model.layers.12.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
66
+ "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
67
+ "model.layers.12.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
68
+ "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
69
+ "model.layers.13.input_layernorm.weight": "model-00004-of-00007.safetensors",
70
+ "model.layers.13.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
71
+ "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
72
+ "model.layers.13.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
73
+ "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
74
+ "model.layers.13.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
75
+ "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
76
+ "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
77
+ "model.layers.13.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
78
+ "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
79
+ "model.layers.13.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
80
+ "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
81
+ "model.layers.14.input_layernorm.weight": "model-00004-of-00007.safetensors",
82
+ "model.layers.14.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
83
+ "model.layers.14.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
84
+ "model.layers.14.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
85
+ "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
86
+ "model.layers.14.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
87
+ "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
88
+ "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
89
+ "model.layers.14.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
90
+ "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
91
+ "model.layers.14.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
92
+ "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
93
+ "model.layers.15.input_layernorm.weight": "model-00004-of-00007.safetensors",
94
+ "model.layers.15.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
95
+ "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
96
+ "model.layers.15.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
97
+ "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
98
+ "model.layers.15.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
99
+ "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
100
+ "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
101
+ "model.layers.15.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
102
+ "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
103
+ "model.layers.15.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
104
+ "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
105
+ "model.layers.16.input_layernorm.weight": "model-00004-of-00007.safetensors",
106
+ "model.layers.16.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
107
+ "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
108
+ "model.layers.16.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
109
+ "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
110
+ "model.layers.16.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
111
+ "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
112
+ "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
113
+ "model.layers.16.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
114
+ "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
115
+ "model.layers.16.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
116
+ "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
117
+ "model.layers.17.input_layernorm.weight": "model-00004-of-00007.safetensors",
118
+ "model.layers.17.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
119
+ "model.layers.17.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
120
+ "model.layers.17.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
121
+ "model.layers.17.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
122
+ "model.layers.17.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
123
+ "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
124
+ "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
125
+ "model.layers.17.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
126
+ "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
127
+ "model.layers.17.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
128
+ "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
129
+ "model.layers.18.input_layernorm.weight": "model-00005-of-00007.safetensors",
130
+ "model.layers.18.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
131
+ "model.layers.18.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
132
+ "model.layers.18.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
133
+ "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
134
+ "model.layers.18.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
135
+ "model.layers.18.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
136
+ "model.layers.18.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
137
+ "model.layers.18.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
138
+ "model.layers.18.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
139
+ "model.layers.18.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
140
+ "model.layers.18.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
141
+ "model.layers.19.input_layernorm.weight": "model-00005-of-00007.safetensors",
142
+ "model.layers.19.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
143
+ "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
144
+ "model.layers.19.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
145
+ "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
146
+ "model.layers.19.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
147
+ "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
148
+ "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
149
+ "model.layers.19.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
150
+ "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
151
+ "model.layers.19.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
152
+ "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
153
+ "model.layers.2.input_layernorm.weight": "model-00001-of-00007.safetensors",
154
+ "model.layers.2.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
155
+ "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
156
+ "model.layers.2.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
157
+ "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
158
+ "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
159
+ "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
160
+ "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
161
+ "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
162
+ "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
163
+ "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
164
+ "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
165
+ "model.layers.20.input_layernorm.weight": "model-00005-of-00007.safetensors",
166
+ "model.layers.20.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
167
+ "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
168
+ "model.layers.20.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
169
+ "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
170
+ "model.layers.20.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
171
+ "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
172
+ "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
173
+ "model.layers.20.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
174
+ "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
175
+ "model.layers.20.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
176
+ "model.layers.20.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
177
+ "model.layers.21.input_layernorm.weight": "model-00005-of-00007.safetensors",
178
+ "model.layers.21.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
179
+ "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
180
+ "model.layers.21.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
181
+ "model.layers.21.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
182
+ "model.layers.21.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
183
+ "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
184
+ "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
185
+ "model.layers.21.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
186
+ "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
187
+ "model.layers.21.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
188
+ "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
189
+ "model.layers.22.input_layernorm.weight": "model-00005-of-00007.safetensors",
190
+ "model.layers.22.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
191
+ "model.layers.22.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
192
+ "model.layers.22.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
193
+ "model.layers.22.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
194
+ "model.layers.22.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
195
+ "model.layers.22.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
196
+ "model.layers.22.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
197
+ "model.layers.22.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
198
+ "model.layers.22.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
199
+ "model.layers.22.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
200
+ "model.layers.22.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
201
+ "model.layers.23.input_layernorm.weight": "model-00005-of-00007.safetensors",
202
+ "model.layers.23.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
203
+ "model.layers.23.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
204
+ "model.layers.23.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
205
+ "model.layers.23.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
206
+ "model.layers.23.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
207
+ "model.layers.23.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
208
+ "model.layers.23.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
209
+ "model.layers.23.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
210
+ "model.layers.23.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
211
+ "model.layers.23.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
212
+ "model.layers.23.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
213
+ "model.layers.24.input_layernorm.weight": "model-00006-of-00007.safetensors",
214
+ "model.layers.24.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
215
+ "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
216
+ "model.layers.24.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
217
+ "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
218
+ "model.layers.24.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
219
+ "model.layers.24.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
220
+ "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
221
+ "model.layers.24.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
222
+ "model.layers.24.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
223
+ "model.layers.24.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
224
+ "model.layers.24.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
225
+ "model.layers.25.input_layernorm.weight": "model-00006-of-00007.safetensors",
226
+ "model.layers.25.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
227
+ "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
228
+ "model.layers.25.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
229
+ "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
230
+ "model.layers.25.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
231
+ "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
232
+ "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
233
+ "model.layers.25.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
234
+ "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
235
+ "model.layers.25.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
236
+ "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
237
+ "model.layers.26.input_layernorm.weight": "model-00006-of-00007.safetensors",
238
+ "model.layers.26.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
239
+ "model.layers.26.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
240
+ "model.layers.26.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
241
+ "model.layers.26.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
242
+ "model.layers.26.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
243
+ "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
244
+ "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
245
+ "model.layers.26.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
246
+ "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
247
+ "model.layers.26.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
248
+ "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
249
+ "model.layers.27.input_layernorm.weight": "model-00006-of-00007.safetensors",
250
+ "model.layers.27.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
251
+ "model.layers.27.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
252
+ "model.layers.27.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
253
+ "model.layers.27.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
254
+ "model.layers.27.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
255
+ "model.layers.27.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
256
+ "model.layers.27.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
257
+ "model.layers.27.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
258
+ "model.layers.27.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
259
+ "model.layers.27.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
260
+ "model.layers.27.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
261
+ "model.layers.3.input_layernorm.weight": "model-00002-of-00007.safetensors",
262
+ "model.layers.3.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
263
+ "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
264
+ "model.layers.3.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
265
+ "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
266
+ "model.layers.3.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
267
+ "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
268
+ "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
269
+ "model.layers.3.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
270
+ "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
271
+ "model.layers.3.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
272
+ "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
273
+ "model.layers.4.input_layernorm.weight": "model-00002-of-00007.safetensors",
274
+ "model.layers.4.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
275
+ "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
276
+ "model.layers.4.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
277
+ "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
278
+ "model.layers.4.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
279
+ "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
280
+ "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
281
+ "model.layers.4.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
282
+ "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
283
+ "model.layers.4.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
284
+ "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
285
+ "model.layers.5.input_layernorm.weight": "model-00002-of-00007.safetensors",
286
+ "model.layers.5.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
287
+ "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
288
+ "model.layers.5.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
289
+ "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
290
+ "model.layers.5.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
291
+ "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
292
+ "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
293
+ "model.layers.5.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
294
+ "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
295
+ "model.layers.5.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
296
+ "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
297
+ "model.layers.6.input_layernorm.weight": "model-00002-of-00007.safetensors",
298
+ "model.layers.6.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
299
+ "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
300
+ "model.layers.6.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
301
+ "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
302
+ "model.layers.6.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
303
+ "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
304
+ "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
305
+ "model.layers.6.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
306
+ "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
307
+ "model.layers.6.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
308
+ "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
309
+ "model.layers.7.input_layernorm.weight": "model-00002-of-00007.safetensors",
310
+ "model.layers.7.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
311
+ "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
312
+ "model.layers.7.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
313
+ "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
314
+ "model.layers.7.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
315
+ "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
316
+ "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
317
+ "model.layers.7.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
318
+ "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
319
+ "model.layers.7.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
320
+ "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
321
+ "model.layers.8.input_layernorm.weight": "model-00003-of-00007.safetensors",
322
+ "model.layers.8.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
323
+ "model.layers.8.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
324
+ "model.layers.8.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
325
+ "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
326
+ "model.layers.8.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
327
+ "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
328
+ "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
329
+ "model.layers.8.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
330
+ "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
331
+ "model.layers.8.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
332
+ "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
333
+ "model.layers.9.input_layernorm.weight": "model-00003-of-00007.safetensors",
334
+ "model.layers.9.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
335
+ "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
336
+ "model.layers.9.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
337
+ "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
338
+ "model.layers.9.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
339
+ "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
340
+ "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
341
+ "model.layers.9.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
342
+ "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
343
+ "model.layers.9.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
344
+ "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
345
+ "model.norm.weight": "model-00006-of-00007.safetensors"
346
+ }
347
+ }
checkpoint-400/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0eb1f7b1acfd920c8cdef076eee48ccfec7d3ed4d8c5d83c2592fbbb4a4f9b38
3
+ size 15429
checkpoint-400/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
+ }
checkpoint-400/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
3
+ size 11421896
checkpoint-400/tokenizer_config.json ADDED
@@ -0,0 +1,209 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": false,
5
+ "added_tokens_decoder": {
6
+ "151643": {
7
+ "content": "<|endoftext|>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "151644": {
15
+ "content": "<|im_start|>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "151645": {
23
+ "content": "<|im_end|>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "151646": {
31
+ "content": "<|object_ref_start|>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "151647": {
39
+ "content": "<|object_ref_end|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "151648": {
47
+ "content": "<|box_start|>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "151649": {
55
+ "content": "<|box_end|>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "151650": {
63
+ "content": "<|quad_start|>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "151651": {
71
+ "content": "<|quad_end|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "151652": {
79
+ "content": "<|vision_start|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "151653": {
87
+ "content": "<|vision_end|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "151654": {
95
+ "content": "<|vision_pad|>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": true
101
+ },
102
+ "151655": {
103
+ "content": "<|image_pad|>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": true
109
+ },
110
+ "151656": {
111
+ "content": "<|video_pad|>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": true
117
+ },
118
+ "151657": {
119
+ "content": "<tool_call>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": false
125
+ },
126
+ "151658": {
127
+ "content": "</tool_call>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": false
133
+ },
134
+ "151659": {
135
+ "content": "<|fim_prefix|>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": false
141
+ },
142
+ "151660": {
143
+ "content": "<|fim_middle|>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": false
149
+ },
150
+ "151661": {
151
+ "content": "<|fim_suffix|>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": false
157
+ },
158
+ "151662": {
159
+ "content": "<|fim_pad|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": false
165
+ },
166
+ "151663": {
167
+ "content": "<|repo_name|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": false
173
+ },
174
+ "151664": {
175
+ "content": "<|file_sep|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": false
181
+ }
182
+ },
183
+ "additional_special_tokens": [
184
+ "<|im_start|>",
185
+ "<|im_end|>",
186
+ "<|object_ref_start|>",
187
+ "<|object_ref_end|>",
188
+ "<|box_start|>",
189
+ "<|box_end|>",
190
+ "<|quad_start|>",
191
+ "<|quad_end|>",
192
+ "<|vision_start|>",
193
+ "<|vision_end|>",
194
+ "<|vision_pad|>",
195
+ "<|image_pad|>",
196
+ "<|video_pad|>"
197
+ ],
198
+ "bos_token": null,
199
+ "clean_up_tokenization_spaces": false,
200
+ "eos_token": "<|im_end|>",
201
+ "errors": "replace",
202
+ "extra_special_tokens": {},
203
+ "model_max_length": 131072,
204
+ "pad_token": "<|endoftext|>",
205
+ "padding_side": "right",
206
+ "split_special_tokens": false,
207
+ "tokenizer_class": "Qwen2Tokenizer",
208
+ "unk_token": null
209
+ }
checkpoint-400/trainer_state.json ADDED
@@ -0,0 +1,2834 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0695884911341587,
6
+ "eval_steps": 500,
7
+ "global_step": 400,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0026764804282368685,
14
+ "grad_norm": 416.7599792480469,
15
+ "learning_rate": 0.0,
16
+ "loss": 3.2698,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.005352960856473737,
21
+ "grad_norm": 316.08477783203125,
22
+ "learning_rate": 2.0000000000000002e-07,
23
+ "loss": 2.9631,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.008029441284710606,
28
+ "grad_norm": 472.76885986328125,
29
+ "learning_rate": 4.0000000000000003e-07,
30
+ "loss": 3.9863,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.010705921712947474,
35
+ "grad_norm": 251.41015625,
36
+ "learning_rate": 6.000000000000001e-07,
37
+ "loss": 3.2654,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.013382402141184342,
42
+ "grad_norm": 217.40838623046875,
43
+ "learning_rate": 8.000000000000001e-07,
44
+ "loss": 3.4505,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.01605888256942121,
49
+ "grad_norm": 136.2932586669922,
50
+ "learning_rate": 1.0000000000000002e-06,
51
+ "loss": 3.2498,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.01873536299765808,
56
+ "grad_norm": 135.7803192138672,
57
+ "learning_rate": 1.2000000000000002e-06,
58
+ "loss": 3.0983,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.021411843425894948,
63
+ "grad_norm": 93.63102722167969,
64
+ "learning_rate": 1.4000000000000001e-06,
65
+ "loss": 2.965,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.024088323854131816,
70
+ "grad_norm": 82.7247085571289,
71
+ "learning_rate": 1.6000000000000001e-06,
72
+ "loss": 2.6341,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.026764804282368684,
77
+ "grad_norm": 94.11923217773438,
78
+ "learning_rate": 1.8000000000000001e-06,
79
+ "loss": 3.2072,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.029441284710605555,
84
+ "grad_norm": 65.76191711425781,
85
+ "learning_rate": 2.0000000000000003e-06,
86
+ "loss": 2.6866,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.03211776513884242,
91
+ "grad_norm": 89.41738891601562,
92
+ "learning_rate": 2.2e-06,
93
+ "loss": 2.8421,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.03479424556707929,
98
+ "grad_norm": 82.1783676147461,
99
+ "learning_rate": 2.4000000000000003e-06,
100
+ "loss": 2.5489,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.03747072599531616,
105
+ "grad_norm": 72.41712951660156,
106
+ "learning_rate": 2.6e-06,
107
+ "loss": 2.9365,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.04014720642355303,
112
+ "grad_norm": 77.541015625,
113
+ "learning_rate": 2.8000000000000003e-06,
114
+ "loss": 2.0516,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.042823686851789895,
119
+ "grad_norm": 125.65862274169922,
120
+ "learning_rate": 3e-06,
121
+ "loss": 2.7676,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.04550016728002677,
126
+ "grad_norm": 94.85375213623047,
127
+ "learning_rate": 3.2000000000000003e-06,
128
+ "loss": 2.5114,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.04817664770826363,
133
+ "grad_norm": 77.89604187011719,
134
+ "learning_rate": 3.4000000000000005e-06,
135
+ "loss": 2.536,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.0508531281365005,
140
+ "grad_norm": 59.268680572509766,
141
+ "learning_rate": 3.6000000000000003e-06,
142
+ "loss": 2.1877,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.05352960856473737,
147
+ "grad_norm": 57.32849884033203,
148
+ "learning_rate": 3.8000000000000005e-06,
149
+ "loss": 2.2084,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.05620608899297424,
154
+ "grad_norm": 58.27803421020508,
155
+ "learning_rate": 4.000000000000001e-06,
156
+ "loss": 2.191,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.05888256942121111,
161
+ "grad_norm": 32.10304260253906,
162
+ "learning_rate": 4.2000000000000004e-06,
163
+ "loss": 2.0009,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.061559049849447975,
168
+ "grad_norm": 17.106046676635742,
169
+ "learning_rate": 4.4e-06,
170
+ "loss": 1.7496,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.06423553027768485,
175
+ "grad_norm": 16.371326446533203,
176
+ "learning_rate": 4.600000000000001e-06,
177
+ "loss": 2.0558,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.06691201070592172,
182
+ "grad_norm": 14.423291206359863,
183
+ "learning_rate": 4.800000000000001e-06,
184
+ "loss": 2.1511,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.06958849113415858,
189
+ "grad_norm": 16.198841094970703,
190
+ "learning_rate": 5e-06,
191
+ "loss": 1.9571,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.07226497156239545,
196
+ "grad_norm": 14.810019493103027,
197
+ "learning_rate": 5.2e-06,
198
+ "loss": 1.9937,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.07494145199063232,
203
+ "grad_norm": 13.464497566223145,
204
+ "learning_rate": 5.400000000000001e-06,
205
+ "loss": 1.7204,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.07761793241886919,
210
+ "grad_norm": 14.364913940429688,
211
+ "learning_rate": 5.600000000000001e-06,
212
+ "loss": 1.7,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.08029441284710606,
217
+ "grad_norm": 12.647878646850586,
218
+ "learning_rate": 5.8e-06,
219
+ "loss": 1.7034,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.08297089327534292,
224
+ "grad_norm": 12.305310249328613,
225
+ "learning_rate": 6e-06,
226
+ "loss": 1.7873,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.08564737370357979,
231
+ "grad_norm": 9.316176414489746,
232
+ "learning_rate": 6.200000000000001e-06,
233
+ "loss": 1.7204,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.08832385413181666,
238
+ "grad_norm": 10.51429271697998,
239
+ "learning_rate": 6.4000000000000006e-06,
240
+ "loss": 1.401,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.09100033456005353,
245
+ "grad_norm": 10.570704460144043,
246
+ "learning_rate": 6.600000000000001e-06,
247
+ "loss": 1.6091,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.0936768149882904,
252
+ "grad_norm": 12.805994987487793,
253
+ "learning_rate": 6.800000000000001e-06,
254
+ "loss": 1.5669,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.09635329541652726,
259
+ "grad_norm": 10.955977439880371,
260
+ "learning_rate": 7e-06,
261
+ "loss": 1.4644,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.09902977584476413,
266
+ "grad_norm": 15.71002197265625,
267
+ "learning_rate": 7.2000000000000005e-06,
268
+ "loss": 1.5497,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.101706256273001,
273
+ "grad_norm": 11.375880241394043,
274
+ "learning_rate": 7.4e-06,
275
+ "loss": 1.5646,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.10438273670123788,
280
+ "grad_norm": 12.987727165222168,
281
+ "learning_rate": 7.600000000000001e-06,
282
+ "loss": 1.4149,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.10705921712947473,
287
+ "grad_norm": 10.491798400878906,
288
+ "learning_rate": 7.800000000000002e-06,
289
+ "loss": 1.5017,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.1097356975577116,
294
+ "grad_norm": 12.43316650390625,
295
+ "learning_rate": 8.000000000000001e-06,
296
+ "loss": 1.4759,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.11241217798594848,
301
+ "grad_norm": 12.077725410461426,
302
+ "learning_rate": 8.2e-06,
303
+ "loss": 1.4892,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.11508865841418535,
308
+ "grad_norm": 12.642361640930176,
309
+ "learning_rate": 8.400000000000001e-06,
310
+ "loss": 1.458,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.11776513884242222,
315
+ "grad_norm": 15.535161018371582,
316
+ "learning_rate": 8.6e-06,
317
+ "loss": 1.6861,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.12044161927065908,
322
+ "grad_norm": 11.678996086120605,
323
+ "learning_rate": 8.8e-06,
324
+ "loss": 1.6968,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.12311809969889595,
329
+ "grad_norm": 11.906571388244629,
330
+ "learning_rate": 9e-06,
331
+ "loss": 1.44,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.12579458012713282,
336
+ "grad_norm": 11.492127418518066,
337
+ "learning_rate": 9.200000000000002e-06,
338
+ "loss": 1.3714,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.1284710605553697,
343
+ "grad_norm": 15.786943435668945,
344
+ "learning_rate": 9.4e-06,
345
+ "loss": 1.4132,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.13114754098360656,
350
+ "grad_norm": 12.31556224822998,
351
+ "learning_rate": 9.600000000000001e-06,
352
+ "loss": 1.166,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.13382402141184344,
357
+ "grad_norm": 12.496630668640137,
358
+ "learning_rate": 9.800000000000001e-06,
359
+ "loss": 1.2859,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.1365005018400803,
364
+ "grad_norm": 11.003588676452637,
365
+ "learning_rate": 1e-05,
366
+ "loss": 1.2614,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.13917698226831715,
371
+ "grad_norm": 12.946131706237793,
372
+ "learning_rate": 9.99994935597953e-06,
373
+ "loss": 1.4555,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.14185346269655402,
378
+ "grad_norm": 13.676822662353516,
379
+ "learning_rate": 9.999797424944041e-06,
380
+ "loss": 1.2167,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.1445299431247909,
385
+ "grad_norm": 14.827473640441895,
386
+ "learning_rate": 9.999544209971299e-06,
387
+ "loss": 1.3456,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.14720642355302777,
392
+ "grad_norm": 11.921989440917969,
393
+ "learning_rate": 9.99918971619083e-06,
394
+ "loss": 1.2535,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.14988290398126464,
399
+ "grad_norm": 12.726682662963867,
400
+ "learning_rate": 9.99873395078383e-06,
401
+ "loss": 1.2172,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.1525593844095015,
406
+ "grad_norm": 18.52472496032715,
407
+ "learning_rate": 9.998176922983017e-06,
408
+ "loss": 1.4418,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.15523586483773838,
413
+ "grad_norm": 13.32449722290039,
414
+ "learning_rate": 9.99751864407244e-06,
415
+ "loss": 1.18,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.15791234526597525,
420
+ "grad_norm": 10.778595924377441,
421
+ "learning_rate": 9.996759127387259e-06,
422
+ "loss": 1.1879,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.16058882569421212,
427
+ "grad_norm": 10.155744552612305,
428
+ "learning_rate": 9.99589838831346e-06,
429
+ "loss": 1.1274,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.16326530612244897,
434
+ "grad_norm": 13.751497268676758,
435
+ "learning_rate": 9.994936444287565e-06,
436
+ "loss": 1.1499,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.16594178655068584,
441
+ "grad_norm": 17.752635955810547,
442
+ "learning_rate": 9.993873314796253e-06,
443
+ "loss": 1.1425,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.1686182669789227,
448
+ "grad_norm": 13.581279754638672,
449
+ "learning_rate": 9.992709021375987e-06,
450
+ "loss": 1.0341,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.17129474740715958,
455
+ "grad_norm": 11.867856979370117,
456
+ "learning_rate": 9.991443587612568e-06,
457
+ "loss": 1.024,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.17397122783539645,
462
+ "grad_norm": 13.20876693725586,
463
+ "learning_rate": 9.990077039140655e-06,
464
+ "loss": 1.0375,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.17664770826363332,
469
+ "grad_norm": 12.211270332336426,
470
+ "learning_rate": 9.988609403643254e-06,
471
+ "loss": 1.1334,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.1793241886918702,
476
+ "grad_norm": 11.89450454711914,
477
+ "learning_rate": 9.987040710851148e-06,
478
+ "loss": 1.0447,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.18200066912010707,
483
+ "grad_norm": 10.132635116577148,
484
+ "learning_rate": 9.9853709925423e-06,
485
+ "loss": 0.7626,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.18467714954834394,
490
+ "grad_norm": 9.661967277526855,
491
+ "learning_rate": 9.983600282541213e-06,
492
+ "loss": 0.968,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.1873536299765808,
497
+ "grad_norm": 11.19580078125,
498
+ "learning_rate": 9.981728616718234e-06,
499
+ "loss": 0.9161,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.19003011040481765,
504
+ "grad_norm": 9.165005683898926,
505
+ "learning_rate": 9.979756032988837e-06,
506
+ "loss": 0.8821,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.19270659083305453,
511
+ "grad_norm": 9.17795181274414,
512
+ "learning_rate": 9.977682571312847e-06,
513
+ "loss": 1.0452,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.1953830712612914,
518
+ "grad_norm": 7.483065128326416,
519
+ "learning_rate": 9.975508273693643e-06,
520
+ "loss": 1.0135,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.19805955168952827,
525
+ "grad_norm": 7.2489728927612305,
526
+ "learning_rate": 9.97323318417729e-06,
527
+ "loss": 0.9675,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.20073603211776514,
532
+ "grad_norm": 7.151744365692139,
533
+ "learning_rate": 9.970857348851667e-06,
534
+ "loss": 0.829,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.203412512546002,
539
+ "grad_norm": 7.535038948059082,
540
+ "learning_rate": 9.968380815845504e-06,
541
+ "loss": 0.9383,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.20608899297423888,
546
+ "grad_norm": 5.609306812286377,
547
+ "learning_rate": 9.965803635327445e-06,
548
+ "loss": 0.8914,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.20876547340247575,
553
+ "grad_norm": 7.203204154968262,
554
+ "learning_rate": 9.963125859505e-06,
555
+ "loss": 1.0889,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.21144195383071263,
560
+ "grad_norm": 4.907174587249756,
561
+ "learning_rate": 9.960347542623506e-06,
562
+ "loss": 0.7694,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.21411843425894947,
567
+ "grad_norm": 5.539646148681641,
568
+ "learning_rate": 9.95746874096501e-06,
569
+ "loss": 0.9965,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.21679491468718634,
574
+ "grad_norm": 5.0046868324279785,
575
+ "learning_rate": 9.954489512847156e-06,
576
+ "loss": 0.7795,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.2194713951154232,
581
+ "grad_norm": 5.138219833374023,
582
+ "learning_rate": 9.951409918621977e-06,
583
+ "loss": 0.8877,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.22214787554366008,
588
+ "grad_norm": 4.388546466827393,
589
+ "learning_rate": 9.948230020674685e-06,
590
+ "loss": 0.9253,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.22482435597189696,
595
+ "grad_norm": 4.456934928894043,
596
+ "learning_rate": 9.944949883422409e-06,
597
+ "loss": 0.7335,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.22750083640013383,
602
+ "grad_norm": 4.645761013031006,
603
+ "learning_rate": 9.941569573312882e-06,
604
+ "loss": 0.8997,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.2301773168283707,
609
+ "grad_norm": 4.329288482666016,
610
+ "learning_rate": 9.938089158823101e-06,
611
+ "loss": 0.852,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.23285379725660757,
616
+ "grad_norm": 5.044316291809082,
617
+ "learning_rate": 9.934508710457944e-06,
618
+ "loss": 0.8236,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.23553027768484444,
623
+ "grad_norm": 5.356631755828857,
624
+ "learning_rate": 9.930828300748726e-06,
625
+ "loss": 0.9572,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.23820675811308128,
630
+ "grad_norm": 3.933661699295044,
631
+ "learning_rate": 9.927048004251748e-06,
632
+ "loss": 0.8479,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.24088323854131816,
637
+ "grad_norm": 5.228504180908203,
638
+ "learning_rate": 9.923167897546773e-06,
639
+ "loss": 0.741,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.24355971896955503,
644
+ "grad_norm": 4.520639419555664,
645
+ "learning_rate": 9.919188059235483e-06,
646
+ "loss": 0.746,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.2462361993977919,
651
+ "grad_norm": 4.191668510437012,
652
+ "learning_rate": 9.915108569939884e-06,
653
+ "loss": 0.6904,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.24891267982602877,
658
+ "grad_norm": 4.300325393676758,
659
+ "learning_rate": 9.910929512300673e-06,
660
+ "loss": 0.8753,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.25158916025426564,
665
+ "grad_norm": 5.979667663574219,
666
+ "learning_rate": 9.90665097097556e-06,
667
+ "loss": 0.7355,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.2542656406825025,
672
+ "grad_norm": 5.374980449676514,
673
+ "learning_rate": 9.902273032637558e-06,
674
+ "loss": 0.7932,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.2569421211107394,
679
+ "grad_norm": 3.9075119495391846,
680
+ "learning_rate": 9.897795785973227e-06,
681
+ "loss": 0.9069,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.25961860153897626,
686
+ "grad_norm": 4.439886569976807,
687
+ "learning_rate": 9.89321932168088e-06,
688
+ "loss": 0.8408,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.26229508196721313,
693
+ "grad_norm": 4.403755187988281,
694
+ "learning_rate": 9.888543732468732e-06,
695
+ "loss": 0.709,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.26497156239545,
700
+ "grad_norm": 5.023388862609863,
701
+ "learning_rate": 9.883769113053039e-06,
702
+ "loss": 0.8928,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.26764804282368687,
707
+ "grad_norm": 5.586976528167725,
708
+ "learning_rate": 9.878895560156172e-06,
709
+ "loss": 0.9644,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 0.27032452325192374,
714
+ "grad_norm": 15.905163764953613,
715
+ "learning_rate": 9.873923172504653e-06,
716
+ "loss": 0.8362,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 0.2730010036801606,
721
+ "grad_norm": 5.6663432121276855,
722
+ "learning_rate": 9.868852050827167e-06,
723
+ "loss": 0.7585,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 0.27567748410839743,
728
+ "grad_norm": 4.5769124031066895,
729
+ "learning_rate": 9.863682297852506e-06,
730
+ "loss": 0.9748,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 0.2783539645366343,
735
+ "grad_norm": 4.305916786193848,
736
+ "learning_rate": 9.858414018307503e-06,
737
+ "loss": 0.805,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 0.2810304449648712,
742
+ "grad_norm": 4.756903648376465,
743
+ "learning_rate": 9.853047318914898e-06,
744
+ "loss": 0.8221,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 0.28370692539310804,
749
+ "grad_norm": 5.709683418273926,
750
+ "learning_rate": 9.847582308391189e-06,
751
+ "loss": 0.7416,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 0.2863834058213449,
756
+ "grad_norm": 3.933176279067993,
757
+ "learning_rate": 9.842019097444414e-06,
758
+ "loss": 0.7298,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 0.2890598862495818,
763
+ "grad_norm": 4.9454264640808105,
764
+ "learning_rate": 9.836357798771922e-06,
765
+ "loss": 0.7683,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 0.29173636667781866,
770
+ "grad_norm": 4.274379730224609,
771
+ "learning_rate": 9.830598527058083e-06,
772
+ "loss": 0.7788,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 0.29441284710605553,
777
+ "grad_norm": 3.6148972511291504,
778
+ "learning_rate": 9.824741398971966e-06,
779
+ "loss": 0.6941,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 0.2970893275342924,
784
+ "grad_norm": 3.8976197242736816,
785
+ "learning_rate": 9.81878653316498e-06,
786
+ "loss": 0.8783,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 0.2997658079625293,
791
+ "grad_norm": 4.712560176849365,
792
+ "learning_rate": 9.81273405026846e-06,
793
+ "loss": 0.9022,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 0.30244228839076615,
798
+ "grad_norm": 5.358889579772949,
799
+ "learning_rate": 9.806584072891234e-06,
800
+ "loss": 0.8587,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 0.305118768819003,
805
+ "grad_norm": 4.296327114105225,
806
+ "learning_rate": 9.800336725617136e-06,
807
+ "loss": 0.8813,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 0.3077952492472399,
812
+ "grad_norm": 4.895114421844482,
813
+ "learning_rate": 9.793992135002476e-06,
814
+ "loss": 0.8184,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 0.31047172967547676,
819
+ "grad_norm": 4.782822132110596,
820
+ "learning_rate": 9.787550429573487e-06,
821
+ "loss": 0.6049,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 0.31314821010371363,
826
+ "grad_norm": 5.41484260559082,
827
+ "learning_rate": 9.781011739823715e-06,
828
+ "loss": 0.7804,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 0.3158246905319505,
833
+ "grad_norm": 3.639780044555664,
834
+ "learning_rate": 9.77437619821137e-06,
835
+ "loss": 0.7241,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 0.3185011709601874,
840
+ "grad_norm": 4.5406880378723145,
841
+ "learning_rate": 9.767643939156658e-06,
842
+ "loss": 0.8398,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 0.32117765138842425,
847
+ "grad_norm": 4.575490951538086,
848
+ "learning_rate": 9.760815099039045e-06,
849
+ "loss": 0.6951,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 0.3238541318166611,
854
+ "grad_norm": 3.868921995162964,
855
+ "learning_rate": 9.753889816194498e-06,
856
+ "loss": 0.706,
857
+ "step": 121
858
+ },
859
+ {
860
+ "epoch": 0.32653061224489793,
861
+ "grad_norm": 4.064291477203369,
862
+ "learning_rate": 9.746868230912683e-06,
863
+ "loss": 0.7386,
864
+ "step": 122
865
+ },
866
+ {
867
+ "epoch": 0.3292070926731348,
868
+ "grad_norm": 4.059055805206299,
869
+ "learning_rate": 9.739750485434126e-06,
870
+ "loss": 0.7171,
871
+ "step": 123
872
+ },
873
+ {
874
+ "epoch": 0.3318835731013717,
875
+ "grad_norm": 4.522129535675049,
876
+ "learning_rate": 9.73253672394732e-06,
877
+ "loss": 0.6679,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 0.33456005352960855,
882
+ "grad_norm": 3.669271945953369,
883
+ "learning_rate": 9.725227092585824e-06,
884
+ "loss": 0.7409,
885
+ "step": 125
886
+ },
887
+ {
888
+ "epoch": 0.3372365339578454,
889
+ "grad_norm": 3.603746175765991,
890
+ "learning_rate": 9.717821739425286e-06,
891
+ "loss": 0.7286,
892
+ "step": 126
893
+ },
894
+ {
895
+ "epoch": 0.3399130143860823,
896
+ "grad_norm": 4.118206977844238,
897
+ "learning_rate": 9.710320814480448e-06,
898
+ "loss": 0.7164,
899
+ "step": 127
900
+ },
901
+ {
902
+ "epoch": 0.34258949481431916,
903
+ "grad_norm": 3.6243247985839844,
904
+ "learning_rate": 9.702724469702107e-06,
905
+ "loss": 0.7772,
906
+ "step": 128
907
+ },
908
+ {
909
+ "epoch": 0.34526597524255603,
910
+ "grad_norm": 3.8515329360961914,
911
+ "learning_rate": 9.695032858974042e-06,
912
+ "loss": 0.7173,
913
+ "step": 129
914
+ },
915
+ {
916
+ "epoch": 0.3479424556707929,
917
+ "grad_norm": 4.385611534118652,
918
+ "learning_rate": 9.687246138109888e-06,
919
+ "loss": 0.8672,
920
+ "step": 130
921
+ },
922
+ {
923
+ "epoch": 0.3506189360990298,
924
+ "grad_norm": 3.6981892585754395,
925
+ "learning_rate": 9.679364464849983e-06,
926
+ "loss": 0.6571,
927
+ "step": 131
928
+ },
929
+ {
930
+ "epoch": 0.35329541652726665,
931
+ "grad_norm": 4.061744213104248,
932
+ "learning_rate": 9.671387998858178e-06,
933
+ "loss": 0.7114,
934
+ "step": 132
935
+ },
936
+ {
937
+ "epoch": 0.3559718969555035,
938
+ "grad_norm": 4.793642997741699,
939
+ "learning_rate": 9.663316901718599e-06,
940
+ "loss": 0.7537,
941
+ "step": 133
942
+ },
943
+ {
944
+ "epoch": 0.3586483773837404,
945
+ "grad_norm": 3.751180648803711,
946
+ "learning_rate": 9.655151336932362e-06,
947
+ "loss": 0.6826,
948
+ "step": 134
949
+ },
950
+ {
951
+ "epoch": 0.36132485781197726,
952
+ "grad_norm": 3.8843295574188232,
953
+ "learning_rate": 9.646891469914285e-06,
954
+ "loss": 0.6868,
955
+ "step": 135
956
+ },
957
+ {
958
+ "epoch": 0.36400133824021413,
959
+ "grad_norm": 4.41197395324707,
960
+ "learning_rate": 9.638537467989517e-06,
961
+ "loss": 0.7224,
962
+ "step": 136
963
+ },
964
+ {
965
+ "epoch": 0.366677818668451,
966
+ "grad_norm": 3.2272307872772217,
967
+ "learning_rate": 9.630089500390154e-06,
968
+ "loss": 0.5735,
969
+ "step": 137
970
+ },
971
+ {
972
+ "epoch": 0.3693542990966879,
973
+ "grad_norm": 3.5232367515563965,
974
+ "learning_rate": 9.621547738251816e-06,
975
+ "loss": 0.7621,
976
+ "step": 138
977
+ },
978
+ {
979
+ "epoch": 0.37203077952492475,
980
+ "grad_norm": 3.4856607913970947,
981
+ "learning_rate": 9.61291235461017e-06,
982
+ "loss": 0.6517,
983
+ "step": 139
984
+ },
985
+ {
986
+ "epoch": 0.3747072599531616,
987
+ "grad_norm": 3.958531379699707,
988
+ "learning_rate": 9.604183524397439e-06,
989
+ "loss": 0.8994,
990
+ "step": 140
991
+ },
992
+ {
993
+ "epoch": 0.37738374038139844,
994
+ "grad_norm": 3.3670690059661865,
995
+ "learning_rate": 9.595361424438841e-06,
996
+ "loss": 0.5975,
997
+ "step": 141
998
+ },
999
+ {
1000
+ "epoch": 0.3800602208096353,
1001
+ "grad_norm": 4.388239860534668,
1002
+ "learning_rate": 9.586446233449024e-06,
1003
+ "loss": 0.7168,
1004
+ "step": 142
1005
+ },
1006
+ {
1007
+ "epoch": 0.3827367012378722,
1008
+ "grad_norm": 3.513667345046997,
1009
+ "learning_rate": 9.577438132028431e-06,
1010
+ "loss": 0.7645,
1011
+ "step": 143
1012
+ },
1013
+ {
1014
+ "epoch": 0.38541318166610905,
1015
+ "grad_norm": 3.383009195327759,
1016
+ "learning_rate": 9.568337302659652e-06,
1017
+ "loss": 0.7159,
1018
+ "step": 144
1019
+ },
1020
+ {
1021
+ "epoch": 0.3880896620943459,
1022
+ "grad_norm": 3.7690727710723877,
1023
+ "learning_rate": 9.559143929703724e-06,
1024
+ "loss": 0.8527,
1025
+ "step": 145
1026
+ },
1027
+ {
1028
+ "epoch": 0.3907661425225828,
1029
+ "grad_norm": 3.1912662982940674,
1030
+ "learning_rate": 9.549858199396394e-06,
1031
+ "loss": 0.6288,
1032
+ "step": 146
1033
+ },
1034
+ {
1035
+ "epoch": 0.39344262295081966,
1036
+ "grad_norm": 3.616091012954712,
1037
+ "learning_rate": 9.540480299844345e-06,
1038
+ "loss": 0.6928,
1039
+ "step": 147
1040
+ },
1041
+ {
1042
+ "epoch": 0.39611910337905654,
1043
+ "grad_norm": 3.7126312255859375,
1044
+ "learning_rate": 9.531010421021396e-06,
1045
+ "loss": 0.7583,
1046
+ "step": 148
1047
+ },
1048
+ {
1049
+ "epoch": 0.3987955838072934,
1050
+ "grad_norm": 3.61950421333313,
1051
+ "learning_rate": 9.52144875476464e-06,
1052
+ "loss": 0.6796,
1053
+ "step": 149
1054
+ },
1055
+ {
1056
+ "epoch": 0.4014720642355303,
1057
+ "grad_norm": 3.0952789783477783,
1058
+ "learning_rate": 9.511795494770563e-06,
1059
+ "loss": 0.6761,
1060
+ "step": 150
1061
+ },
1062
+ {
1063
+ "epoch": 0.40414854466376715,
1064
+ "grad_norm": 3.4935519695281982,
1065
+ "learning_rate": 9.50205083659113e-06,
1066
+ "loss": 0.622,
1067
+ "step": 151
1068
+ },
1069
+ {
1070
+ "epoch": 0.406825025092004,
1071
+ "grad_norm": 4.367190837860107,
1072
+ "learning_rate": 9.492214977629804e-06,
1073
+ "loss": 0.8146,
1074
+ "step": 152
1075
+ },
1076
+ {
1077
+ "epoch": 0.4095015055202409,
1078
+ "grad_norm": 3.501554489135742,
1079
+ "learning_rate": 9.482288117137561e-06,
1080
+ "loss": 0.6101,
1081
+ "step": 153
1082
+ },
1083
+ {
1084
+ "epoch": 0.41217798594847777,
1085
+ "grad_norm": 3.7483065128326416,
1086
+ "learning_rate": 9.472270456208856e-06,
1087
+ "loss": 0.6095,
1088
+ "step": 154
1089
+ },
1090
+ {
1091
+ "epoch": 0.41485446637671464,
1092
+ "grad_norm": 2.7559807300567627,
1093
+ "learning_rate": 9.462162197777533e-06,
1094
+ "loss": 0.6167,
1095
+ "step": 155
1096
+ },
1097
+ {
1098
+ "epoch": 0.4175309468049515,
1099
+ "grad_norm": 3.520737409591675,
1100
+ "learning_rate": 9.451963546612737e-06,
1101
+ "loss": 0.5388,
1102
+ "step": 156
1103
+ },
1104
+ {
1105
+ "epoch": 0.4202074272331884,
1106
+ "grad_norm": 4.042897701263428,
1107
+ "learning_rate": 9.441674709314743e-06,
1108
+ "loss": 0.7228,
1109
+ "step": 157
1110
+ },
1111
+ {
1112
+ "epoch": 0.42288390766142525,
1113
+ "grad_norm": 4.090766429901123,
1114
+ "learning_rate": 9.431295894310786e-06,
1115
+ "loss": 0.5569,
1116
+ "step": 158
1117
+ },
1118
+ {
1119
+ "epoch": 0.42556038808966207,
1120
+ "grad_norm": 4.141130447387695,
1121
+ "learning_rate": 9.420827311850836e-06,
1122
+ "loss": 0.6297,
1123
+ "step": 159
1124
+ },
1125
+ {
1126
+ "epoch": 0.42823686851789894,
1127
+ "grad_norm": 3.472973346710205,
1128
+ "learning_rate": 9.410269174003333e-06,
1129
+ "loss": 0.6868,
1130
+ "step": 160
1131
+ },
1132
+ {
1133
+ "epoch": 0.4309133489461358,
1134
+ "grad_norm": 3.8000664710998535,
1135
+ "learning_rate": 9.399621694650898e-06,
1136
+ "loss": 0.6547,
1137
+ "step": 161
1138
+ },
1139
+ {
1140
+ "epoch": 0.4335898293743727,
1141
+ "grad_norm": 4.582136154174805,
1142
+ "learning_rate": 9.388885089485995e-06,
1143
+ "loss": 0.7552,
1144
+ "step": 162
1145
+ },
1146
+ {
1147
+ "epoch": 0.43626630980260955,
1148
+ "grad_norm": 3.4165501594543457,
1149
+ "learning_rate": 9.378059576006567e-06,
1150
+ "loss": 0.622,
1151
+ "step": 163
1152
+ },
1153
+ {
1154
+ "epoch": 0.4389427902308464,
1155
+ "grad_norm": 3.642669916152954,
1156
+ "learning_rate": 9.36714537351162e-06,
1157
+ "loss": 0.6302,
1158
+ "step": 164
1159
+ },
1160
+ {
1161
+ "epoch": 0.4416192706590833,
1162
+ "grad_norm": 3.53709077835083,
1163
+ "learning_rate": 9.356142703096793e-06,
1164
+ "loss": 0.5357,
1165
+ "step": 165
1166
+ },
1167
+ {
1168
+ "epoch": 0.44429575108732017,
1169
+ "grad_norm": 3.6043941974639893,
1170
+ "learning_rate": 9.345051787649877e-06,
1171
+ "loss": 0.6978,
1172
+ "step": 166
1173
+ },
1174
+ {
1175
+ "epoch": 0.44697223151555704,
1176
+ "grad_norm": 4.232686519622803,
1177
+ "learning_rate": 9.333872851846285e-06,
1178
+ "loss": 0.6698,
1179
+ "step": 167
1180
+ },
1181
+ {
1182
+ "epoch": 0.4496487119437939,
1183
+ "grad_norm": 3.1254665851593018,
1184
+ "learning_rate": 9.322606122144524e-06,
1185
+ "loss": 0.5173,
1186
+ "step": 168
1187
+ },
1188
+ {
1189
+ "epoch": 0.4523251923720308,
1190
+ "grad_norm": 4.0060930252075195,
1191
+ "learning_rate": 9.311251826781587e-06,
1192
+ "loss": 0.6674,
1193
+ "step": 169
1194
+ },
1195
+ {
1196
+ "epoch": 0.45500167280026765,
1197
+ "grad_norm": 4.720844268798828,
1198
+ "learning_rate": 9.299810195768341e-06,
1199
+ "loss": 0.8349,
1200
+ "step": 170
1201
+ },
1202
+ {
1203
+ "epoch": 0.4576781532285045,
1204
+ "grad_norm": 4.017560005187988,
1205
+ "learning_rate": 9.288281460884864e-06,
1206
+ "loss": 0.8489,
1207
+ "step": 171
1208
+ },
1209
+ {
1210
+ "epoch": 0.4603546336567414,
1211
+ "grad_norm": 4.482756614685059,
1212
+ "learning_rate": 9.276665855675751e-06,
1213
+ "loss": 0.6877,
1214
+ "step": 172
1215
+ },
1216
+ {
1217
+ "epoch": 0.46303111408497827,
1218
+ "grad_norm": 3.9189202785491943,
1219
+ "learning_rate": 9.264963615445378e-06,
1220
+ "loss": 0.58,
1221
+ "step": 173
1222
+ },
1223
+ {
1224
+ "epoch": 0.46570759451321514,
1225
+ "grad_norm": 3.285618543624878,
1226
+ "learning_rate": 9.25317497725315e-06,
1227
+ "loss": 0.585,
1228
+ "step": 174
1229
+ },
1230
+ {
1231
+ "epoch": 0.468384074941452,
1232
+ "grad_norm": 3.4484951496124268,
1233
+ "learning_rate": 9.241300179908672e-06,
1234
+ "loss": 0.5898,
1235
+ "step": 175
1236
+ },
1237
+ {
1238
+ "epoch": 0.4710605553696889,
1239
+ "grad_norm": 3.892336845397949,
1240
+ "learning_rate": 9.229339463966942e-06,
1241
+ "loss": 0.7691,
1242
+ "step": 176
1243
+ },
1244
+ {
1245
+ "epoch": 0.47373703579792575,
1246
+ "grad_norm": 3.3191475868225098,
1247
+ "learning_rate": 9.217293071723455e-06,
1248
+ "loss": 0.6468,
1249
+ "step": 177
1250
+ },
1251
+ {
1252
+ "epoch": 0.47641351622616257,
1253
+ "grad_norm": 3.2376253604888916,
1254
+ "learning_rate": 9.205161247209303e-06,
1255
+ "loss": 0.6233,
1256
+ "step": 178
1257
+ },
1258
+ {
1259
+ "epoch": 0.47908999665439944,
1260
+ "grad_norm": 3.029269218444824,
1261
+ "learning_rate": 9.192944236186237e-06,
1262
+ "loss": 0.5845,
1263
+ "step": 179
1264
+ },
1265
+ {
1266
+ "epoch": 0.4817664770826363,
1267
+ "grad_norm": 3.5170111656188965,
1268
+ "learning_rate": 9.180642286141678e-06,
1269
+ "loss": 0.6815,
1270
+ "step": 180
1271
+ },
1272
+ {
1273
+ "epoch": 0.4844429575108732,
1274
+ "grad_norm": 4.0202155113220215,
1275
+ "learning_rate": 9.16825564628371e-06,
1276
+ "loss": 0.7131,
1277
+ "step": 181
1278
+ },
1279
+ {
1280
+ "epoch": 0.48711943793911006,
1281
+ "grad_norm": 3.566044330596924,
1282
+ "learning_rate": 9.15578456753603e-06,
1283
+ "loss": 0.6628,
1284
+ "step": 182
1285
+ },
1286
+ {
1287
+ "epoch": 0.4897959183673469,
1288
+ "grad_norm": 3.9581563472747803,
1289
+ "learning_rate": 9.143229302532866e-06,
1290
+ "loss": 0.6497,
1291
+ "step": 183
1292
+ },
1293
+ {
1294
+ "epoch": 0.4924723987955838,
1295
+ "grad_norm": 3.2238948345184326,
1296
+ "learning_rate": 9.130590105613854e-06,
1297
+ "loss": 0.5773,
1298
+ "step": 184
1299
+ },
1300
+ {
1301
+ "epoch": 0.49514887922382067,
1302
+ "grad_norm": 4.199437618255615,
1303
+ "learning_rate": 9.117867232818897e-06,
1304
+ "loss": 0.5552,
1305
+ "step": 185
1306
+ },
1307
+ {
1308
+ "epoch": 0.49782535965205754,
1309
+ "grad_norm": 3.223834753036499,
1310
+ "learning_rate": 9.105060941882966e-06,
1311
+ "loss": 0.4985,
1312
+ "step": 186
1313
+ },
1314
+ {
1315
+ "epoch": 0.5005018400802944,
1316
+ "grad_norm": 3.2973475456237793,
1317
+ "learning_rate": 9.092171492230883e-06,
1318
+ "loss": 0.6637,
1319
+ "step": 187
1320
+ },
1321
+ {
1322
+ "epoch": 0.5031783205085313,
1323
+ "grad_norm": 3.9001471996307373,
1324
+ "learning_rate": 9.079199144972072e-06,
1325
+ "loss": 0.7487,
1326
+ "step": 188
1327
+ },
1328
+ {
1329
+ "epoch": 0.5058548009367682,
1330
+ "grad_norm": 3.913593292236328,
1331
+ "learning_rate": 9.066144162895259e-06,
1332
+ "loss": 0.5707,
1333
+ "step": 189
1334
+ },
1335
+ {
1336
+ "epoch": 0.508531281365005,
1337
+ "grad_norm": 2.6204168796539307,
1338
+ "learning_rate": 9.053006810463156e-06,
1339
+ "loss": 0.4685,
1340
+ "step": 190
1341
+ },
1342
+ {
1343
+ "epoch": 0.5112077617932419,
1344
+ "grad_norm": 3.507556200027466,
1345
+ "learning_rate": 9.039787353807101e-06,
1346
+ "loss": 0.5564,
1347
+ "step": 191
1348
+ },
1349
+ {
1350
+ "epoch": 0.5138842422214788,
1351
+ "grad_norm": 3.072106122970581,
1352
+ "learning_rate": 9.026486060721668e-06,
1353
+ "loss": 0.5427,
1354
+ "step": 192
1355
+ },
1356
+ {
1357
+ "epoch": 0.5165607226497156,
1358
+ "grad_norm": 2.578962802886963,
1359
+ "learning_rate": 9.01310320065924e-06,
1360
+ "loss": 0.3969,
1361
+ "step": 193
1362
+ },
1363
+ {
1364
+ "epoch": 0.5192372030779525,
1365
+ "grad_norm": 3.3628005981445312,
1366
+ "learning_rate": 8.999639044724555e-06,
1367
+ "loss": 0.6076,
1368
+ "step": 194
1369
+ },
1370
+ {
1371
+ "epoch": 0.5219136835061894,
1372
+ "grad_norm": 3.64412784576416,
1373
+ "learning_rate": 8.986093865669205e-06,
1374
+ "loss": 0.6826,
1375
+ "step": 195
1376
+ },
1377
+ {
1378
+ "epoch": 0.5245901639344263,
1379
+ "grad_norm": 3.7886106967926025,
1380
+ "learning_rate": 8.972467937886122e-06,
1381
+ "loss": 0.5279,
1382
+ "step": 196
1383
+ },
1384
+ {
1385
+ "epoch": 0.5272666443626631,
1386
+ "grad_norm": 3.543992757797241,
1387
+ "learning_rate": 8.958761537404012e-06,
1388
+ "loss": 0.5748,
1389
+ "step": 197
1390
+ },
1391
+ {
1392
+ "epoch": 0.5299431247909,
1393
+ "grad_norm": 4.291012763977051,
1394
+ "learning_rate": 8.944974941881766e-06,
1395
+ "loss": 0.6919,
1396
+ "step": 198
1397
+ },
1398
+ {
1399
+ "epoch": 0.5326196052191369,
1400
+ "grad_norm": 4.906763553619385,
1401
+ "learning_rate": 8.931108430602834e-06,
1402
+ "loss": 0.5633,
1403
+ "step": 199
1404
+ },
1405
+ {
1406
+ "epoch": 0.5352960856473737,
1407
+ "grad_norm": 3.177494764328003,
1408
+ "learning_rate": 8.917162284469569e-06,
1409
+ "loss": 0.5792,
1410
+ "step": 200
1411
+ },
1412
+ {
1413
+ "epoch": 0.5379725660756106,
1414
+ "grad_norm": 3.5298800468444824,
1415
+ "learning_rate": 8.903136785997533e-06,
1416
+ "loss": 0.5616,
1417
+ "step": 201
1418
+ },
1419
+ {
1420
+ "epoch": 0.5406490465038475,
1421
+ "grad_norm": 3.5532355308532715,
1422
+ "learning_rate": 8.889032219309781e-06,
1423
+ "loss": 0.6682,
1424
+ "step": 202
1425
+ },
1426
+ {
1427
+ "epoch": 0.5433255269320844,
1428
+ "grad_norm": 4.2544026374816895,
1429
+ "learning_rate": 8.874848870131098e-06,
1430
+ "loss": 0.544,
1431
+ "step": 203
1432
+ },
1433
+ {
1434
+ "epoch": 0.5460020073603212,
1435
+ "grad_norm": 3.0393524169921875,
1436
+ "learning_rate": 8.860587025782215e-06,
1437
+ "loss": 0.6333,
1438
+ "step": 204
1439
+ },
1440
+ {
1441
+ "epoch": 0.5486784877885581,
1442
+ "grad_norm": 4.1217122077941895,
1443
+ "learning_rate": 8.846246975173985e-06,
1444
+ "loss": 0.597,
1445
+ "step": 205
1446
+ },
1447
+ {
1448
+ "epoch": 0.5513549682167949,
1449
+ "grad_norm": 3.7197227478027344,
1450
+ "learning_rate": 8.831829008801536e-06,
1451
+ "loss": 0.6568,
1452
+ "step": 206
1453
+ },
1454
+ {
1455
+ "epoch": 0.5540314486450317,
1456
+ "grad_norm": 3.254206657409668,
1457
+ "learning_rate": 8.817333418738382e-06,
1458
+ "loss": 0.5321,
1459
+ "step": 207
1460
+ },
1461
+ {
1462
+ "epoch": 0.5567079290732686,
1463
+ "grad_norm": 3.084484577178955,
1464
+ "learning_rate": 8.802760498630507e-06,
1465
+ "loss": 0.4752,
1466
+ "step": 208
1467
+ },
1468
+ {
1469
+ "epoch": 0.5593844095015055,
1470
+ "grad_norm": 4.343369483947754,
1471
+ "learning_rate": 8.788110543690415e-06,
1472
+ "loss": 0.6444,
1473
+ "step": 209
1474
+ },
1475
+ {
1476
+ "epoch": 0.5620608899297423,
1477
+ "grad_norm": 3.27128529548645,
1478
+ "learning_rate": 8.773383850691155e-06,
1479
+ "loss": 0.5282,
1480
+ "step": 210
1481
+ },
1482
+ {
1483
+ "epoch": 0.5647373703579792,
1484
+ "grad_norm": 3.0274081230163574,
1485
+ "learning_rate": 8.758580717960303e-06,
1486
+ "loss": 0.6607,
1487
+ "step": 211
1488
+ },
1489
+ {
1490
+ "epoch": 0.5674138507862161,
1491
+ "grad_norm": 3.622535228729248,
1492
+ "learning_rate": 8.743701445373922e-06,
1493
+ "loss": 0.5787,
1494
+ "step": 212
1495
+ },
1496
+ {
1497
+ "epoch": 0.570090331214453,
1498
+ "grad_norm": 3.6681504249572754,
1499
+ "learning_rate": 8.728746334350483e-06,
1500
+ "loss": 0.5879,
1501
+ "step": 213
1502
+ },
1503
+ {
1504
+ "epoch": 0.5727668116426898,
1505
+ "grad_norm": 3.163616895675659,
1506
+ "learning_rate": 8.713715687844772e-06,
1507
+ "loss": 0.5888,
1508
+ "step": 214
1509
+ },
1510
+ {
1511
+ "epoch": 0.5754432920709267,
1512
+ "grad_norm": 3.0883967876434326,
1513
+ "learning_rate": 8.698609810341733e-06,
1514
+ "loss": 0.5189,
1515
+ "step": 215
1516
+ },
1517
+ {
1518
+ "epoch": 0.5781197724991636,
1519
+ "grad_norm": 3.443776845932007,
1520
+ "learning_rate": 8.683429007850313e-06,
1521
+ "loss": 0.5303,
1522
+ "step": 216
1523
+ },
1524
+ {
1525
+ "epoch": 0.5807962529274004,
1526
+ "grad_norm": 3.3351690769195557,
1527
+ "learning_rate": 8.668173587897261e-06,
1528
+ "loss": 0.5296,
1529
+ "step": 217
1530
+ },
1531
+ {
1532
+ "epoch": 0.5834727333556373,
1533
+ "grad_norm": 3.8925840854644775,
1534
+ "learning_rate": 8.6528438595209e-06,
1535
+ "loss": 0.499,
1536
+ "step": 218
1537
+ },
1538
+ {
1539
+ "epoch": 0.5861492137838742,
1540
+ "grad_norm": 3.4368419647216797,
1541
+ "learning_rate": 8.637440133264858e-06,
1542
+ "loss": 0.4166,
1543
+ "step": 219
1544
+ },
1545
+ {
1546
+ "epoch": 0.5888256942121111,
1547
+ "grad_norm": 3.439969539642334,
1548
+ "learning_rate": 8.621962721171789e-06,
1549
+ "loss": 0.5417,
1550
+ "step": 220
1551
+ },
1552
+ {
1553
+ "epoch": 0.5915021746403479,
1554
+ "grad_norm": 3.2590739727020264,
1555
+ "learning_rate": 8.60641193677704e-06,
1556
+ "loss": 0.4925,
1557
+ "step": 221
1558
+ },
1559
+ {
1560
+ "epoch": 0.5941786550685848,
1561
+ "grad_norm": 3.918884754180908,
1562
+ "learning_rate": 8.59078809510231e-06,
1563
+ "loss": 0.75,
1564
+ "step": 222
1565
+ },
1566
+ {
1567
+ "epoch": 0.5968551354968217,
1568
+ "grad_norm": 4.21658182144165,
1569
+ "learning_rate": 8.57509151264926e-06,
1570
+ "loss": 0.5138,
1571
+ "step": 223
1572
+ },
1573
+ {
1574
+ "epoch": 0.5995316159250585,
1575
+ "grad_norm": 3.2700538635253906,
1576
+ "learning_rate": 8.55932250739311e-06,
1577
+ "loss": 0.4855,
1578
+ "step": 224
1579
+ },
1580
+ {
1581
+ "epoch": 0.6022080963532954,
1582
+ "grad_norm": 3.5396180152893066,
1583
+ "learning_rate": 8.543481398776188e-06,
1584
+ "loss": 0.626,
1585
+ "step": 225
1586
+ },
1587
+ {
1588
+ "epoch": 0.6048845767815323,
1589
+ "grad_norm": 3.267137289047241,
1590
+ "learning_rate": 8.527568507701467e-06,
1591
+ "loss": 0.5043,
1592
+ "step": 226
1593
+ },
1594
+ {
1595
+ "epoch": 0.6075610572097692,
1596
+ "grad_norm": 4.908371448516846,
1597
+ "learning_rate": 8.511584156526059e-06,
1598
+ "loss": 0.6942,
1599
+ "step": 227
1600
+ },
1601
+ {
1602
+ "epoch": 0.610237537638006,
1603
+ "grad_norm": 3.7445895671844482,
1604
+ "learning_rate": 8.495528669054688e-06,
1605
+ "loss": 0.5475,
1606
+ "step": 228
1607
+ },
1608
+ {
1609
+ "epoch": 0.6129140180662429,
1610
+ "grad_norm": 4.3304033279418945,
1611
+ "learning_rate": 8.479402370533127e-06,
1612
+ "loss": 0.6943,
1613
+ "step": 229
1614
+ },
1615
+ {
1616
+ "epoch": 0.6155904984944798,
1617
+ "grad_norm": 3.6123831272125244,
1618
+ "learning_rate": 8.463205587641614e-06,
1619
+ "loss": 0.5365,
1620
+ "step": 230
1621
+ },
1622
+ {
1623
+ "epoch": 0.6182669789227166,
1624
+ "grad_norm": 4.3004045486450195,
1625
+ "learning_rate": 8.44693864848823e-06,
1626
+ "loss": 0.6377,
1627
+ "step": 231
1628
+ },
1629
+ {
1630
+ "epoch": 0.6209434593509535,
1631
+ "grad_norm": 3.1220314502716064,
1632
+ "learning_rate": 8.430601882602256e-06,
1633
+ "loss": 0.4646,
1634
+ "step": 232
1635
+ },
1636
+ {
1637
+ "epoch": 0.6236199397791904,
1638
+ "grad_norm": 4.166379451751709,
1639
+ "learning_rate": 8.414195620927491e-06,
1640
+ "loss": 0.5738,
1641
+ "step": 233
1642
+ },
1643
+ {
1644
+ "epoch": 0.6262964202074273,
1645
+ "grad_norm": 4.061609745025635,
1646
+ "learning_rate": 8.397720195815561e-06,
1647
+ "loss": 0.4654,
1648
+ "step": 234
1649
+ },
1650
+ {
1651
+ "epoch": 0.6289729006356641,
1652
+ "grad_norm": 3.618021249771118,
1653
+ "learning_rate": 8.381175941019171e-06,
1654
+ "loss": 0.5699,
1655
+ "step": 235
1656
+ },
1657
+ {
1658
+ "epoch": 0.631649381063901,
1659
+ "grad_norm": 3.2099597454071045,
1660
+ "learning_rate": 8.364563191685348e-06,
1661
+ "loss": 0.5481,
1662
+ "step": 236
1663
+ },
1664
+ {
1665
+ "epoch": 0.6343258614921379,
1666
+ "grad_norm": 3.3331351280212402,
1667
+ "learning_rate": 8.347882284348665e-06,
1668
+ "loss": 0.5351,
1669
+ "step": 237
1670
+ },
1671
+ {
1672
+ "epoch": 0.6370023419203747,
1673
+ "grad_norm": 3.91636323928833,
1674
+ "learning_rate": 8.331133556924404e-06,
1675
+ "loss": 0.6588,
1676
+ "step": 238
1677
+ },
1678
+ {
1679
+ "epoch": 0.6396788223486116,
1680
+ "grad_norm": 3.613023519515991,
1681
+ "learning_rate": 8.314317348701724e-06,
1682
+ "loss": 0.6157,
1683
+ "step": 239
1684
+ },
1685
+ {
1686
+ "epoch": 0.6423553027768485,
1687
+ "grad_norm": 3.674731731414795,
1688
+ "learning_rate": 8.297434000336781e-06,
1689
+ "loss": 0.5251,
1690
+ "step": 240
1691
+ },
1692
+ {
1693
+ "epoch": 0.6450317832050854,
1694
+ "grad_norm": 3.4246506690979004,
1695
+ "learning_rate": 8.280483853845831e-06,
1696
+ "loss": 0.5449,
1697
+ "step": 241
1698
+ },
1699
+ {
1700
+ "epoch": 0.6477082636333222,
1701
+ "grad_norm": 3.308436155319214,
1702
+ "learning_rate": 8.263467252598303e-06,
1703
+ "loss": 0.6432,
1704
+ "step": 242
1705
+ },
1706
+ {
1707
+ "epoch": 0.650384744061559,
1708
+ "grad_norm": 3.3527414798736572,
1709
+ "learning_rate": 8.246384541309835e-06,
1710
+ "loss": 0.6035,
1711
+ "step": 243
1712
+ },
1713
+ {
1714
+ "epoch": 0.6530612244897959,
1715
+ "grad_norm": 3.5172030925750732,
1716
+ "learning_rate": 8.2292360660353e-06,
1717
+ "loss": 0.522,
1718
+ "step": 244
1719
+ },
1720
+ {
1721
+ "epoch": 0.6557377049180327,
1722
+ "grad_norm": 3.086130142211914,
1723
+ "learning_rate": 8.21202217416179e-06,
1724
+ "loss": 0.4867,
1725
+ "step": 245
1726
+ },
1727
+ {
1728
+ "epoch": 0.6584141853462696,
1729
+ "grad_norm": 4.54761266708374,
1730
+ "learning_rate": 8.194743214401587e-06,
1731
+ "loss": 0.5607,
1732
+ "step": 246
1733
+ },
1734
+ {
1735
+ "epoch": 0.6610906657745065,
1736
+ "grad_norm": 3.190037250518799,
1737
+ "learning_rate": 8.17739953678508e-06,
1738
+ "loss": 0.4674,
1739
+ "step": 247
1740
+ },
1741
+ {
1742
+ "epoch": 0.6637671462027434,
1743
+ "grad_norm": 3.303924083709717,
1744
+ "learning_rate": 8.1599914926537e-06,
1745
+ "loss": 0.5116,
1746
+ "step": 248
1747
+ },
1748
+ {
1749
+ "epoch": 0.6664436266309802,
1750
+ "grad_norm": 3.0905206203460693,
1751
+ "learning_rate": 8.142519434652782e-06,
1752
+ "loss": 0.5041,
1753
+ "step": 249
1754
+ },
1755
+ {
1756
+ "epoch": 0.6691201070592171,
1757
+ "grad_norm": 3.9465432167053223,
1758
+ "learning_rate": 8.124983716724434e-06,
1759
+ "loss": 0.5009,
1760
+ "step": 250
1761
+ },
1762
+ {
1763
+ "epoch": 0.671796587487454,
1764
+ "grad_norm": 3.0109636783599854,
1765
+ "learning_rate": 8.107384694100355e-06,
1766
+ "loss": 0.5041,
1767
+ "step": 251
1768
+ },
1769
+ {
1770
+ "epoch": 0.6744730679156908,
1771
+ "grad_norm": 3.4623425006866455,
1772
+ "learning_rate": 8.089722723294654e-06,
1773
+ "loss": 0.5509,
1774
+ "step": 252
1775
+ },
1776
+ {
1777
+ "epoch": 0.6771495483439277,
1778
+ "grad_norm": 4.234053134918213,
1779
+ "learning_rate": 8.071998162096613e-06,
1780
+ "loss": 0.6164,
1781
+ "step": 253
1782
+ },
1783
+ {
1784
+ "epoch": 0.6798260287721646,
1785
+ "grad_norm": 3.2587759494781494,
1786
+ "learning_rate": 8.054211369563448e-06,
1787
+ "loss": 0.5531,
1788
+ "step": 254
1789
+ },
1790
+ {
1791
+ "epoch": 0.6825025092004015,
1792
+ "grad_norm": 3.6140904426574707,
1793
+ "learning_rate": 8.036362706013033e-06,
1794
+ "loss": 0.5566,
1795
+ "step": 255
1796
+ },
1797
+ {
1798
+ "epoch": 0.6851789896286383,
1799
+ "grad_norm": 2.9970974922180176,
1800
+ "learning_rate": 8.018452533016604e-06,
1801
+ "loss": 0.4614,
1802
+ "step": 256
1803
+ },
1804
+ {
1805
+ "epoch": 0.6878554700568752,
1806
+ "grad_norm": 3.6728920936584473,
1807
+ "learning_rate": 8.000481213391422e-06,
1808
+ "loss": 0.5529,
1809
+ "step": 257
1810
+ },
1811
+ {
1812
+ "epoch": 0.6905319504851121,
1813
+ "grad_norm": 4.451485633850098,
1814
+ "learning_rate": 7.982449111193445e-06,
1815
+ "loss": 0.5997,
1816
+ "step": 258
1817
+ },
1818
+ {
1819
+ "epoch": 0.6932084309133489,
1820
+ "grad_norm": 3.2221109867095947,
1821
+ "learning_rate": 7.96435659170993e-06,
1822
+ "loss": 0.4782,
1823
+ "step": 259
1824
+ },
1825
+ {
1826
+ "epoch": 0.6958849113415858,
1827
+ "grad_norm": 3.3525805473327637,
1828
+ "learning_rate": 7.946204021452049e-06,
1829
+ "loss": 0.5988,
1830
+ "step": 260
1831
+ },
1832
+ {
1833
+ "epoch": 0.6985613917698227,
1834
+ "grad_norm": 3.346829652786255,
1835
+ "learning_rate": 7.92799176814746e-06,
1836
+ "loss": 0.461,
1837
+ "step": 261
1838
+ },
1839
+ {
1840
+ "epoch": 0.7012378721980596,
1841
+ "grad_norm": 3.4841763973236084,
1842
+ "learning_rate": 7.90972020073285e-06,
1843
+ "loss": 0.5972,
1844
+ "step": 262
1845
+ },
1846
+ {
1847
+ "epoch": 0.7039143526262964,
1848
+ "grad_norm": 3.1365015506744385,
1849
+ "learning_rate": 7.891389689346479e-06,
1850
+ "loss": 0.4601,
1851
+ "step": 263
1852
+ },
1853
+ {
1854
+ "epoch": 0.7065908330545333,
1855
+ "grad_norm": 2.762209177017212,
1856
+ "learning_rate": 7.873000605320658e-06,
1857
+ "loss": 0.4713,
1858
+ "step": 264
1859
+ },
1860
+ {
1861
+ "epoch": 0.7092673134827702,
1862
+ "grad_norm": 3.6923787593841553,
1863
+ "learning_rate": 7.85455332117425e-06,
1864
+ "loss": 0.5868,
1865
+ "step": 265
1866
+ },
1867
+ {
1868
+ "epoch": 0.711943793911007,
1869
+ "grad_norm": 3.483915090560913,
1870
+ "learning_rate": 7.836048210605109e-06,
1871
+ "loss": 0.5488,
1872
+ "step": 266
1873
+ },
1874
+ {
1875
+ "epoch": 0.7146202743392439,
1876
+ "grad_norm": 3.7506582736968994,
1877
+ "learning_rate": 7.817485648482514e-06,
1878
+ "loss": 0.506,
1879
+ "step": 267
1880
+ },
1881
+ {
1882
+ "epoch": 0.7172967547674808,
1883
+ "grad_norm": 3.148474931716919,
1884
+ "learning_rate": 7.798866010839577e-06,
1885
+ "loss": 0.4793,
1886
+ "step": 268
1887
+ },
1888
+ {
1889
+ "epoch": 0.7199732351957177,
1890
+ "grad_norm": 3.194519281387329,
1891
+ "learning_rate": 7.780189674865617e-06,
1892
+ "loss": 0.5573,
1893
+ "step": 269
1894
+ },
1895
+ {
1896
+ "epoch": 0.7226497156239545,
1897
+ "grad_norm": 3.321631669998169,
1898
+ "learning_rate": 7.761457018898536e-06,
1899
+ "loss": 0.4949,
1900
+ "step": 270
1901
+ },
1902
+ {
1903
+ "epoch": 0.7253261960521914,
1904
+ "grad_norm": 3.322242021560669,
1905
+ "learning_rate": 7.742668422417137e-06,
1906
+ "loss": 0.5329,
1907
+ "step": 271
1908
+ },
1909
+ {
1910
+ "epoch": 0.7280026764804283,
1911
+ "grad_norm": 3.460432291030884,
1912
+ "learning_rate": 7.723824266033444e-06,
1913
+ "loss": 0.5798,
1914
+ "step": 272
1915
+ },
1916
+ {
1917
+ "epoch": 0.7306791569086651,
1918
+ "grad_norm": 3.0027997493743896,
1919
+ "learning_rate": 7.704924931484997e-06,
1920
+ "loss": 0.5512,
1921
+ "step": 273
1922
+ },
1923
+ {
1924
+ "epoch": 0.733355637336902,
1925
+ "grad_norm": 3.621713399887085,
1926
+ "learning_rate": 7.685970801627108e-06,
1927
+ "loss": 0.6221,
1928
+ "step": 274
1929
+ },
1930
+ {
1931
+ "epoch": 0.7360321177651389,
1932
+ "grad_norm": 2.8740806579589844,
1933
+ "learning_rate": 7.666962260425113e-06,
1934
+ "loss": 0.532,
1935
+ "step": 275
1936
+ },
1937
+ {
1938
+ "epoch": 0.7387085981933758,
1939
+ "grad_norm": 2.9761555194854736,
1940
+ "learning_rate": 7.647899692946594e-06,
1941
+ "loss": 0.4211,
1942
+ "step": 276
1943
+ },
1944
+ {
1945
+ "epoch": 0.7413850786216126,
1946
+ "grad_norm": 2.8922770023345947,
1947
+ "learning_rate": 7.628783485353573e-06,
1948
+ "loss": 0.4465,
1949
+ "step": 277
1950
+ },
1951
+ {
1952
+ "epoch": 0.7440615590498495,
1953
+ "grad_norm": 2.775923252105713,
1954
+ "learning_rate": 7.609614024894694e-06,
1955
+ "loss": 0.46,
1956
+ "step": 278
1957
+ },
1958
+ {
1959
+ "epoch": 0.7467380394780864,
1960
+ "grad_norm": 3.38106632232666,
1961
+ "learning_rate": 7.5903916998973745e-06,
1962
+ "loss": 0.6024,
1963
+ "step": 279
1964
+ },
1965
+ {
1966
+ "epoch": 0.7494145199063232,
1967
+ "grad_norm": 2.9495768547058105,
1968
+ "learning_rate": 7.571116899759945e-06,
1969
+ "loss": 0.4792,
1970
+ "step": 280
1971
+ },
1972
+ {
1973
+ "epoch": 0.75209100033456,
1974
+ "grad_norm": 3.18288516998291,
1975
+ "learning_rate": 7.551790014943752e-06,
1976
+ "loss": 0.4439,
1977
+ "step": 281
1978
+ },
1979
+ {
1980
+ "epoch": 0.7547674807627969,
1981
+ "grad_norm": 3.185044288635254,
1982
+ "learning_rate": 7.532411436965258e-06,
1983
+ "loss": 0.5251,
1984
+ "step": 282
1985
+ },
1986
+ {
1987
+ "epoch": 0.7574439611910337,
1988
+ "grad_norm": 2.8244149684906006,
1989
+ "learning_rate": 7.512981558388101e-06,
1990
+ "loss": 0.4443,
1991
+ "step": 283
1992
+ },
1993
+ {
1994
+ "epoch": 0.7601204416192706,
1995
+ "grad_norm": 3.172478437423706,
1996
+ "learning_rate": 7.49350077281515e-06,
1997
+ "loss": 0.5057,
1998
+ "step": 284
1999
+ },
2000
+ {
2001
+ "epoch": 0.7627969220475075,
2002
+ "grad_norm": 3.2806384563446045,
2003
+ "learning_rate": 7.473969474880527e-06,
2004
+ "loss": 0.5944,
2005
+ "step": 285
2006
+ },
2007
+ {
2008
+ "epoch": 0.7654734024757444,
2009
+ "grad_norm": 2.7359542846679688,
2010
+ "learning_rate": 7.45438806024161e-06,
2011
+ "loss": 0.4296,
2012
+ "step": 286
2013
+ },
2014
+ {
2015
+ "epoch": 0.7681498829039812,
2016
+ "grad_norm": 2.7269349098205566,
2017
+ "learning_rate": 7.4347569255710254e-06,
2018
+ "loss": 0.4004,
2019
+ "step": 287
2020
+ },
2021
+ {
2022
+ "epoch": 0.7708263633322181,
2023
+ "grad_norm": 3.779465913772583,
2024
+ "learning_rate": 7.41507646854861e-06,
2025
+ "loss": 0.5674,
2026
+ "step": 288
2027
+ },
2028
+ {
2029
+ "epoch": 0.773502843760455,
2030
+ "grad_norm": 2.7215170860290527,
2031
+ "learning_rate": 7.395347087853349e-06,
2032
+ "loss": 0.3874,
2033
+ "step": 289
2034
+ },
2035
+ {
2036
+ "epoch": 0.7761793241886918,
2037
+ "grad_norm": 4.240821838378906,
2038
+ "learning_rate": 7.375569183155306e-06,
2039
+ "loss": 0.5878,
2040
+ "step": 290
2041
+ },
2042
+ {
2043
+ "epoch": 0.7788558046169287,
2044
+ "grad_norm": 3.43664288520813,
2045
+ "learning_rate": 7.355743155107526e-06,
2046
+ "loss": 0.4482,
2047
+ "step": 291
2048
+ },
2049
+ {
2050
+ "epoch": 0.7815322850451656,
2051
+ "grad_norm": 3.580437660217285,
2052
+ "learning_rate": 7.335869405337919e-06,
2053
+ "loss": 0.5135,
2054
+ "step": 292
2055
+ },
2056
+ {
2057
+ "epoch": 0.7842087654734025,
2058
+ "grad_norm": 3.794210433959961,
2059
+ "learning_rate": 7.3159483364411175e-06,
2060
+ "loss": 0.4546,
2061
+ "step": 293
2062
+ },
2063
+ {
2064
+ "epoch": 0.7868852459016393,
2065
+ "grad_norm": 2.8526501655578613,
2066
+ "learning_rate": 7.295980351970331e-06,
2067
+ "loss": 0.416,
2068
+ "step": 294
2069
+ },
2070
+ {
2071
+ "epoch": 0.7895617263298762,
2072
+ "grad_norm": 3.249462127685547,
2073
+ "learning_rate": 7.275965856429167e-06,
2074
+ "loss": 0.4575,
2075
+ "step": 295
2076
+ },
2077
+ {
2078
+ "epoch": 0.7922382067581131,
2079
+ "grad_norm": 4.065032482147217,
2080
+ "learning_rate": 7.255905255263434e-06,
2081
+ "loss": 0.4073,
2082
+ "step": 296
2083
+ },
2084
+ {
2085
+ "epoch": 0.7949146871863499,
2086
+ "grad_norm": 2.9181621074676514,
2087
+ "learning_rate": 7.235798954852929e-06,
2088
+ "loss": 0.467,
2089
+ "step": 297
2090
+ },
2091
+ {
2092
+ "epoch": 0.7975911676145868,
2093
+ "grad_norm": 3.012425661087036,
2094
+ "learning_rate": 7.2156473625032075e-06,
2095
+ "loss": 0.5308,
2096
+ "step": 298
2097
+ },
2098
+ {
2099
+ "epoch": 0.8002676480428237,
2100
+ "grad_norm": 3.2600486278533936,
2101
+ "learning_rate": 7.195450886437334e-06,
2102
+ "loss": 0.5594,
2103
+ "step": 299
2104
+ },
2105
+ {
2106
+ "epoch": 0.8029441284710606,
2107
+ "grad_norm": 3.5234806537628174,
2108
+ "learning_rate": 7.175209935787605e-06,
2109
+ "loss": 0.4377,
2110
+ "step": 300
2111
+ },
2112
+ {
2113
+ "epoch": 0.8056206088992974,
2114
+ "grad_norm": 3.3013510704040527,
2115
+ "learning_rate": 7.154924920587269e-06,
2116
+ "loss": 0.5039,
2117
+ "step": 301
2118
+ },
2119
+ {
2120
+ "epoch": 0.8082970893275343,
2121
+ "grad_norm": 3.6095550060272217,
2122
+ "learning_rate": 7.134596251762217e-06,
2123
+ "loss": 0.5333,
2124
+ "step": 302
2125
+ },
2126
+ {
2127
+ "epoch": 0.8109735697557712,
2128
+ "grad_norm": 3.533142328262329,
2129
+ "learning_rate": 7.114224341122655e-06,
2130
+ "loss": 0.4358,
2131
+ "step": 303
2132
+ },
2133
+ {
2134
+ "epoch": 0.813650050184008,
2135
+ "grad_norm": 3.41983699798584,
2136
+ "learning_rate": 7.093809601354769e-06,
2137
+ "loss": 0.5369,
2138
+ "step": 304
2139
+ },
2140
+ {
2141
+ "epoch": 0.8163265306122449,
2142
+ "grad_norm": 2.5269010066986084,
2143
+ "learning_rate": 7.073352446012357e-06,
2144
+ "loss": 0.4014,
2145
+ "step": 305
2146
+ },
2147
+ {
2148
+ "epoch": 0.8190030110404818,
2149
+ "grad_norm": 3.3515594005584717,
2150
+ "learning_rate": 7.052853289508458e-06,
2151
+ "loss": 0.4681,
2152
+ "step": 306
2153
+ },
2154
+ {
2155
+ "epoch": 0.8216794914687187,
2156
+ "grad_norm": 3.920032501220703,
2157
+ "learning_rate": 7.03231254710695e-06,
2158
+ "loss": 0.5876,
2159
+ "step": 307
2160
+ },
2161
+ {
2162
+ "epoch": 0.8243559718969555,
2163
+ "grad_norm": 2.734036684036255,
2164
+ "learning_rate": 7.0117306349141485e-06,
2165
+ "loss": 0.4294,
2166
+ "step": 308
2167
+ },
2168
+ {
2169
+ "epoch": 0.8270324523251924,
2170
+ "grad_norm": 2.5630767345428467,
2171
+ "learning_rate": 6.991107969870363e-06,
2172
+ "loss": 0.3869,
2173
+ "step": 309
2174
+ },
2175
+ {
2176
+ "epoch": 0.8297089327534293,
2177
+ "grad_norm": 2.897214889526367,
2178
+ "learning_rate": 6.970444969741462e-06,
2179
+ "loss": 0.4756,
2180
+ "step": 310
2181
+ },
2182
+ {
2183
+ "epoch": 0.8323854131816661,
2184
+ "grad_norm": 3.645385980606079,
2185
+ "learning_rate": 6.949742053110408e-06,
2186
+ "loss": 0.465,
2187
+ "step": 311
2188
+ },
2189
+ {
2190
+ "epoch": 0.835061893609903,
2191
+ "grad_norm": 2.9441306591033936,
2192
+ "learning_rate": 6.928999639368773e-06,
2193
+ "loss": 0.4803,
2194
+ "step": 312
2195
+ },
2196
+ {
2197
+ "epoch": 0.8377383740381399,
2198
+ "grad_norm": 3.1111350059509277,
2199
+ "learning_rate": 6.908218148708248e-06,
2200
+ "loss": 0.4119,
2201
+ "step": 313
2202
+ },
2203
+ {
2204
+ "epoch": 0.8404148544663768,
2205
+ "grad_norm": 2.8598246574401855,
2206
+ "learning_rate": 6.887398002112129e-06,
2207
+ "loss": 0.5,
2208
+ "step": 314
2209
+ },
2210
+ {
2211
+ "epoch": 0.8430913348946136,
2212
+ "grad_norm": 2.703573703765869,
2213
+ "learning_rate": 6.866539621346786e-06,
2214
+ "loss": 0.3387,
2215
+ "step": 315
2216
+ },
2217
+ {
2218
+ "epoch": 0.8457678153228505,
2219
+ "grad_norm": 2.924374580383301,
2220
+ "learning_rate": 6.845643428953127e-06,
2221
+ "loss": 0.4498,
2222
+ "step": 316
2223
+ },
2224
+ {
2225
+ "epoch": 0.8484442957510874,
2226
+ "grad_norm": 3.2768142223358154,
2227
+ "learning_rate": 6.824709848238028e-06,
2228
+ "loss": 0.4514,
2229
+ "step": 317
2230
+ },
2231
+ {
2232
+ "epoch": 0.8511207761793241,
2233
+ "grad_norm": 3.422415256500244,
2234
+ "learning_rate": 6.8037393032657665e-06,
2235
+ "loss": 0.4806,
2236
+ "step": 318
2237
+ },
2238
+ {
2239
+ "epoch": 0.853797256607561,
2240
+ "grad_norm": 3.6226656436920166,
2241
+ "learning_rate": 6.782732218849425e-06,
2242
+ "loss": 0.431,
2243
+ "step": 319
2244
+ },
2245
+ {
2246
+ "epoch": 0.8564737370357979,
2247
+ "grad_norm": 3.154444932937622,
2248
+ "learning_rate": 6.761689020542288e-06,
2249
+ "loss": 0.4059,
2250
+ "step": 320
2251
+ },
2252
+ {
2253
+ "epoch": 0.8591502174640347,
2254
+ "grad_norm": 3.5488181114196777,
2255
+ "learning_rate": 6.740610134629224e-06,
2256
+ "loss": 0.5335,
2257
+ "step": 321
2258
+ },
2259
+ {
2260
+ "epoch": 0.8618266978922716,
2261
+ "grad_norm": 3.483675718307495,
2262
+ "learning_rate": 6.719495988118043e-06,
2263
+ "loss": 0.5676,
2264
+ "step": 322
2265
+ },
2266
+ {
2267
+ "epoch": 0.8645031783205085,
2268
+ "grad_norm": 3.83905291557312,
2269
+ "learning_rate": 6.698347008730854e-06,
2270
+ "loss": 0.5997,
2271
+ "step": 323
2272
+ },
2273
+ {
2274
+ "epoch": 0.8671796587487454,
2275
+ "grad_norm": 3.451472759246826,
2276
+ "learning_rate": 6.677163624895393e-06,
2277
+ "loss": 0.4285,
2278
+ "step": 324
2279
+ },
2280
+ {
2281
+ "epoch": 0.8698561391769822,
2282
+ "grad_norm": 3.9657328128814697,
2283
+ "learning_rate": 6.6559462657363525e-06,
2284
+ "loss": 0.5505,
2285
+ "step": 325
2286
+ },
2287
+ {
2288
+ "epoch": 0.8725326196052191,
2289
+ "grad_norm": 3.474752902984619,
2290
+ "learning_rate": 6.634695361066679e-06,
2291
+ "loss": 0.577,
2292
+ "step": 326
2293
+ },
2294
+ {
2295
+ "epoch": 0.875209100033456,
2296
+ "grad_norm": 3.0896995067596436,
2297
+ "learning_rate": 6.613411341378872e-06,
2298
+ "loss": 0.4052,
2299
+ "step": 327
2300
+ },
2301
+ {
2302
+ "epoch": 0.8778855804616928,
2303
+ "grad_norm": 2.541867256164551,
2304
+ "learning_rate": 6.592094637836266e-06,
2305
+ "loss": 0.3609,
2306
+ "step": 328
2307
+ },
2308
+ {
2309
+ "epoch": 0.8805620608899297,
2310
+ "grad_norm": 3.222644329071045,
2311
+ "learning_rate": 6.570745682264288e-06,
2312
+ "loss": 0.4346,
2313
+ "step": 329
2314
+ },
2315
+ {
2316
+ "epoch": 0.8832385413181666,
2317
+ "grad_norm": 2.9932363033294678,
2318
+ "learning_rate": 6.549364907141713e-06,
2319
+ "loss": 0.3862,
2320
+ "step": 330
2321
+ },
2322
+ {
2323
+ "epoch": 0.8859150217464035,
2324
+ "grad_norm": 2.610246181488037,
2325
+ "learning_rate": 6.527952745591911e-06,
2326
+ "loss": 0.3075,
2327
+ "step": 331
2328
+ },
2329
+ {
2330
+ "epoch": 0.8885915021746403,
2331
+ "grad_norm": 3.310206174850464,
2332
+ "learning_rate": 6.506509631374056e-06,
2333
+ "loss": 0.4288,
2334
+ "step": 332
2335
+ },
2336
+ {
2337
+ "epoch": 0.8912679826028772,
2338
+ "grad_norm": 3.0393242835998535,
2339
+ "learning_rate": 6.485035998874356e-06,
2340
+ "loss": 0.5224,
2341
+ "step": 333
2342
+ },
2343
+ {
2344
+ "epoch": 0.8939444630311141,
2345
+ "grad_norm": 3.029529571533203,
2346
+ "learning_rate": 6.4635322830972465e-06,
2347
+ "loss": 0.5228,
2348
+ "step": 334
2349
+ },
2350
+ {
2351
+ "epoch": 0.896620943459351,
2352
+ "grad_norm": 2.9739904403686523,
2353
+ "learning_rate": 6.441998919656575e-06,
2354
+ "loss": 0.4464,
2355
+ "step": 335
2356
+ },
2357
+ {
2358
+ "epoch": 0.8992974238875878,
2359
+ "grad_norm": 3.4808287620544434,
2360
+ "learning_rate": 6.420436344766781e-06,
2361
+ "loss": 0.4787,
2362
+ "step": 336
2363
+ },
2364
+ {
2365
+ "epoch": 0.9019739043158247,
2366
+ "grad_norm": 3.4027199745178223,
2367
+ "learning_rate": 6.398844995234057e-06,
2368
+ "loss": 0.4964,
2369
+ "step": 337
2370
+ },
2371
+ {
2372
+ "epoch": 0.9046503847440616,
2373
+ "grad_norm": 3.5456416606903076,
2374
+ "learning_rate": 6.377225308447503e-06,
2375
+ "loss": 0.5533,
2376
+ "step": 338
2377
+ },
2378
+ {
2379
+ "epoch": 0.9073268651722984,
2380
+ "grad_norm": 3.416761875152588,
2381
+ "learning_rate": 6.355577722370264e-06,
2382
+ "loss": 0.4679,
2383
+ "step": 339
2384
+ },
2385
+ {
2386
+ "epoch": 0.9100033456005353,
2387
+ "grad_norm": 3.634335994720459,
2388
+ "learning_rate": 6.333902675530657e-06,
2389
+ "loss": 0.4683,
2390
+ "step": 340
2391
+ },
2392
+ {
2393
+ "epoch": 0.9126798260287722,
2394
+ "grad_norm": 2.5649242401123047,
2395
+ "learning_rate": 6.312200607013287e-06,
2396
+ "loss": 0.3649,
2397
+ "step": 341
2398
+ },
2399
+ {
2400
+ "epoch": 0.915356306457009,
2401
+ "grad_norm": 3.457108736038208,
2402
+ "learning_rate": 6.2904719564501545e-06,
2403
+ "loss": 0.559,
2404
+ "step": 342
2405
+ },
2406
+ {
2407
+ "epoch": 0.9180327868852459,
2408
+ "grad_norm": 3.2710628509521484,
2409
+ "learning_rate": 6.268717164011751e-06,
2410
+ "loss": 0.4526,
2411
+ "step": 343
2412
+ },
2413
+ {
2414
+ "epoch": 0.9207092673134828,
2415
+ "grad_norm": 2.584167003631592,
2416
+ "learning_rate": 6.246936670398136e-06,
2417
+ "loss": 0.3223,
2418
+ "step": 344
2419
+ },
2420
+ {
2421
+ "epoch": 0.9233857477417197,
2422
+ "grad_norm": 3.2143352031707764,
2423
+ "learning_rate": 6.225130916830017e-06,
2424
+ "loss": 0.3996,
2425
+ "step": 345
2426
+ },
2427
+ {
2428
+ "epoch": 0.9260622281699565,
2429
+ "grad_norm": 2.900538206100464,
2430
+ "learning_rate": 6.203300345039804e-06,
2431
+ "loss": 0.4241,
2432
+ "step": 346
2433
+ },
2434
+ {
2435
+ "epoch": 0.9287387085981934,
2436
+ "grad_norm": 2.566956043243408,
2437
+ "learning_rate": 6.181445397262671e-06,
2438
+ "loss": 0.3754,
2439
+ "step": 347
2440
+ },
2441
+ {
2442
+ "epoch": 0.9314151890264303,
2443
+ "grad_norm": 3.051973819732666,
2444
+ "learning_rate": 6.159566516227582e-06,
2445
+ "loss": 0.464,
2446
+ "step": 348
2447
+ },
2448
+ {
2449
+ "epoch": 0.9340916694546672,
2450
+ "grad_norm": 2.648303985595703,
2451
+ "learning_rate": 6.137664145148339e-06,
2452
+ "loss": 0.3669,
2453
+ "step": 349
2454
+ },
2455
+ {
2456
+ "epoch": 0.936768149882904,
2457
+ "grad_norm": 3.3887152671813965,
2458
+ "learning_rate": 6.115738727714593e-06,
2459
+ "loss": 0.4347,
2460
+ "step": 350
2461
+ },
2462
+ {
2463
+ "epoch": 0.9394446303111409,
2464
+ "grad_norm": 2.8967878818511963,
2465
+ "learning_rate": 6.093790708082861e-06,
2466
+ "loss": 0.4126,
2467
+ "step": 351
2468
+ },
2469
+ {
2470
+ "epoch": 0.9421211107393778,
2471
+ "grad_norm": 3.301456928253174,
2472
+ "learning_rate": 6.071820530867524e-06,
2473
+ "loss": 0.4932,
2474
+ "step": 352
2475
+ },
2476
+ {
2477
+ "epoch": 0.9447975911676146,
2478
+ "grad_norm": 2.467073678970337,
2479
+ "learning_rate": 6.0498286411318255e-06,
2480
+ "loss": 0.3714,
2481
+ "step": 353
2482
+ },
2483
+ {
2484
+ "epoch": 0.9474740715958515,
2485
+ "grad_norm": 2.737948417663574,
2486
+ "learning_rate": 6.027815484378848e-06,
2487
+ "loss": 0.3832,
2488
+ "step": 354
2489
+ },
2490
+ {
2491
+ "epoch": 0.9501505520240883,
2492
+ "grad_norm": 3.305079221725464,
2493
+ "learning_rate": 6.005781506542498e-06,
2494
+ "loss": 0.3813,
2495
+ "step": 355
2496
+ },
2497
+ {
2498
+ "epoch": 0.9528270324523251,
2499
+ "grad_norm": 3.1508371829986572,
2500
+ "learning_rate": 5.983727153978467e-06,
2501
+ "loss": 0.5005,
2502
+ "step": 356
2503
+ },
2504
+ {
2505
+ "epoch": 0.955503512880562,
2506
+ "grad_norm": 3.4938628673553467,
2507
+ "learning_rate": 5.961652873455186e-06,
2508
+ "loss": 0.4586,
2509
+ "step": 357
2510
+ },
2511
+ {
2512
+ "epoch": 0.9581799933087989,
2513
+ "grad_norm": 2.7031075954437256,
2514
+ "learning_rate": 5.939559112144781e-06,
2515
+ "loss": 0.3995,
2516
+ "step": 358
2517
+ },
2518
+ {
2519
+ "epoch": 0.9608564737370358,
2520
+ "grad_norm": 3.1249899864196777,
2521
+ "learning_rate": 5.917446317614012e-06,
2522
+ "loss": 0.4738,
2523
+ "step": 359
2524
+ },
2525
+ {
2526
+ "epoch": 0.9635329541652726,
2527
+ "grad_norm": 3.1881392002105713,
2528
+ "learning_rate": 5.895314937815206e-06,
2529
+ "loss": 0.4234,
2530
+ "step": 360
2531
+ },
2532
+ {
2533
+ "epoch": 0.9662094345935095,
2534
+ "grad_norm": 3.020379066467285,
2535
+ "learning_rate": 5.873165421077186e-06,
2536
+ "loss": 0.347,
2537
+ "step": 361
2538
+ },
2539
+ {
2540
+ "epoch": 0.9688859150217464,
2541
+ "grad_norm": 2.737119436264038,
2542
+ "learning_rate": 5.850998216096181e-06,
2543
+ "loss": 0.3805,
2544
+ "step": 362
2545
+ },
2546
+ {
2547
+ "epoch": 0.9715623954499832,
2548
+ "grad_norm": 3.5535435676574707,
2549
+ "learning_rate": 5.828813771926746e-06,
2550
+ "loss": 0.5056,
2551
+ "step": 363
2552
+ },
2553
+ {
2554
+ "epoch": 0.9742388758782201,
2555
+ "grad_norm": 3.3006703853607178,
2556
+ "learning_rate": 5.806612537972658e-06,
2557
+ "loss": 0.3342,
2558
+ "step": 364
2559
+ },
2560
+ {
2561
+ "epoch": 0.976915356306457,
2562
+ "grad_norm": 3.176393508911133,
2563
+ "learning_rate": 5.784394963977815e-06,
2564
+ "loss": 0.4834,
2565
+ "step": 365
2566
+ },
2567
+ {
2568
+ "epoch": 0.9795918367346939,
2569
+ "grad_norm": 3.3865625858306885,
2570
+ "learning_rate": 5.762161500017128e-06,
2571
+ "loss": 0.4573,
2572
+ "step": 366
2573
+ },
2574
+ {
2575
+ "epoch": 0.9822683171629307,
2576
+ "grad_norm": 2.942143678665161,
2577
+ "learning_rate": 5.739912596487396e-06,
2578
+ "loss": 0.4826,
2579
+ "step": 367
2580
+ },
2581
+ {
2582
+ "epoch": 0.9849447975911676,
2583
+ "grad_norm": 2.834310531616211,
2584
+ "learning_rate": 5.717648704098191e-06,
2585
+ "loss": 0.444,
2586
+ "step": 368
2587
+ },
2588
+ {
2589
+ "epoch": 0.9876212780194045,
2590
+ "grad_norm": 2.967898368835449,
2591
+ "learning_rate": 5.6953702738627215e-06,
2592
+ "loss": 0.3811,
2593
+ "step": 369
2594
+ },
2595
+ {
2596
+ "epoch": 0.9902977584476413,
2597
+ "grad_norm": 2.998699426651001,
2598
+ "learning_rate": 5.6730777570887e-06,
2599
+ "loss": 0.4694,
2600
+ "step": 370
2601
+ },
2602
+ {
2603
+ "epoch": 0.9929742388758782,
2604
+ "grad_norm": 3.34025502204895,
2605
+ "learning_rate": 5.6507716053691916e-06,
2606
+ "loss": 0.5136,
2607
+ "step": 371
2608
+ },
2609
+ {
2610
+ "epoch": 0.9956507193041151,
2611
+ "grad_norm": 2.5834507942199707,
2612
+ "learning_rate": 5.628452270573483e-06,
2613
+ "loss": 0.3702,
2614
+ "step": 372
2615
+ },
2616
+ {
2617
+ "epoch": 0.998327199732352,
2618
+ "grad_norm": 2.72577166557312,
2619
+ "learning_rate": 5.6061202048379125e-06,
2620
+ "loss": 0.3757,
2621
+ "step": 373
2622
+ },
2623
+ {
2624
+ "epoch": 1.0,
2625
+ "grad_norm": 3.7107036113739014,
2626
+ "learning_rate": 5.583775860556717e-06,
2627
+ "loss": 0.3578,
2628
+ "step": 374
2629
+ },
2630
+ {
2631
+ "epoch": 1.0026764804282369,
2632
+ "grad_norm": 2.568798542022705,
2633
+ "learning_rate": 5.561419690372869e-06,
2634
+ "loss": 0.1397,
2635
+ "step": 375
2636
+ },
2637
+ {
2638
+ "epoch": 1.0053529608564737,
2639
+ "grad_norm": 1.9347107410430908,
2640
+ "learning_rate": 5.539052147168903e-06,
2641
+ "loss": 0.2252,
2642
+ "step": 376
2643
+ },
2644
+ {
2645
+ "epoch": 1.0080294412847106,
2646
+ "grad_norm": 2.0152370929718018,
2647
+ "learning_rate": 5.516673684057747e-06,
2648
+ "loss": 0.2206,
2649
+ "step": 377
2650
+ },
2651
+ {
2652
+ "epoch": 1.0107059217129475,
2653
+ "grad_norm": 2.0257060527801514,
2654
+ "learning_rate": 5.494284754373538e-06,
2655
+ "loss": 0.1776,
2656
+ "step": 378
2657
+ },
2658
+ {
2659
+ "epoch": 1.0133824021411844,
2660
+ "grad_norm": 2.107743263244629,
2661
+ "learning_rate": 5.471885811662442e-06,
2662
+ "loss": 0.1799,
2663
+ "step": 379
2664
+ },
2665
+ {
2666
+ "epoch": 1.0160588825694212,
2667
+ "grad_norm": 2.221683979034424,
2668
+ "learning_rate": 5.449477309673462e-06,
2669
+ "loss": 0.1642,
2670
+ "step": 380
2671
+ },
2672
+ {
2673
+ "epoch": 1.018735362997658,
2674
+ "grad_norm": 2.8024303913116455,
2675
+ "learning_rate": 5.427059702349255e-06,
2676
+ "loss": 0.1712,
2677
+ "step": 381
2678
+ },
2679
+ {
2680
+ "epoch": 1.021411843425895,
2681
+ "grad_norm": 2.7197983264923096,
2682
+ "learning_rate": 5.4046334438169245e-06,
2683
+ "loss": 0.2057,
2684
+ "step": 382
2685
+ },
2686
+ {
2687
+ "epoch": 1.0240883238541318,
2688
+ "grad_norm": 1.90606689453125,
2689
+ "learning_rate": 5.382198988378829e-06,
2690
+ "loss": 0.1793,
2691
+ "step": 383
2692
+ },
2693
+ {
2694
+ "epoch": 1.0267648042823687,
2695
+ "grad_norm": 2.782726764678955,
2696
+ "learning_rate": 5.359756790503376e-06,
2697
+ "loss": 0.2036,
2698
+ "step": 384
2699
+ },
2700
+ {
2701
+ "epoch": 1.0294412847106056,
2702
+ "grad_norm": 2.3498716354370117,
2703
+ "learning_rate": 5.337307304815817e-06,
2704
+ "loss": 0.2134,
2705
+ "step": 385
2706
+ },
2707
+ {
2708
+ "epoch": 1.0321177651388425,
2709
+ "grad_norm": 3.6276447772979736,
2710
+ "learning_rate": 5.31485098608904e-06,
2711
+ "loss": 0.1908,
2712
+ "step": 386
2713
+ },
2714
+ {
2715
+ "epoch": 1.0347942455670793,
2716
+ "grad_norm": 3.2593178749084473,
2717
+ "learning_rate": 5.292388289234349e-06,
2718
+ "loss": 0.2326,
2719
+ "step": 387
2720
+ },
2721
+ {
2722
+ "epoch": 1.0374707259953162,
2723
+ "grad_norm": 3.1643850803375244,
2724
+ "learning_rate": 5.2699196692922546e-06,
2725
+ "loss": 0.2165,
2726
+ "step": 388
2727
+ },
2728
+ {
2729
+ "epoch": 1.040147206423553,
2730
+ "grad_norm": 2.73394513130188,
2731
+ "learning_rate": 5.247445581423257e-06,
2732
+ "loss": 0.1906,
2733
+ "step": 389
2734
+ },
2735
+ {
2736
+ "epoch": 1.04282368685179,
2737
+ "grad_norm": 3.0761959552764893,
2738
+ "learning_rate": 5.224966480898624e-06,
2739
+ "loss": 0.1903,
2740
+ "step": 390
2741
+ },
2742
+ {
2743
+ "epoch": 1.0455001672800268,
2744
+ "grad_norm": 3.4978442192077637,
2745
+ "learning_rate": 5.202482823091165e-06,
2746
+ "loss": 0.211,
2747
+ "step": 391
2748
+ },
2749
+ {
2750
+ "epoch": 1.0481766477082637,
2751
+ "grad_norm": 2.406022548675537,
2752
+ "learning_rate": 5.179995063466011e-06,
2753
+ "loss": 0.2095,
2754
+ "step": 392
2755
+ },
2756
+ {
2757
+ "epoch": 1.0508531281365006,
2758
+ "grad_norm": 3.4254419803619385,
2759
+ "learning_rate": 5.157503657571386e-06,
2760
+ "loss": 0.2339,
2761
+ "step": 393
2762
+ },
2763
+ {
2764
+ "epoch": 1.0535296085647374,
2765
+ "grad_norm": 3.1183996200561523,
2766
+ "learning_rate": 5.1350090610293765e-06,
2767
+ "loss": 0.1881,
2768
+ "step": 394
2769
+ },
2770
+ {
2771
+ "epoch": 1.0562060889929743,
2772
+ "grad_norm": 2.5185012817382812,
2773
+ "learning_rate": 5.112511729526708e-06,
2774
+ "loss": 0.1261,
2775
+ "step": 395
2776
+ },
2777
+ {
2778
+ "epoch": 1.0588825694212112,
2779
+ "grad_norm": 2.3276853561401367,
2780
+ "learning_rate": 5.090012118805505e-06,
2781
+ "loss": 0.1423,
2782
+ "step": 396
2783
+ },
2784
+ {
2785
+ "epoch": 1.061559049849448,
2786
+ "grad_norm": 2.8159985542297363,
2787
+ "learning_rate": 5.067510684654069e-06,
2788
+ "loss": 0.1786,
2789
+ "step": 397
2790
+ },
2791
+ {
2792
+ "epoch": 1.064235530277685,
2793
+ "grad_norm": 2.6511971950531006,
2794
+ "learning_rate": 5.0450078828976326e-06,
2795
+ "loss": 0.2184,
2796
+ "step": 398
2797
+ },
2798
+ {
2799
+ "epoch": 1.0669120107059218,
2800
+ "grad_norm": 2.8638663291931152,
2801
+ "learning_rate": 5.02250416938914e-06,
2802
+ "loss": 0.1745,
2803
+ "step": 399
2804
+ },
2805
+ {
2806
+ "epoch": 1.0695884911341587,
2807
+ "grad_norm": 2.6509993076324463,
2808
+ "learning_rate": 5e-06,
2809
+ "loss": 0.1678,
2810
+ "step": 400
2811
+ }
2812
+ ],
2813
+ "logging_steps": 1,
2814
+ "max_steps": 748,
2815
+ "num_input_tokens_seen": 0,
2816
+ "num_train_epochs": 2,
2817
+ "save_steps": 100,
2818
+ "stateful_callbacks": {
2819
+ "TrainerControl": {
2820
+ "args": {
2821
+ "should_epoch_stop": false,
2822
+ "should_evaluate": false,
2823
+ "should_log": false,
2824
+ "should_save": true,
2825
+ "should_training_stop": false
2826
+ },
2827
+ "attributes": {}
2828
+ }
2829
+ },
2830
+ "total_flos": 4.392429255582024e+17,
2831
+ "train_batch_size": 1,
2832
+ "trial_name": null,
2833
+ "trial_params": null
2834
+ }
checkpoint-500/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
+ }
checkpoint-500/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'Please reason step by step, and put your final answer within \\boxed{}.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nPlease reason step by step, and put your final answer within \\boxed{}.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
checkpoint-500/config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen2ForCausalLM"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "dtype": "float32",
7
+ "eos_token_id": 151645,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 3584,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 18944,
12
+ "layer_types": [
13
+ "full_attention",
14
+ "full_attention",
15
+ "full_attention",
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention"
41
+ ],
42
+ "max_position_embeddings": 4096,
43
+ "max_window_layers": 28,
44
+ "model_type": "qwen2",
45
+ "num_attention_heads": 28,
46
+ "num_hidden_layers": 28,
47
+ "num_key_value_heads": 4,
48
+ "pad_token_id": 151643,
49
+ "rms_norm_eps": 1e-06,
50
+ "rope_scaling": null,
51
+ "rope_theta": 10000,
52
+ "sliding_window": null,
53
+ "tie_word_embeddings": false,
54
+ "transformers_version": "4.57.1",
55
+ "use_cache": false,
56
+ "use_mrope": false,
57
+ "use_sliding_window": false,
58
+ "vocab_size": 152064
59
+ }
checkpoint-500/generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "eos_token_id": [
3
+ 151645,
4
+ 151643
5
+ ],
6
+ "max_new_tokens": 2048,
7
+ "pad_token_id": 151643,
8
+ "transformers_version": "4.57.1"
9
+ }
checkpoint-500/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-500/model.safetensors.index.json ADDED
@@ -0,0 +1,347 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_parameters": 7615616512,
4
+ "total_size": 30462466048
5
+ },
6
+ "weight_map": {
7
+ "lm_head.weight": "model-00007-of-00007.safetensors",
8
+ "model.embed_tokens.weight": "model-00001-of-00007.safetensors",
9
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00007.safetensors",
10
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
11
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
12
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
13
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
14
+ "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
15
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
16
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
17
+ "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
18
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
19
+ "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
20
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
21
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00007.safetensors",
22
+ "model.layers.1.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
23
+ "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
24
+ "model.layers.1.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
25
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
26
+ "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
27
+ "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
28
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
29
+ "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
30
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
31
+ "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
32
+ "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
33
+ "model.layers.10.input_layernorm.weight": "model-00003-of-00007.safetensors",
34
+ "model.layers.10.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
35
+ "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
36
+ "model.layers.10.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
37
+ "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
38
+ "model.layers.10.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
39
+ "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
40
+ "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
41
+ "model.layers.10.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
42
+ "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
43
+ "model.layers.10.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
44
+ "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
45
+ "model.layers.11.input_layernorm.weight": "model-00003-of-00007.safetensors",
46
+ "model.layers.11.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
47
+ "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
48
+ "model.layers.11.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
49
+ "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
50
+ "model.layers.11.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
51
+ "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
52
+ "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
53
+ "model.layers.11.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
54
+ "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
55
+ "model.layers.11.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
56
+ "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
57
+ "model.layers.12.input_layernorm.weight": "model-00003-of-00007.safetensors",
58
+ "model.layers.12.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
59
+ "model.layers.12.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
60
+ "model.layers.12.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
61
+ "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
62
+ "model.layers.12.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
63
+ "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
64
+ "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
65
+ "model.layers.12.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
66
+ "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
67
+ "model.layers.12.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
68
+ "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
69
+ "model.layers.13.input_layernorm.weight": "model-00004-of-00007.safetensors",
70
+ "model.layers.13.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
71
+ "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
72
+ "model.layers.13.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
73
+ "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
74
+ "model.layers.13.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
75
+ "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
76
+ "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
77
+ "model.layers.13.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
78
+ "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
79
+ "model.layers.13.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
80
+ "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
81
+ "model.layers.14.input_layernorm.weight": "model-00004-of-00007.safetensors",
82
+ "model.layers.14.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
83
+ "model.layers.14.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
84
+ "model.layers.14.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
85
+ "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
86
+ "model.layers.14.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
87
+ "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
88
+ "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
89
+ "model.layers.14.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
90
+ "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
91
+ "model.layers.14.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
92
+ "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
93
+ "model.layers.15.input_layernorm.weight": "model-00004-of-00007.safetensors",
94
+ "model.layers.15.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
95
+ "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
96
+ "model.layers.15.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
97
+ "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
98
+ "model.layers.15.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
99
+ "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
100
+ "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
101
+ "model.layers.15.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
102
+ "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
103
+ "model.layers.15.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
104
+ "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
105
+ "model.layers.16.input_layernorm.weight": "model-00004-of-00007.safetensors",
106
+ "model.layers.16.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
107
+ "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
108
+ "model.layers.16.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
109
+ "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
110
+ "model.layers.16.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
111
+ "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
112
+ "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
113
+ "model.layers.16.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
114
+ "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
115
+ "model.layers.16.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
116
+ "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
117
+ "model.layers.17.input_layernorm.weight": "model-00004-of-00007.safetensors",
118
+ "model.layers.17.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
119
+ "model.layers.17.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
120
+ "model.layers.17.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
121
+ "model.layers.17.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
122
+ "model.layers.17.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
123
+ "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
124
+ "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
125
+ "model.layers.17.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
126
+ "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
127
+ "model.layers.17.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
128
+ "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
129
+ "model.layers.18.input_layernorm.weight": "model-00005-of-00007.safetensors",
130
+ "model.layers.18.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
131
+ "model.layers.18.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
132
+ "model.layers.18.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
133
+ "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
134
+ "model.layers.18.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
135
+ "model.layers.18.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
136
+ "model.layers.18.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
137
+ "model.layers.18.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
138
+ "model.layers.18.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
139
+ "model.layers.18.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
140
+ "model.layers.18.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
141
+ "model.layers.19.input_layernorm.weight": "model-00005-of-00007.safetensors",
142
+ "model.layers.19.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
143
+ "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
144
+ "model.layers.19.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
145
+ "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
146
+ "model.layers.19.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
147
+ "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
148
+ "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
149
+ "model.layers.19.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
150
+ "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
151
+ "model.layers.19.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
152
+ "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
153
+ "model.layers.2.input_layernorm.weight": "model-00001-of-00007.safetensors",
154
+ "model.layers.2.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
155
+ "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
156
+ "model.layers.2.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
157
+ "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
158
+ "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
159
+ "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
160
+ "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
161
+ "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
162
+ "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
163
+ "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
164
+ "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
165
+ "model.layers.20.input_layernorm.weight": "model-00005-of-00007.safetensors",
166
+ "model.layers.20.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
167
+ "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
168
+ "model.layers.20.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
169
+ "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
170
+ "model.layers.20.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
171
+ "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
172
+ "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
173
+ "model.layers.20.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
174
+ "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
175
+ "model.layers.20.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
176
+ "model.layers.20.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
177
+ "model.layers.21.input_layernorm.weight": "model-00005-of-00007.safetensors",
178
+ "model.layers.21.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
179
+ "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
180
+ "model.layers.21.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
181
+ "model.layers.21.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
182
+ "model.layers.21.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
183
+ "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
184
+ "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
185
+ "model.layers.21.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
186
+ "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
187
+ "model.layers.21.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
188
+ "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
189
+ "model.layers.22.input_layernorm.weight": "model-00005-of-00007.safetensors",
190
+ "model.layers.22.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
191
+ "model.layers.22.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
192
+ "model.layers.22.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
193
+ "model.layers.22.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
194
+ "model.layers.22.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
195
+ "model.layers.22.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
196
+ "model.layers.22.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
197
+ "model.layers.22.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
198
+ "model.layers.22.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
199
+ "model.layers.22.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
200
+ "model.layers.22.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
201
+ "model.layers.23.input_layernorm.weight": "model-00005-of-00007.safetensors",
202
+ "model.layers.23.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
203
+ "model.layers.23.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
204
+ "model.layers.23.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
205
+ "model.layers.23.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
206
+ "model.layers.23.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
207
+ "model.layers.23.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
208
+ "model.layers.23.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
209
+ "model.layers.23.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
210
+ "model.layers.23.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
211
+ "model.layers.23.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
212
+ "model.layers.23.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
213
+ "model.layers.24.input_layernorm.weight": "model-00006-of-00007.safetensors",
214
+ "model.layers.24.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
215
+ "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
216
+ "model.layers.24.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
217
+ "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
218
+ "model.layers.24.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
219
+ "model.layers.24.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
220
+ "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
221
+ "model.layers.24.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
222
+ "model.layers.24.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
223
+ "model.layers.24.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
224
+ "model.layers.24.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
225
+ "model.layers.25.input_layernorm.weight": "model-00006-of-00007.safetensors",
226
+ "model.layers.25.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
227
+ "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
228
+ "model.layers.25.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
229
+ "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
230
+ "model.layers.25.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
231
+ "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
232
+ "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
233
+ "model.layers.25.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
234
+ "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
235
+ "model.layers.25.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
236
+ "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
237
+ "model.layers.26.input_layernorm.weight": "model-00006-of-00007.safetensors",
238
+ "model.layers.26.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
239
+ "model.layers.26.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
240
+ "model.layers.26.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
241
+ "model.layers.26.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
242
+ "model.layers.26.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
243
+ "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
244
+ "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
245
+ "model.layers.26.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
246
+ "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
247
+ "model.layers.26.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
248
+ "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
249
+ "model.layers.27.input_layernorm.weight": "model-00006-of-00007.safetensors",
250
+ "model.layers.27.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
251
+ "model.layers.27.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
252
+ "model.layers.27.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
253
+ "model.layers.27.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
254
+ "model.layers.27.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
255
+ "model.layers.27.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
256
+ "model.layers.27.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
257
+ "model.layers.27.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
258
+ "model.layers.27.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
259
+ "model.layers.27.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
260
+ "model.layers.27.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
261
+ "model.layers.3.input_layernorm.weight": "model-00002-of-00007.safetensors",
262
+ "model.layers.3.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
263
+ "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
264
+ "model.layers.3.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
265
+ "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
266
+ "model.layers.3.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
267
+ "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
268
+ "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
269
+ "model.layers.3.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
270
+ "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
271
+ "model.layers.3.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
272
+ "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
273
+ "model.layers.4.input_layernorm.weight": "model-00002-of-00007.safetensors",
274
+ "model.layers.4.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
275
+ "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
276
+ "model.layers.4.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
277
+ "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
278
+ "model.layers.4.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
279
+ "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
280
+ "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
281
+ "model.layers.4.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
282
+ "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
283
+ "model.layers.4.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
284
+ "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
285
+ "model.layers.5.input_layernorm.weight": "model-00002-of-00007.safetensors",
286
+ "model.layers.5.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
287
+ "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
288
+ "model.layers.5.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
289
+ "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
290
+ "model.layers.5.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
291
+ "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
292
+ "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
293
+ "model.layers.5.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
294
+ "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
295
+ "model.layers.5.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
296
+ "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
297
+ "model.layers.6.input_layernorm.weight": "model-00002-of-00007.safetensors",
298
+ "model.layers.6.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
299
+ "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
300
+ "model.layers.6.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
301
+ "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
302
+ "model.layers.6.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
303
+ "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
304
+ "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
305
+ "model.layers.6.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
306
+ "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
307
+ "model.layers.6.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
308
+ "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
309
+ "model.layers.7.input_layernorm.weight": "model-00002-of-00007.safetensors",
310
+ "model.layers.7.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
311
+ "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
312
+ "model.layers.7.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
313
+ "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
314
+ "model.layers.7.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
315
+ "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
316
+ "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
317
+ "model.layers.7.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
318
+ "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
319
+ "model.layers.7.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
320
+ "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
321
+ "model.layers.8.input_layernorm.weight": "model-00003-of-00007.safetensors",
322
+ "model.layers.8.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
323
+ "model.layers.8.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
324
+ "model.layers.8.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
325
+ "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
326
+ "model.layers.8.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
327
+ "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
328
+ "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
329
+ "model.layers.8.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
330
+ "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
331
+ "model.layers.8.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
332
+ "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
333
+ "model.layers.9.input_layernorm.weight": "model-00003-of-00007.safetensors",
334
+ "model.layers.9.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
335
+ "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
336
+ "model.layers.9.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
337
+ "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
338
+ "model.layers.9.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
339
+ "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
340
+ "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
341
+ "model.layers.9.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
342
+ "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
343
+ "model.layers.9.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
344
+ "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
345
+ "model.norm.weight": "model-00006-of-00007.safetensors"
346
+ }
347
+ }
checkpoint-500/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
+ }
checkpoint-500/tokenizer_config.json ADDED
@@ -0,0 +1,209 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": false,
5
+ "added_tokens_decoder": {
6
+ "151643": {
7
+ "content": "<|endoftext|>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "151644": {
15
+ "content": "<|im_start|>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "151645": {
23
+ "content": "<|im_end|>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "151646": {
31
+ "content": "<|object_ref_start|>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "151647": {
39
+ "content": "<|object_ref_end|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "151648": {
47
+ "content": "<|box_start|>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "151649": {
55
+ "content": "<|box_end|>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "151650": {
63
+ "content": "<|quad_start|>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "151651": {
71
+ "content": "<|quad_end|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "151652": {
79
+ "content": "<|vision_start|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "151653": {
87
+ "content": "<|vision_end|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "151654": {
95
+ "content": "<|vision_pad|>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": true
101
+ },
102
+ "151655": {
103
+ "content": "<|image_pad|>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": true
109
+ },
110
+ "151656": {
111
+ "content": "<|video_pad|>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": true
117
+ },
118
+ "151657": {
119
+ "content": "<tool_call>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": false
125
+ },
126
+ "151658": {
127
+ "content": "</tool_call>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": false
133
+ },
134
+ "151659": {
135
+ "content": "<|fim_prefix|>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": false
141
+ },
142
+ "151660": {
143
+ "content": "<|fim_middle|>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": false
149
+ },
150
+ "151661": {
151
+ "content": "<|fim_suffix|>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": false
157
+ },
158
+ "151662": {
159
+ "content": "<|fim_pad|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": false
165
+ },
166
+ "151663": {
167
+ "content": "<|repo_name|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": false
173
+ },
174
+ "151664": {
175
+ "content": "<|file_sep|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": false
181
+ }
182
+ },
183
+ "additional_special_tokens": [
184
+ "<|im_start|>",
185
+ "<|im_end|>",
186
+ "<|object_ref_start|>",
187
+ "<|object_ref_end|>",
188
+ "<|box_start|>",
189
+ "<|box_end|>",
190
+ "<|quad_start|>",
191
+ "<|quad_end|>",
192
+ "<|vision_start|>",
193
+ "<|vision_end|>",
194
+ "<|vision_pad|>",
195
+ "<|image_pad|>",
196
+ "<|video_pad|>"
197
+ ],
198
+ "bos_token": null,
199
+ "clean_up_tokenization_spaces": false,
200
+ "eos_token": "<|im_end|>",
201
+ "errors": "replace",
202
+ "extra_special_tokens": {},
203
+ "model_max_length": 131072,
204
+ "pad_token": "<|endoftext|>",
205
+ "padding_side": "right",
206
+ "split_special_tokens": false,
207
+ "tokenizer_class": "Qwen2Tokenizer",
208
+ "unk_token": null
209
+ }
checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,3534 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.3372365339578454,
6
+ "eval_steps": 500,
7
+ "global_step": 500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0026764804282368685,
14
+ "grad_norm": 416.7599792480469,
15
+ "learning_rate": 0.0,
16
+ "loss": 3.2698,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.005352960856473737,
21
+ "grad_norm": 316.08477783203125,
22
+ "learning_rate": 2.0000000000000002e-07,
23
+ "loss": 2.9631,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.008029441284710606,
28
+ "grad_norm": 472.76885986328125,
29
+ "learning_rate": 4.0000000000000003e-07,
30
+ "loss": 3.9863,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.010705921712947474,
35
+ "grad_norm": 251.41015625,
36
+ "learning_rate": 6.000000000000001e-07,
37
+ "loss": 3.2654,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.013382402141184342,
42
+ "grad_norm": 217.40838623046875,
43
+ "learning_rate": 8.000000000000001e-07,
44
+ "loss": 3.4505,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.01605888256942121,
49
+ "grad_norm": 136.2932586669922,
50
+ "learning_rate": 1.0000000000000002e-06,
51
+ "loss": 3.2498,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.01873536299765808,
56
+ "grad_norm": 135.7803192138672,
57
+ "learning_rate": 1.2000000000000002e-06,
58
+ "loss": 3.0983,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.021411843425894948,
63
+ "grad_norm": 93.63102722167969,
64
+ "learning_rate": 1.4000000000000001e-06,
65
+ "loss": 2.965,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.024088323854131816,
70
+ "grad_norm": 82.7247085571289,
71
+ "learning_rate": 1.6000000000000001e-06,
72
+ "loss": 2.6341,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.026764804282368684,
77
+ "grad_norm": 94.11923217773438,
78
+ "learning_rate": 1.8000000000000001e-06,
79
+ "loss": 3.2072,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.029441284710605555,
84
+ "grad_norm": 65.76191711425781,
85
+ "learning_rate": 2.0000000000000003e-06,
86
+ "loss": 2.6866,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.03211776513884242,
91
+ "grad_norm": 89.41738891601562,
92
+ "learning_rate": 2.2e-06,
93
+ "loss": 2.8421,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.03479424556707929,
98
+ "grad_norm": 82.1783676147461,
99
+ "learning_rate": 2.4000000000000003e-06,
100
+ "loss": 2.5489,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.03747072599531616,
105
+ "grad_norm": 72.41712951660156,
106
+ "learning_rate": 2.6e-06,
107
+ "loss": 2.9365,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.04014720642355303,
112
+ "grad_norm": 77.541015625,
113
+ "learning_rate": 2.8000000000000003e-06,
114
+ "loss": 2.0516,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.042823686851789895,
119
+ "grad_norm": 125.65862274169922,
120
+ "learning_rate": 3e-06,
121
+ "loss": 2.7676,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.04550016728002677,
126
+ "grad_norm": 94.85375213623047,
127
+ "learning_rate": 3.2000000000000003e-06,
128
+ "loss": 2.5114,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.04817664770826363,
133
+ "grad_norm": 77.89604187011719,
134
+ "learning_rate": 3.4000000000000005e-06,
135
+ "loss": 2.536,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.0508531281365005,
140
+ "grad_norm": 59.268680572509766,
141
+ "learning_rate": 3.6000000000000003e-06,
142
+ "loss": 2.1877,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.05352960856473737,
147
+ "grad_norm": 57.32849884033203,
148
+ "learning_rate": 3.8000000000000005e-06,
149
+ "loss": 2.2084,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.05620608899297424,
154
+ "grad_norm": 58.27803421020508,
155
+ "learning_rate": 4.000000000000001e-06,
156
+ "loss": 2.191,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.05888256942121111,
161
+ "grad_norm": 32.10304260253906,
162
+ "learning_rate": 4.2000000000000004e-06,
163
+ "loss": 2.0009,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.061559049849447975,
168
+ "grad_norm": 17.106046676635742,
169
+ "learning_rate": 4.4e-06,
170
+ "loss": 1.7496,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.06423553027768485,
175
+ "grad_norm": 16.371326446533203,
176
+ "learning_rate": 4.600000000000001e-06,
177
+ "loss": 2.0558,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.06691201070592172,
182
+ "grad_norm": 14.423291206359863,
183
+ "learning_rate": 4.800000000000001e-06,
184
+ "loss": 2.1511,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.06958849113415858,
189
+ "grad_norm": 16.198841094970703,
190
+ "learning_rate": 5e-06,
191
+ "loss": 1.9571,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.07226497156239545,
196
+ "grad_norm": 14.810019493103027,
197
+ "learning_rate": 5.2e-06,
198
+ "loss": 1.9937,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.07494145199063232,
203
+ "grad_norm": 13.464497566223145,
204
+ "learning_rate": 5.400000000000001e-06,
205
+ "loss": 1.7204,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.07761793241886919,
210
+ "grad_norm": 14.364913940429688,
211
+ "learning_rate": 5.600000000000001e-06,
212
+ "loss": 1.7,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.08029441284710606,
217
+ "grad_norm": 12.647878646850586,
218
+ "learning_rate": 5.8e-06,
219
+ "loss": 1.7034,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.08297089327534292,
224
+ "grad_norm": 12.305310249328613,
225
+ "learning_rate": 6e-06,
226
+ "loss": 1.7873,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.08564737370357979,
231
+ "grad_norm": 9.316176414489746,
232
+ "learning_rate": 6.200000000000001e-06,
233
+ "loss": 1.7204,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.08832385413181666,
238
+ "grad_norm": 10.51429271697998,
239
+ "learning_rate": 6.4000000000000006e-06,
240
+ "loss": 1.401,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.09100033456005353,
245
+ "grad_norm": 10.570704460144043,
246
+ "learning_rate": 6.600000000000001e-06,
247
+ "loss": 1.6091,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.0936768149882904,
252
+ "grad_norm": 12.805994987487793,
253
+ "learning_rate": 6.800000000000001e-06,
254
+ "loss": 1.5669,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.09635329541652726,
259
+ "grad_norm": 10.955977439880371,
260
+ "learning_rate": 7e-06,
261
+ "loss": 1.4644,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.09902977584476413,
266
+ "grad_norm": 15.71002197265625,
267
+ "learning_rate": 7.2000000000000005e-06,
268
+ "loss": 1.5497,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.101706256273001,
273
+ "grad_norm": 11.375880241394043,
274
+ "learning_rate": 7.4e-06,
275
+ "loss": 1.5646,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.10438273670123788,
280
+ "grad_norm": 12.987727165222168,
281
+ "learning_rate": 7.600000000000001e-06,
282
+ "loss": 1.4149,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.10705921712947473,
287
+ "grad_norm": 10.491798400878906,
288
+ "learning_rate": 7.800000000000002e-06,
289
+ "loss": 1.5017,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.1097356975577116,
294
+ "grad_norm": 12.43316650390625,
295
+ "learning_rate": 8.000000000000001e-06,
296
+ "loss": 1.4759,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.11241217798594848,
301
+ "grad_norm": 12.077725410461426,
302
+ "learning_rate": 8.2e-06,
303
+ "loss": 1.4892,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.11508865841418535,
308
+ "grad_norm": 12.642361640930176,
309
+ "learning_rate": 8.400000000000001e-06,
310
+ "loss": 1.458,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.11776513884242222,
315
+ "grad_norm": 15.535161018371582,
316
+ "learning_rate": 8.6e-06,
317
+ "loss": 1.6861,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.12044161927065908,
322
+ "grad_norm": 11.678996086120605,
323
+ "learning_rate": 8.8e-06,
324
+ "loss": 1.6968,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.12311809969889595,
329
+ "grad_norm": 11.906571388244629,
330
+ "learning_rate": 9e-06,
331
+ "loss": 1.44,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.12579458012713282,
336
+ "grad_norm": 11.492127418518066,
337
+ "learning_rate": 9.200000000000002e-06,
338
+ "loss": 1.3714,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.1284710605553697,
343
+ "grad_norm": 15.786943435668945,
344
+ "learning_rate": 9.4e-06,
345
+ "loss": 1.4132,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.13114754098360656,
350
+ "grad_norm": 12.31556224822998,
351
+ "learning_rate": 9.600000000000001e-06,
352
+ "loss": 1.166,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.13382402141184344,
357
+ "grad_norm": 12.496630668640137,
358
+ "learning_rate": 9.800000000000001e-06,
359
+ "loss": 1.2859,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.1365005018400803,
364
+ "grad_norm": 11.003588676452637,
365
+ "learning_rate": 1e-05,
366
+ "loss": 1.2614,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.13917698226831715,
371
+ "grad_norm": 12.946131706237793,
372
+ "learning_rate": 9.99994935597953e-06,
373
+ "loss": 1.4555,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.14185346269655402,
378
+ "grad_norm": 13.676822662353516,
379
+ "learning_rate": 9.999797424944041e-06,
380
+ "loss": 1.2167,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.1445299431247909,
385
+ "grad_norm": 14.827473640441895,
386
+ "learning_rate": 9.999544209971299e-06,
387
+ "loss": 1.3456,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.14720642355302777,
392
+ "grad_norm": 11.921989440917969,
393
+ "learning_rate": 9.99918971619083e-06,
394
+ "loss": 1.2535,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.14988290398126464,
399
+ "grad_norm": 12.726682662963867,
400
+ "learning_rate": 9.99873395078383e-06,
401
+ "loss": 1.2172,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.1525593844095015,
406
+ "grad_norm": 18.52472496032715,
407
+ "learning_rate": 9.998176922983017e-06,
408
+ "loss": 1.4418,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.15523586483773838,
413
+ "grad_norm": 13.32449722290039,
414
+ "learning_rate": 9.99751864407244e-06,
415
+ "loss": 1.18,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.15791234526597525,
420
+ "grad_norm": 10.778595924377441,
421
+ "learning_rate": 9.996759127387259e-06,
422
+ "loss": 1.1879,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.16058882569421212,
427
+ "grad_norm": 10.155744552612305,
428
+ "learning_rate": 9.99589838831346e-06,
429
+ "loss": 1.1274,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.16326530612244897,
434
+ "grad_norm": 13.751497268676758,
435
+ "learning_rate": 9.994936444287565e-06,
436
+ "loss": 1.1499,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.16594178655068584,
441
+ "grad_norm": 17.752635955810547,
442
+ "learning_rate": 9.993873314796253e-06,
443
+ "loss": 1.1425,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.1686182669789227,
448
+ "grad_norm": 13.581279754638672,
449
+ "learning_rate": 9.992709021375987e-06,
450
+ "loss": 1.0341,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.17129474740715958,
455
+ "grad_norm": 11.867856979370117,
456
+ "learning_rate": 9.991443587612568e-06,
457
+ "loss": 1.024,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.17397122783539645,
462
+ "grad_norm": 13.20876693725586,
463
+ "learning_rate": 9.990077039140655e-06,
464
+ "loss": 1.0375,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.17664770826363332,
469
+ "grad_norm": 12.211270332336426,
470
+ "learning_rate": 9.988609403643254e-06,
471
+ "loss": 1.1334,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.1793241886918702,
476
+ "grad_norm": 11.89450454711914,
477
+ "learning_rate": 9.987040710851148e-06,
478
+ "loss": 1.0447,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.18200066912010707,
483
+ "grad_norm": 10.132635116577148,
484
+ "learning_rate": 9.9853709925423e-06,
485
+ "loss": 0.7626,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.18467714954834394,
490
+ "grad_norm": 9.661967277526855,
491
+ "learning_rate": 9.983600282541213e-06,
492
+ "loss": 0.968,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.1873536299765808,
497
+ "grad_norm": 11.19580078125,
498
+ "learning_rate": 9.981728616718234e-06,
499
+ "loss": 0.9161,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.19003011040481765,
504
+ "grad_norm": 9.165005683898926,
505
+ "learning_rate": 9.979756032988837e-06,
506
+ "loss": 0.8821,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.19270659083305453,
511
+ "grad_norm": 9.17795181274414,
512
+ "learning_rate": 9.977682571312847e-06,
513
+ "loss": 1.0452,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.1953830712612914,
518
+ "grad_norm": 7.483065128326416,
519
+ "learning_rate": 9.975508273693643e-06,
520
+ "loss": 1.0135,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.19805955168952827,
525
+ "grad_norm": 7.2489728927612305,
526
+ "learning_rate": 9.97323318417729e-06,
527
+ "loss": 0.9675,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.20073603211776514,
532
+ "grad_norm": 7.151744365692139,
533
+ "learning_rate": 9.970857348851667e-06,
534
+ "loss": 0.829,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.203412512546002,
539
+ "grad_norm": 7.535038948059082,
540
+ "learning_rate": 9.968380815845504e-06,
541
+ "loss": 0.9383,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.20608899297423888,
546
+ "grad_norm": 5.609306812286377,
547
+ "learning_rate": 9.965803635327445e-06,
548
+ "loss": 0.8914,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.20876547340247575,
553
+ "grad_norm": 7.203204154968262,
554
+ "learning_rate": 9.963125859505e-06,
555
+ "loss": 1.0889,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.21144195383071263,
560
+ "grad_norm": 4.907174587249756,
561
+ "learning_rate": 9.960347542623506e-06,
562
+ "loss": 0.7694,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.21411843425894947,
567
+ "grad_norm": 5.539646148681641,
568
+ "learning_rate": 9.95746874096501e-06,
569
+ "loss": 0.9965,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.21679491468718634,
574
+ "grad_norm": 5.0046868324279785,
575
+ "learning_rate": 9.954489512847156e-06,
576
+ "loss": 0.7795,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.2194713951154232,
581
+ "grad_norm": 5.138219833374023,
582
+ "learning_rate": 9.951409918621977e-06,
583
+ "loss": 0.8877,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.22214787554366008,
588
+ "grad_norm": 4.388546466827393,
589
+ "learning_rate": 9.948230020674685e-06,
590
+ "loss": 0.9253,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.22482435597189696,
595
+ "grad_norm": 4.456934928894043,
596
+ "learning_rate": 9.944949883422409e-06,
597
+ "loss": 0.7335,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.22750083640013383,
602
+ "grad_norm": 4.645761013031006,
603
+ "learning_rate": 9.941569573312882e-06,
604
+ "loss": 0.8997,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.2301773168283707,
609
+ "grad_norm": 4.329288482666016,
610
+ "learning_rate": 9.938089158823101e-06,
611
+ "loss": 0.852,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.23285379725660757,
616
+ "grad_norm": 5.044316291809082,
617
+ "learning_rate": 9.934508710457944e-06,
618
+ "loss": 0.8236,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.23553027768484444,
623
+ "grad_norm": 5.356631755828857,
624
+ "learning_rate": 9.930828300748726e-06,
625
+ "loss": 0.9572,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.23820675811308128,
630
+ "grad_norm": 3.933661699295044,
631
+ "learning_rate": 9.927048004251748e-06,
632
+ "loss": 0.8479,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.24088323854131816,
637
+ "grad_norm": 5.228504180908203,
638
+ "learning_rate": 9.923167897546773e-06,
639
+ "loss": 0.741,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.24355971896955503,
644
+ "grad_norm": 4.520639419555664,
645
+ "learning_rate": 9.919188059235483e-06,
646
+ "loss": 0.746,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.2462361993977919,
651
+ "grad_norm": 4.191668510437012,
652
+ "learning_rate": 9.915108569939884e-06,
653
+ "loss": 0.6904,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.24891267982602877,
658
+ "grad_norm": 4.300325393676758,
659
+ "learning_rate": 9.910929512300673e-06,
660
+ "loss": 0.8753,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.25158916025426564,
665
+ "grad_norm": 5.979667663574219,
666
+ "learning_rate": 9.90665097097556e-06,
667
+ "loss": 0.7355,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.2542656406825025,
672
+ "grad_norm": 5.374980449676514,
673
+ "learning_rate": 9.902273032637558e-06,
674
+ "loss": 0.7932,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.2569421211107394,
679
+ "grad_norm": 3.9075119495391846,
680
+ "learning_rate": 9.897795785973227e-06,
681
+ "loss": 0.9069,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.25961860153897626,
686
+ "grad_norm": 4.439886569976807,
687
+ "learning_rate": 9.89321932168088e-06,
688
+ "loss": 0.8408,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.26229508196721313,
693
+ "grad_norm": 4.403755187988281,
694
+ "learning_rate": 9.888543732468732e-06,
695
+ "loss": 0.709,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.26497156239545,
700
+ "grad_norm": 5.023388862609863,
701
+ "learning_rate": 9.883769113053039e-06,
702
+ "loss": 0.8928,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.26764804282368687,
707
+ "grad_norm": 5.586976528167725,
708
+ "learning_rate": 9.878895560156172e-06,
709
+ "loss": 0.9644,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 0.27032452325192374,
714
+ "grad_norm": 15.905163764953613,
715
+ "learning_rate": 9.873923172504653e-06,
716
+ "loss": 0.8362,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 0.2730010036801606,
721
+ "grad_norm": 5.6663432121276855,
722
+ "learning_rate": 9.868852050827167e-06,
723
+ "loss": 0.7585,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 0.27567748410839743,
728
+ "grad_norm": 4.5769124031066895,
729
+ "learning_rate": 9.863682297852506e-06,
730
+ "loss": 0.9748,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 0.2783539645366343,
735
+ "grad_norm": 4.305916786193848,
736
+ "learning_rate": 9.858414018307503e-06,
737
+ "loss": 0.805,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 0.2810304449648712,
742
+ "grad_norm": 4.756903648376465,
743
+ "learning_rate": 9.853047318914898e-06,
744
+ "loss": 0.8221,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 0.28370692539310804,
749
+ "grad_norm": 5.709683418273926,
750
+ "learning_rate": 9.847582308391189e-06,
751
+ "loss": 0.7416,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 0.2863834058213449,
756
+ "grad_norm": 3.933176279067993,
757
+ "learning_rate": 9.842019097444414e-06,
758
+ "loss": 0.7298,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 0.2890598862495818,
763
+ "grad_norm": 4.9454264640808105,
764
+ "learning_rate": 9.836357798771922e-06,
765
+ "loss": 0.7683,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 0.29173636667781866,
770
+ "grad_norm": 4.274379730224609,
771
+ "learning_rate": 9.830598527058083e-06,
772
+ "loss": 0.7788,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 0.29441284710605553,
777
+ "grad_norm": 3.6148972511291504,
778
+ "learning_rate": 9.824741398971966e-06,
779
+ "loss": 0.6941,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 0.2970893275342924,
784
+ "grad_norm": 3.8976197242736816,
785
+ "learning_rate": 9.81878653316498e-06,
786
+ "loss": 0.8783,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 0.2997658079625293,
791
+ "grad_norm": 4.712560176849365,
792
+ "learning_rate": 9.81273405026846e-06,
793
+ "loss": 0.9022,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 0.30244228839076615,
798
+ "grad_norm": 5.358889579772949,
799
+ "learning_rate": 9.806584072891234e-06,
800
+ "loss": 0.8587,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 0.305118768819003,
805
+ "grad_norm": 4.296327114105225,
806
+ "learning_rate": 9.800336725617136e-06,
807
+ "loss": 0.8813,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 0.3077952492472399,
812
+ "grad_norm": 4.895114421844482,
813
+ "learning_rate": 9.793992135002476e-06,
814
+ "loss": 0.8184,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 0.31047172967547676,
819
+ "grad_norm": 4.782822132110596,
820
+ "learning_rate": 9.787550429573487e-06,
821
+ "loss": 0.6049,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 0.31314821010371363,
826
+ "grad_norm": 5.41484260559082,
827
+ "learning_rate": 9.781011739823715e-06,
828
+ "loss": 0.7804,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 0.3158246905319505,
833
+ "grad_norm": 3.639780044555664,
834
+ "learning_rate": 9.77437619821137e-06,
835
+ "loss": 0.7241,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 0.3185011709601874,
840
+ "grad_norm": 4.5406880378723145,
841
+ "learning_rate": 9.767643939156658e-06,
842
+ "loss": 0.8398,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 0.32117765138842425,
847
+ "grad_norm": 4.575490951538086,
848
+ "learning_rate": 9.760815099039045e-06,
849
+ "loss": 0.6951,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 0.3238541318166611,
854
+ "grad_norm": 3.868921995162964,
855
+ "learning_rate": 9.753889816194498e-06,
856
+ "loss": 0.706,
857
+ "step": 121
858
+ },
859
+ {
860
+ "epoch": 0.32653061224489793,
861
+ "grad_norm": 4.064291477203369,
862
+ "learning_rate": 9.746868230912683e-06,
863
+ "loss": 0.7386,
864
+ "step": 122
865
+ },
866
+ {
867
+ "epoch": 0.3292070926731348,
868
+ "grad_norm": 4.059055805206299,
869
+ "learning_rate": 9.739750485434126e-06,
870
+ "loss": 0.7171,
871
+ "step": 123
872
+ },
873
+ {
874
+ "epoch": 0.3318835731013717,
875
+ "grad_norm": 4.522129535675049,
876
+ "learning_rate": 9.73253672394732e-06,
877
+ "loss": 0.6679,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 0.33456005352960855,
882
+ "grad_norm": 3.669271945953369,
883
+ "learning_rate": 9.725227092585824e-06,
884
+ "loss": 0.7409,
885
+ "step": 125
886
+ },
887
+ {
888
+ "epoch": 0.3372365339578454,
889
+ "grad_norm": 3.603746175765991,
890
+ "learning_rate": 9.717821739425286e-06,
891
+ "loss": 0.7286,
892
+ "step": 126
893
+ },
894
+ {
895
+ "epoch": 0.3399130143860823,
896
+ "grad_norm": 4.118206977844238,
897
+ "learning_rate": 9.710320814480448e-06,
898
+ "loss": 0.7164,
899
+ "step": 127
900
+ },
901
+ {
902
+ "epoch": 0.34258949481431916,
903
+ "grad_norm": 3.6243247985839844,
904
+ "learning_rate": 9.702724469702107e-06,
905
+ "loss": 0.7772,
906
+ "step": 128
907
+ },
908
+ {
909
+ "epoch": 0.34526597524255603,
910
+ "grad_norm": 3.8515329360961914,
911
+ "learning_rate": 9.695032858974042e-06,
912
+ "loss": 0.7173,
913
+ "step": 129
914
+ },
915
+ {
916
+ "epoch": 0.3479424556707929,
917
+ "grad_norm": 4.385611534118652,
918
+ "learning_rate": 9.687246138109888e-06,
919
+ "loss": 0.8672,
920
+ "step": 130
921
+ },
922
+ {
923
+ "epoch": 0.3506189360990298,
924
+ "grad_norm": 3.6981892585754395,
925
+ "learning_rate": 9.679364464849983e-06,
926
+ "loss": 0.6571,
927
+ "step": 131
928
+ },
929
+ {
930
+ "epoch": 0.35329541652726665,
931
+ "grad_norm": 4.061744213104248,
932
+ "learning_rate": 9.671387998858178e-06,
933
+ "loss": 0.7114,
934
+ "step": 132
935
+ },
936
+ {
937
+ "epoch": 0.3559718969555035,
938
+ "grad_norm": 4.793642997741699,
939
+ "learning_rate": 9.663316901718599e-06,
940
+ "loss": 0.7537,
941
+ "step": 133
942
+ },
943
+ {
944
+ "epoch": 0.3586483773837404,
945
+ "grad_norm": 3.751180648803711,
946
+ "learning_rate": 9.655151336932362e-06,
947
+ "loss": 0.6826,
948
+ "step": 134
949
+ },
950
+ {
951
+ "epoch": 0.36132485781197726,
952
+ "grad_norm": 3.8843295574188232,
953
+ "learning_rate": 9.646891469914285e-06,
954
+ "loss": 0.6868,
955
+ "step": 135
956
+ },
957
+ {
958
+ "epoch": 0.36400133824021413,
959
+ "grad_norm": 4.41197395324707,
960
+ "learning_rate": 9.638537467989517e-06,
961
+ "loss": 0.7224,
962
+ "step": 136
963
+ },
964
+ {
965
+ "epoch": 0.366677818668451,
966
+ "grad_norm": 3.2272307872772217,
967
+ "learning_rate": 9.630089500390154e-06,
968
+ "loss": 0.5735,
969
+ "step": 137
970
+ },
971
+ {
972
+ "epoch": 0.3693542990966879,
973
+ "grad_norm": 3.5232367515563965,
974
+ "learning_rate": 9.621547738251816e-06,
975
+ "loss": 0.7621,
976
+ "step": 138
977
+ },
978
+ {
979
+ "epoch": 0.37203077952492475,
980
+ "grad_norm": 3.4856607913970947,
981
+ "learning_rate": 9.61291235461017e-06,
982
+ "loss": 0.6517,
983
+ "step": 139
984
+ },
985
+ {
986
+ "epoch": 0.3747072599531616,
987
+ "grad_norm": 3.958531379699707,
988
+ "learning_rate": 9.604183524397439e-06,
989
+ "loss": 0.8994,
990
+ "step": 140
991
+ },
992
+ {
993
+ "epoch": 0.37738374038139844,
994
+ "grad_norm": 3.3670690059661865,
995
+ "learning_rate": 9.595361424438841e-06,
996
+ "loss": 0.5975,
997
+ "step": 141
998
+ },
999
+ {
1000
+ "epoch": 0.3800602208096353,
1001
+ "grad_norm": 4.388239860534668,
1002
+ "learning_rate": 9.586446233449024e-06,
1003
+ "loss": 0.7168,
1004
+ "step": 142
1005
+ },
1006
+ {
1007
+ "epoch": 0.3827367012378722,
1008
+ "grad_norm": 3.513667345046997,
1009
+ "learning_rate": 9.577438132028431e-06,
1010
+ "loss": 0.7645,
1011
+ "step": 143
1012
+ },
1013
+ {
1014
+ "epoch": 0.38541318166610905,
1015
+ "grad_norm": 3.383009195327759,
1016
+ "learning_rate": 9.568337302659652e-06,
1017
+ "loss": 0.7159,
1018
+ "step": 144
1019
+ },
1020
+ {
1021
+ "epoch": 0.3880896620943459,
1022
+ "grad_norm": 3.7690727710723877,
1023
+ "learning_rate": 9.559143929703724e-06,
1024
+ "loss": 0.8527,
1025
+ "step": 145
1026
+ },
1027
+ {
1028
+ "epoch": 0.3907661425225828,
1029
+ "grad_norm": 3.1912662982940674,
1030
+ "learning_rate": 9.549858199396394e-06,
1031
+ "loss": 0.6288,
1032
+ "step": 146
1033
+ },
1034
+ {
1035
+ "epoch": 0.39344262295081966,
1036
+ "grad_norm": 3.616091012954712,
1037
+ "learning_rate": 9.540480299844345e-06,
1038
+ "loss": 0.6928,
1039
+ "step": 147
1040
+ },
1041
+ {
1042
+ "epoch": 0.39611910337905654,
1043
+ "grad_norm": 3.7126312255859375,
1044
+ "learning_rate": 9.531010421021396e-06,
1045
+ "loss": 0.7583,
1046
+ "step": 148
1047
+ },
1048
+ {
1049
+ "epoch": 0.3987955838072934,
1050
+ "grad_norm": 3.61950421333313,
1051
+ "learning_rate": 9.52144875476464e-06,
1052
+ "loss": 0.6796,
1053
+ "step": 149
1054
+ },
1055
+ {
1056
+ "epoch": 0.4014720642355303,
1057
+ "grad_norm": 3.0952789783477783,
1058
+ "learning_rate": 9.511795494770563e-06,
1059
+ "loss": 0.6761,
1060
+ "step": 150
1061
+ },
1062
+ {
1063
+ "epoch": 0.40414854466376715,
1064
+ "grad_norm": 3.4935519695281982,
1065
+ "learning_rate": 9.50205083659113e-06,
1066
+ "loss": 0.622,
1067
+ "step": 151
1068
+ },
1069
+ {
1070
+ "epoch": 0.406825025092004,
1071
+ "grad_norm": 4.367190837860107,
1072
+ "learning_rate": 9.492214977629804e-06,
1073
+ "loss": 0.8146,
1074
+ "step": 152
1075
+ },
1076
+ {
1077
+ "epoch": 0.4095015055202409,
1078
+ "grad_norm": 3.501554489135742,
1079
+ "learning_rate": 9.482288117137561e-06,
1080
+ "loss": 0.6101,
1081
+ "step": 153
1082
+ },
1083
+ {
1084
+ "epoch": 0.41217798594847777,
1085
+ "grad_norm": 3.7483065128326416,
1086
+ "learning_rate": 9.472270456208856e-06,
1087
+ "loss": 0.6095,
1088
+ "step": 154
1089
+ },
1090
+ {
1091
+ "epoch": 0.41485446637671464,
1092
+ "grad_norm": 2.7559807300567627,
1093
+ "learning_rate": 9.462162197777533e-06,
1094
+ "loss": 0.6167,
1095
+ "step": 155
1096
+ },
1097
+ {
1098
+ "epoch": 0.4175309468049515,
1099
+ "grad_norm": 3.520737409591675,
1100
+ "learning_rate": 9.451963546612737e-06,
1101
+ "loss": 0.5388,
1102
+ "step": 156
1103
+ },
1104
+ {
1105
+ "epoch": 0.4202074272331884,
1106
+ "grad_norm": 4.042897701263428,
1107
+ "learning_rate": 9.441674709314743e-06,
1108
+ "loss": 0.7228,
1109
+ "step": 157
1110
+ },
1111
+ {
1112
+ "epoch": 0.42288390766142525,
1113
+ "grad_norm": 4.090766429901123,
1114
+ "learning_rate": 9.431295894310786e-06,
1115
+ "loss": 0.5569,
1116
+ "step": 158
1117
+ },
1118
+ {
1119
+ "epoch": 0.42556038808966207,
1120
+ "grad_norm": 4.141130447387695,
1121
+ "learning_rate": 9.420827311850836e-06,
1122
+ "loss": 0.6297,
1123
+ "step": 159
1124
+ },
1125
+ {
1126
+ "epoch": 0.42823686851789894,
1127
+ "grad_norm": 3.472973346710205,
1128
+ "learning_rate": 9.410269174003333e-06,
1129
+ "loss": 0.6868,
1130
+ "step": 160
1131
+ },
1132
+ {
1133
+ "epoch": 0.4309133489461358,
1134
+ "grad_norm": 3.8000664710998535,
1135
+ "learning_rate": 9.399621694650898e-06,
1136
+ "loss": 0.6547,
1137
+ "step": 161
1138
+ },
1139
+ {
1140
+ "epoch": 0.4335898293743727,
1141
+ "grad_norm": 4.582136154174805,
1142
+ "learning_rate": 9.388885089485995e-06,
1143
+ "loss": 0.7552,
1144
+ "step": 162
1145
+ },
1146
+ {
1147
+ "epoch": 0.43626630980260955,
1148
+ "grad_norm": 3.4165501594543457,
1149
+ "learning_rate": 9.378059576006567e-06,
1150
+ "loss": 0.622,
1151
+ "step": 163
1152
+ },
1153
+ {
1154
+ "epoch": 0.4389427902308464,
1155
+ "grad_norm": 3.642669916152954,
1156
+ "learning_rate": 9.36714537351162e-06,
1157
+ "loss": 0.6302,
1158
+ "step": 164
1159
+ },
1160
+ {
1161
+ "epoch": 0.4416192706590833,
1162
+ "grad_norm": 3.53709077835083,
1163
+ "learning_rate": 9.356142703096793e-06,
1164
+ "loss": 0.5357,
1165
+ "step": 165
1166
+ },
1167
+ {
1168
+ "epoch": 0.44429575108732017,
1169
+ "grad_norm": 3.6043941974639893,
1170
+ "learning_rate": 9.345051787649877e-06,
1171
+ "loss": 0.6978,
1172
+ "step": 166
1173
+ },
1174
+ {
1175
+ "epoch": 0.44697223151555704,
1176
+ "grad_norm": 4.232686519622803,
1177
+ "learning_rate": 9.333872851846285e-06,
1178
+ "loss": 0.6698,
1179
+ "step": 167
1180
+ },
1181
+ {
1182
+ "epoch": 0.4496487119437939,
1183
+ "grad_norm": 3.1254665851593018,
1184
+ "learning_rate": 9.322606122144524e-06,
1185
+ "loss": 0.5173,
1186
+ "step": 168
1187
+ },
1188
+ {
1189
+ "epoch": 0.4523251923720308,
1190
+ "grad_norm": 4.0060930252075195,
1191
+ "learning_rate": 9.311251826781587e-06,
1192
+ "loss": 0.6674,
1193
+ "step": 169
1194
+ },
1195
+ {
1196
+ "epoch": 0.45500167280026765,
1197
+ "grad_norm": 4.720844268798828,
1198
+ "learning_rate": 9.299810195768341e-06,
1199
+ "loss": 0.8349,
1200
+ "step": 170
1201
+ },
1202
+ {
1203
+ "epoch": 0.4576781532285045,
1204
+ "grad_norm": 4.017560005187988,
1205
+ "learning_rate": 9.288281460884864e-06,
1206
+ "loss": 0.8489,
1207
+ "step": 171
1208
+ },
1209
+ {
1210
+ "epoch": 0.4603546336567414,
1211
+ "grad_norm": 4.482756614685059,
1212
+ "learning_rate": 9.276665855675751e-06,
1213
+ "loss": 0.6877,
1214
+ "step": 172
1215
+ },
1216
+ {
1217
+ "epoch": 0.46303111408497827,
1218
+ "grad_norm": 3.9189202785491943,
1219
+ "learning_rate": 9.264963615445378e-06,
1220
+ "loss": 0.58,
1221
+ "step": 173
1222
+ },
1223
+ {
1224
+ "epoch": 0.46570759451321514,
1225
+ "grad_norm": 3.285618543624878,
1226
+ "learning_rate": 9.25317497725315e-06,
1227
+ "loss": 0.585,
1228
+ "step": 174
1229
+ },
1230
+ {
1231
+ "epoch": 0.468384074941452,
1232
+ "grad_norm": 3.4484951496124268,
1233
+ "learning_rate": 9.241300179908672e-06,
1234
+ "loss": 0.5898,
1235
+ "step": 175
1236
+ },
1237
+ {
1238
+ "epoch": 0.4710605553696889,
1239
+ "grad_norm": 3.892336845397949,
1240
+ "learning_rate": 9.229339463966942e-06,
1241
+ "loss": 0.7691,
1242
+ "step": 176
1243
+ },
1244
+ {
1245
+ "epoch": 0.47373703579792575,
1246
+ "grad_norm": 3.3191475868225098,
1247
+ "learning_rate": 9.217293071723455e-06,
1248
+ "loss": 0.6468,
1249
+ "step": 177
1250
+ },
1251
+ {
1252
+ "epoch": 0.47641351622616257,
1253
+ "grad_norm": 3.2376253604888916,
1254
+ "learning_rate": 9.205161247209303e-06,
1255
+ "loss": 0.6233,
1256
+ "step": 178
1257
+ },
1258
+ {
1259
+ "epoch": 0.47908999665439944,
1260
+ "grad_norm": 3.029269218444824,
1261
+ "learning_rate": 9.192944236186237e-06,
1262
+ "loss": 0.5845,
1263
+ "step": 179
1264
+ },
1265
+ {
1266
+ "epoch": 0.4817664770826363,
1267
+ "grad_norm": 3.5170111656188965,
1268
+ "learning_rate": 9.180642286141678e-06,
1269
+ "loss": 0.6815,
1270
+ "step": 180
1271
+ },
1272
+ {
1273
+ "epoch": 0.4844429575108732,
1274
+ "grad_norm": 4.0202155113220215,
1275
+ "learning_rate": 9.16825564628371e-06,
1276
+ "loss": 0.7131,
1277
+ "step": 181
1278
+ },
1279
+ {
1280
+ "epoch": 0.48711943793911006,
1281
+ "grad_norm": 3.566044330596924,
1282
+ "learning_rate": 9.15578456753603e-06,
1283
+ "loss": 0.6628,
1284
+ "step": 182
1285
+ },
1286
+ {
1287
+ "epoch": 0.4897959183673469,
1288
+ "grad_norm": 3.9581563472747803,
1289
+ "learning_rate": 9.143229302532866e-06,
1290
+ "loss": 0.6497,
1291
+ "step": 183
1292
+ },
1293
+ {
1294
+ "epoch": 0.4924723987955838,
1295
+ "grad_norm": 3.2238948345184326,
1296
+ "learning_rate": 9.130590105613854e-06,
1297
+ "loss": 0.5773,
1298
+ "step": 184
1299
+ },
1300
+ {
1301
+ "epoch": 0.49514887922382067,
1302
+ "grad_norm": 4.199437618255615,
1303
+ "learning_rate": 9.117867232818897e-06,
1304
+ "loss": 0.5552,
1305
+ "step": 185
1306
+ },
1307
+ {
1308
+ "epoch": 0.49782535965205754,
1309
+ "grad_norm": 3.223834753036499,
1310
+ "learning_rate": 9.105060941882966e-06,
1311
+ "loss": 0.4985,
1312
+ "step": 186
1313
+ },
1314
+ {
1315
+ "epoch": 0.5005018400802944,
1316
+ "grad_norm": 3.2973475456237793,
1317
+ "learning_rate": 9.092171492230883e-06,
1318
+ "loss": 0.6637,
1319
+ "step": 187
1320
+ },
1321
+ {
1322
+ "epoch": 0.5031783205085313,
1323
+ "grad_norm": 3.9001471996307373,
1324
+ "learning_rate": 9.079199144972072e-06,
1325
+ "loss": 0.7487,
1326
+ "step": 188
1327
+ },
1328
+ {
1329
+ "epoch": 0.5058548009367682,
1330
+ "grad_norm": 3.913593292236328,
1331
+ "learning_rate": 9.066144162895259e-06,
1332
+ "loss": 0.5707,
1333
+ "step": 189
1334
+ },
1335
+ {
1336
+ "epoch": 0.508531281365005,
1337
+ "grad_norm": 2.6204168796539307,
1338
+ "learning_rate": 9.053006810463156e-06,
1339
+ "loss": 0.4685,
1340
+ "step": 190
1341
+ },
1342
+ {
1343
+ "epoch": 0.5112077617932419,
1344
+ "grad_norm": 3.507556200027466,
1345
+ "learning_rate": 9.039787353807101e-06,
1346
+ "loss": 0.5564,
1347
+ "step": 191
1348
+ },
1349
+ {
1350
+ "epoch": 0.5138842422214788,
1351
+ "grad_norm": 3.072106122970581,
1352
+ "learning_rate": 9.026486060721668e-06,
1353
+ "loss": 0.5427,
1354
+ "step": 192
1355
+ },
1356
+ {
1357
+ "epoch": 0.5165607226497156,
1358
+ "grad_norm": 2.578962802886963,
1359
+ "learning_rate": 9.01310320065924e-06,
1360
+ "loss": 0.3969,
1361
+ "step": 193
1362
+ },
1363
+ {
1364
+ "epoch": 0.5192372030779525,
1365
+ "grad_norm": 3.3628005981445312,
1366
+ "learning_rate": 8.999639044724555e-06,
1367
+ "loss": 0.6076,
1368
+ "step": 194
1369
+ },
1370
+ {
1371
+ "epoch": 0.5219136835061894,
1372
+ "grad_norm": 3.64412784576416,
1373
+ "learning_rate": 8.986093865669205e-06,
1374
+ "loss": 0.6826,
1375
+ "step": 195
1376
+ },
1377
+ {
1378
+ "epoch": 0.5245901639344263,
1379
+ "grad_norm": 3.7886106967926025,
1380
+ "learning_rate": 8.972467937886122e-06,
1381
+ "loss": 0.5279,
1382
+ "step": 196
1383
+ },
1384
+ {
1385
+ "epoch": 0.5272666443626631,
1386
+ "grad_norm": 3.543992757797241,
1387
+ "learning_rate": 8.958761537404012e-06,
1388
+ "loss": 0.5748,
1389
+ "step": 197
1390
+ },
1391
+ {
1392
+ "epoch": 0.5299431247909,
1393
+ "grad_norm": 4.291012763977051,
1394
+ "learning_rate": 8.944974941881766e-06,
1395
+ "loss": 0.6919,
1396
+ "step": 198
1397
+ },
1398
+ {
1399
+ "epoch": 0.5326196052191369,
1400
+ "grad_norm": 4.906763553619385,
1401
+ "learning_rate": 8.931108430602834e-06,
1402
+ "loss": 0.5633,
1403
+ "step": 199
1404
+ },
1405
+ {
1406
+ "epoch": 0.5352960856473737,
1407
+ "grad_norm": 3.177494764328003,
1408
+ "learning_rate": 8.917162284469569e-06,
1409
+ "loss": 0.5792,
1410
+ "step": 200
1411
+ },
1412
+ {
1413
+ "epoch": 0.5379725660756106,
1414
+ "grad_norm": 3.5298800468444824,
1415
+ "learning_rate": 8.903136785997533e-06,
1416
+ "loss": 0.5616,
1417
+ "step": 201
1418
+ },
1419
+ {
1420
+ "epoch": 0.5406490465038475,
1421
+ "grad_norm": 3.5532355308532715,
1422
+ "learning_rate": 8.889032219309781e-06,
1423
+ "loss": 0.6682,
1424
+ "step": 202
1425
+ },
1426
+ {
1427
+ "epoch": 0.5433255269320844,
1428
+ "grad_norm": 4.2544026374816895,
1429
+ "learning_rate": 8.874848870131098e-06,
1430
+ "loss": 0.544,
1431
+ "step": 203
1432
+ },
1433
+ {
1434
+ "epoch": 0.5460020073603212,
1435
+ "grad_norm": 3.0393524169921875,
1436
+ "learning_rate": 8.860587025782215e-06,
1437
+ "loss": 0.6333,
1438
+ "step": 204
1439
+ },
1440
+ {
1441
+ "epoch": 0.5486784877885581,
1442
+ "grad_norm": 4.1217122077941895,
1443
+ "learning_rate": 8.846246975173985e-06,
1444
+ "loss": 0.597,
1445
+ "step": 205
1446
+ },
1447
+ {
1448
+ "epoch": 0.5513549682167949,
1449
+ "grad_norm": 3.7197227478027344,
1450
+ "learning_rate": 8.831829008801536e-06,
1451
+ "loss": 0.6568,
1452
+ "step": 206
1453
+ },
1454
+ {
1455
+ "epoch": 0.5540314486450317,
1456
+ "grad_norm": 3.254206657409668,
1457
+ "learning_rate": 8.817333418738382e-06,
1458
+ "loss": 0.5321,
1459
+ "step": 207
1460
+ },
1461
+ {
1462
+ "epoch": 0.5567079290732686,
1463
+ "grad_norm": 3.084484577178955,
1464
+ "learning_rate": 8.802760498630507e-06,
1465
+ "loss": 0.4752,
1466
+ "step": 208
1467
+ },
1468
+ {
1469
+ "epoch": 0.5593844095015055,
1470
+ "grad_norm": 4.343369483947754,
1471
+ "learning_rate": 8.788110543690415e-06,
1472
+ "loss": 0.6444,
1473
+ "step": 209
1474
+ },
1475
+ {
1476
+ "epoch": 0.5620608899297423,
1477
+ "grad_norm": 3.27128529548645,
1478
+ "learning_rate": 8.773383850691155e-06,
1479
+ "loss": 0.5282,
1480
+ "step": 210
1481
+ },
1482
+ {
1483
+ "epoch": 0.5647373703579792,
1484
+ "grad_norm": 3.0274081230163574,
1485
+ "learning_rate": 8.758580717960303e-06,
1486
+ "loss": 0.6607,
1487
+ "step": 211
1488
+ },
1489
+ {
1490
+ "epoch": 0.5674138507862161,
1491
+ "grad_norm": 3.622535228729248,
1492
+ "learning_rate": 8.743701445373922e-06,
1493
+ "loss": 0.5787,
1494
+ "step": 212
1495
+ },
1496
+ {
1497
+ "epoch": 0.570090331214453,
1498
+ "grad_norm": 3.6681504249572754,
1499
+ "learning_rate": 8.728746334350483e-06,
1500
+ "loss": 0.5879,
1501
+ "step": 213
1502
+ },
1503
+ {
1504
+ "epoch": 0.5727668116426898,
1505
+ "grad_norm": 3.163616895675659,
1506
+ "learning_rate": 8.713715687844772e-06,
1507
+ "loss": 0.5888,
1508
+ "step": 214
1509
+ },
1510
+ {
1511
+ "epoch": 0.5754432920709267,
1512
+ "grad_norm": 3.0883967876434326,
1513
+ "learning_rate": 8.698609810341733e-06,
1514
+ "loss": 0.5189,
1515
+ "step": 215
1516
+ },
1517
+ {
1518
+ "epoch": 0.5781197724991636,
1519
+ "grad_norm": 3.443776845932007,
1520
+ "learning_rate": 8.683429007850313e-06,
1521
+ "loss": 0.5303,
1522
+ "step": 216
1523
+ },
1524
+ {
1525
+ "epoch": 0.5807962529274004,
1526
+ "grad_norm": 3.3351690769195557,
1527
+ "learning_rate": 8.668173587897261e-06,
1528
+ "loss": 0.5296,
1529
+ "step": 217
1530
+ },
1531
+ {
1532
+ "epoch": 0.5834727333556373,
1533
+ "grad_norm": 3.8925840854644775,
1534
+ "learning_rate": 8.6528438595209e-06,
1535
+ "loss": 0.499,
1536
+ "step": 218
1537
+ },
1538
+ {
1539
+ "epoch": 0.5861492137838742,
1540
+ "grad_norm": 3.4368419647216797,
1541
+ "learning_rate": 8.637440133264858e-06,
1542
+ "loss": 0.4166,
1543
+ "step": 219
1544
+ },
1545
+ {
1546
+ "epoch": 0.5888256942121111,
1547
+ "grad_norm": 3.439969539642334,
1548
+ "learning_rate": 8.621962721171789e-06,
1549
+ "loss": 0.5417,
1550
+ "step": 220
1551
+ },
1552
+ {
1553
+ "epoch": 0.5915021746403479,
1554
+ "grad_norm": 3.2590739727020264,
1555
+ "learning_rate": 8.60641193677704e-06,
1556
+ "loss": 0.4925,
1557
+ "step": 221
1558
+ },
1559
+ {
1560
+ "epoch": 0.5941786550685848,
1561
+ "grad_norm": 3.918884754180908,
1562
+ "learning_rate": 8.59078809510231e-06,
1563
+ "loss": 0.75,
1564
+ "step": 222
1565
+ },
1566
+ {
1567
+ "epoch": 0.5968551354968217,
1568
+ "grad_norm": 4.21658182144165,
1569
+ "learning_rate": 8.57509151264926e-06,
1570
+ "loss": 0.5138,
1571
+ "step": 223
1572
+ },
1573
+ {
1574
+ "epoch": 0.5995316159250585,
1575
+ "grad_norm": 3.2700538635253906,
1576
+ "learning_rate": 8.55932250739311e-06,
1577
+ "loss": 0.4855,
1578
+ "step": 224
1579
+ },
1580
+ {
1581
+ "epoch": 0.6022080963532954,
1582
+ "grad_norm": 3.5396180152893066,
1583
+ "learning_rate": 8.543481398776188e-06,
1584
+ "loss": 0.626,
1585
+ "step": 225
1586
+ },
1587
+ {
1588
+ "epoch": 0.6048845767815323,
1589
+ "grad_norm": 3.267137289047241,
1590
+ "learning_rate": 8.527568507701467e-06,
1591
+ "loss": 0.5043,
1592
+ "step": 226
1593
+ },
1594
+ {
1595
+ "epoch": 0.6075610572097692,
1596
+ "grad_norm": 4.908371448516846,
1597
+ "learning_rate": 8.511584156526059e-06,
1598
+ "loss": 0.6942,
1599
+ "step": 227
1600
+ },
1601
+ {
1602
+ "epoch": 0.610237537638006,
1603
+ "grad_norm": 3.7445895671844482,
1604
+ "learning_rate": 8.495528669054688e-06,
1605
+ "loss": 0.5475,
1606
+ "step": 228
1607
+ },
1608
+ {
1609
+ "epoch": 0.6129140180662429,
1610
+ "grad_norm": 4.3304033279418945,
1611
+ "learning_rate": 8.479402370533127e-06,
1612
+ "loss": 0.6943,
1613
+ "step": 229
1614
+ },
1615
+ {
1616
+ "epoch": 0.6155904984944798,
1617
+ "grad_norm": 3.6123831272125244,
1618
+ "learning_rate": 8.463205587641614e-06,
1619
+ "loss": 0.5365,
1620
+ "step": 230
1621
+ },
1622
+ {
1623
+ "epoch": 0.6182669789227166,
1624
+ "grad_norm": 4.3004045486450195,
1625
+ "learning_rate": 8.44693864848823e-06,
1626
+ "loss": 0.6377,
1627
+ "step": 231
1628
+ },
1629
+ {
1630
+ "epoch": 0.6209434593509535,
1631
+ "grad_norm": 3.1220314502716064,
1632
+ "learning_rate": 8.430601882602256e-06,
1633
+ "loss": 0.4646,
1634
+ "step": 232
1635
+ },
1636
+ {
1637
+ "epoch": 0.6236199397791904,
1638
+ "grad_norm": 4.166379451751709,
1639
+ "learning_rate": 8.414195620927491e-06,
1640
+ "loss": 0.5738,
1641
+ "step": 233
1642
+ },
1643
+ {
1644
+ "epoch": 0.6262964202074273,
1645
+ "grad_norm": 4.061609745025635,
1646
+ "learning_rate": 8.397720195815561e-06,
1647
+ "loss": 0.4654,
1648
+ "step": 234
1649
+ },
1650
+ {
1651
+ "epoch": 0.6289729006356641,
1652
+ "grad_norm": 3.618021249771118,
1653
+ "learning_rate": 8.381175941019171e-06,
1654
+ "loss": 0.5699,
1655
+ "step": 235
1656
+ },
1657
+ {
1658
+ "epoch": 0.631649381063901,
1659
+ "grad_norm": 3.2099597454071045,
1660
+ "learning_rate": 8.364563191685348e-06,
1661
+ "loss": 0.5481,
1662
+ "step": 236
1663
+ },
1664
+ {
1665
+ "epoch": 0.6343258614921379,
1666
+ "grad_norm": 3.3331351280212402,
1667
+ "learning_rate": 8.347882284348665e-06,
1668
+ "loss": 0.5351,
1669
+ "step": 237
1670
+ },
1671
+ {
1672
+ "epoch": 0.6370023419203747,
1673
+ "grad_norm": 3.91636323928833,
1674
+ "learning_rate": 8.331133556924404e-06,
1675
+ "loss": 0.6588,
1676
+ "step": 238
1677
+ },
1678
+ {
1679
+ "epoch": 0.6396788223486116,
1680
+ "grad_norm": 3.613023519515991,
1681
+ "learning_rate": 8.314317348701724e-06,
1682
+ "loss": 0.6157,
1683
+ "step": 239
1684
+ },
1685
+ {
1686
+ "epoch": 0.6423553027768485,
1687
+ "grad_norm": 3.674731731414795,
1688
+ "learning_rate": 8.297434000336781e-06,
1689
+ "loss": 0.5251,
1690
+ "step": 240
1691
+ },
1692
+ {
1693
+ "epoch": 0.6450317832050854,
1694
+ "grad_norm": 3.4246506690979004,
1695
+ "learning_rate": 8.280483853845831e-06,
1696
+ "loss": 0.5449,
1697
+ "step": 241
1698
+ },
1699
+ {
1700
+ "epoch": 0.6477082636333222,
1701
+ "grad_norm": 3.308436155319214,
1702
+ "learning_rate": 8.263467252598303e-06,
1703
+ "loss": 0.6432,
1704
+ "step": 242
1705
+ },
1706
+ {
1707
+ "epoch": 0.650384744061559,
1708
+ "grad_norm": 3.3527414798736572,
1709
+ "learning_rate": 8.246384541309835e-06,
1710
+ "loss": 0.6035,
1711
+ "step": 243
1712
+ },
1713
+ {
1714
+ "epoch": 0.6530612244897959,
1715
+ "grad_norm": 3.5172030925750732,
1716
+ "learning_rate": 8.2292360660353e-06,
1717
+ "loss": 0.522,
1718
+ "step": 244
1719
+ },
1720
+ {
1721
+ "epoch": 0.6557377049180327,
1722
+ "grad_norm": 3.086130142211914,
1723
+ "learning_rate": 8.21202217416179e-06,
1724
+ "loss": 0.4867,
1725
+ "step": 245
1726
+ },
1727
+ {
1728
+ "epoch": 0.6584141853462696,
1729
+ "grad_norm": 4.54761266708374,
1730
+ "learning_rate": 8.194743214401587e-06,
1731
+ "loss": 0.5607,
1732
+ "step": 246
1733
+ },
1734
+ {
1735
+ "epoch": 0.6610906657745065,
1736
+ "grad_norm": 3.190037250518799,
1737
+ "learning_rate": 8.17739953678508e-06,
1738
+ "loss": 0.4674,
1739
+ "step": 247
1740
+ },
1741
+ {
1742
+ "epoch": 0.6637671462027434,
1743
+ "grad_norm": 3.303924083709717,
1744
+ "learning_rate": 8.1599914926537e-06,
1745
+ "loss": 0.5116,
1746
+ "step": 248
1747
+ },
1748
+ {
1749
+ "epoch": 0.6664436266309802,
1750
+ "grad_norm": 3.0905206203460693,
1751
+ "learning_rate": 8.142519434652782e-06,
1752
+ "loss": 0.5041,
1753
+ "step": 249
1754
+ },
1755
+ {
1756
+ "epoch": 0.6691201070592171,
1757
+ "grad_norm": 3.9465432167053223,
1758
+ "learning_rate": 8.124983716724434e-06,
1759
+ "loss": 0.5009,
1760
+ "step": 250
1761
+ },
1762
+ {
1763
+ "epoch": 0.671796587487454,
1764
+ "grad_norm": 3.0109636783599854,
1765
+ "learning_rate": 8.107384694100355e-06,
1766
+ "loss": 0.5041,
1767
+ "step": 251
1768
+ },
1769
+ {
1770
+ "epoch": 0.6744730679156908,
1771
+ "grad_norm": 3.4623425006866455,
1772
+ "learning_rate": 8.089722723294654e-06,
1773
+ "loss": 0.5509,
1774
+ "step": 252
1775
+ },
1776
+ {
1777
+ "epoch": 0.6771495483439277,
1778
+ "grad_norm": 4.234053134918213,
1779
+ "learning_rate": 8.071998162096613e-06,
1780
+ "loss": 0.6164,
1781
+ "step": 253
1782
+ },
1783
+ {
1784
+ "epoch": 0.6798260287721646,
1785
+ "grad_norm": 3.2587759494781494,
1786
+ "learning_rate": 8.054211369563448e-06,
1787
+ "loss": 0.5531,
1788
+ "step": 254
1789
+ },
1790
+ {
1791
+ "epoch": 0.6825025092004015,
1792
+ "grad_norm": 3.6140904426574707,
1793
+ "learning_rate": 8.036362706013033e-06,
1794
+ "loss": 0.5566,
1795
+ "step": 255
1796
+ },
1797
+ {
1798
+ "epoch": 0.6851789896286383,
1799
+ "grad_norm": 2.9970974922180176,
1800
+ "learning_rate": 8.018452533016604e-06,
1801
+ "loss": 0.4614,
1802
+ "step": 256
1803
+ },
1804
+ {
1805
+ "epoch": 0.6878554700568752,
1806
+ "grad_norm": 3.6728920936584473,
1807
+ "learning_rate": 8.000481213391422e-06,
1808
+ "loss": 0.5529,
1809
+ "step": 257
1810
+ },
1811
+ {
1812
+ "epoch": 0.6905319504851121,
1813
+ "grad_norm": 4.451485633850098,
1814
+ "learning_rate": 7.982449111193445e-06,
1815
+ "loss": 0.5997,
1816
+ "step": 258
1817
+ },
1818
+ {
1819
+ "epoch": 0.6932084309133489,
1820
+ "grad_norm": 3.2221109867095947,
1821
+ "learning_rate": 7.96435659170993e-06,
1822
+ "loss": 0.4782,
1823
+ "step": 259
1824
+ },
1825
+ {
1826
+ "epoch": 0.6958849113415858,
1827
+ "grad_norm": 3.3525805473327637,
1828
+ "learning_rate": 7.946204021452049e-06,
1829
+ "loss": 0.5988,
1830
+ "step": 260
1831
+ },
1832
+ {
1833
+ "epoch": 0.6985613917698227,
1834
+ "grad_norm": 3.346829652786255,
1835
+ "learning_rate": 7.92799176814746e-06,
1836
+ "loss": 0.461,
1837
+ "step": 261
1838
+ },
1839
+ {
1840
+ "epoch": 0.7012378721980596,
1841
+ "grad_norm": 3.4841763973236084,
1842
+ "learning_rate": 7.90972020073285e-06,
1843
+ "loss": 0.5972,
1844
+ "step": 262
1845
+ },
1846
+ {
1847
+ "epoch": 0.7039143526262964,
1848
+ "grad_norm": 3.1365015506744385,
1849
+ "learning_rate": 7.891389689346479e-06,
1850
+ "loss": 0.4601,
1851
+ "step": 263
1852
+ },
1853
+ {
1854
+ "epoch": 0.7065908330545333,
1855
+ "grad_norm": 2.762209177017212,
1856
+ "learning_rate": 7.873000605320658e-06,
1857
+ "loss": 0.4713,
1858
+ "step": 264
1859
+ },
1860
+ {
1861
+ "epoch": 0.7092673134827702,
1862
+ "grad_norm": 3.6923787593841553,
1863
+ "learning_rate": 7.85455332117425e-06,
1864
+ "loss": 0.5868,
1865
+ "step": 265
1866
+ },
1867
+ {
1868
+ "epoch": 0.711943793911007,
1869
+ "grad_norm": 3.483915090560913,
1870
+ "learning_rate": 7.836048210605109e-06,
1871
+ "loss": 0.5488,
1872
+ "step": 266
1873
+ },
1874
+ {
1875
+ "epoch": 0.7146202743392439,
1876
+ "grad_norm": 3.7506582736968994,
1877
+ "learning_rate": 7.817485648482514e-06,
1878
+ "loss": 0.506,
1879
+ "step": 267
1880
+ },
1881
+ {
1882
+ "epoch": 0.7172967547674808,
1883
+ "grad_norm": 3.148474931716919,
1884
+ "learning_rate": 7.798866010839577e-06,
1885
+ "loss": 0.4793,
1886
+ "step": 268
1887
+ },
1888
+ {
1889
+ "epoch": 0.7199732351957177,
1890
+ "grad_norm": 3.194519281387329,
1891
+ "learning_rate": 7.780189674865617e-06,
1892
+ "loss": 0.5573,
1893
+ "step": 269
1894
+ },
1895
+ {
1896
+ "epoch": 0.7226497156239545,
1897
+ "grad_norm": 3.321631669998169,
1898
+ "learning_rate": 7.761457018898536e-06,
1899
+ "loss": 0.4949,
1900
+ "step": 270
1901
+ },
1902
+ {
1903
+ "epoch": 0.7253261960521914,
1904
+ "grad_norm": 3.322242021560669,
1905
+ "learning_rate": 7.742668422417137e-06,
1906
+ "loss": 0.5329,
1907
+ "step": 271
1908
+ },
1909
+ {
1910
+ "epoch": 0.7280026764804283,
1911
+ "grad_norm": 3.460432291030884,
1912
+ "learning_rate": 7.723824266033444e-06,
1913
+ "loss": 0.5798,
1914
+ "step": 272
1915
+ },
1916
+ {
1917
+ "epoch": 0.7306791569086651,
1918
+ "grad_norm": 3.0027997493743896,
1919
+ "learning_rate": 7.704924931484997e-06,
1920
+ "loss": 0.5512,
1921
+ "step": 273
1922
+ },
1923
+ {
1924
+ "epoch": 0.733355637336902,
1925
+ "grad_norm": 3.621713399887085,
1926
+ "learning_rate": 7.685970801627108e-06,
1927
+ "loss": 0.6221,
1928
+ "step": 274
1929
+ },
1930
+ {
1931
+ "epoch": 0.7360321177651389,
1932
+ "grad_norm": 2.8740806579589844,
1933
+ "learning_rate": 7.666962260425113e-06,
1934
+ "loss": 0.532,
1935
+ "step": 275
1936
+ },
1937
+ {
1938
+ "epoch": 0.7387085981933758,
1939
+ "grad_norm": 2.9761555194854736,
1940
+ "learning_rate": 7.647899692946594e-06,
1941
+ "loss": 0.4211,
1942
+ "step": 276
1943
+ },
1944
+ {
1945
+ "epoch": 0.7413850786216126,
1946
+ "grad_norm": 2.8922770023345947,
1947
+ "learning_rate": 7.628783485353573e-06,
1948
+ "loss": 0.4465,
1949
+ "step": 277
1950
+ },
1951
+ {
1952
+ "epoch": 0.7440615590498495,
1953
+ "grad_norm": 2.775923252105713,
1954
+ "learning_rate": 7.609614024894694e-06,
1955
+ "loss": 0.46,
1956
+ "step": 278
1957
+ },
1958
+ {
1959
+ "epoch": 0.7467380394780864,
1960
+ "grad_norm": 3.38106632232666,
1961
+ "learning_rate": 7.5903916998973745e-06,
1962
+ "loss": 0.6024,
1963
+ "step": 279
1964
+ },
1965
+ {
1966
+ "epoch": 0.7494145199063232,
1967
+ "grad_norm": 2.9495768547058105,
1968
+ "learning_rate": 7.571116899759945e-06,
1969
+ "loss": 0.4792,
1970
+ "step": 280
1971
+ },
1972
+ {
1973
+ "epoch": 0.75209100033456,
1974
+ "grad_norm": 3.18288516998291,
1975
+ "learning_rate": 7.551790014943752e-06,
1976
+ "loss": 0.4439,
1977
+ "step": 281
1978
+ },
1979
+ {
1980
+ "epoch": 0.7547674807627969,
1981
+ "grad_norm": 3.185044288635254,
1982
+ "learning_rate": 7.532411436965258e-06,
1983
+ "loss": 0.5251,
1984
+ "step": 282
1985
+ },
1986
+ {
1987
+ "epoch": 0.7574439611910337,
1988
+ "grad_norm": 2.8244149684906006,
1989
+ "learning_rate": 7.512981558388101e-06,
1990
+ "loss": 0.4443,
1991
+ "step": 283
1992
+ },
1993
+ {
1994
+ "epoch": 0.7601204416192706,
1995
+ "grad_norm": 3.172478437423706,
1996
+ "learning_rate": 7.49350077281515e-06,
1997
+ "loss": 0.5057,
1998
+ "step": 284
1999
+ },
2000
+ {
2001
+ "epoch": 0.7627969220475075,
2002
+ "grad_norm": 3.2806384563446045,
2003
+ "learning_rate": 7.473969474880527e-06,
2004
+ "loss": 0.5944,
2005
+ "step": 285
2006
+ },
2007
+ {
2008
+ "epoch": 0.7654734024757444,
2009
+ "grad_norm": 2.7359542846679688,
2010
+ "learning_rate": 7.45438806024161e-06,
2011
+ "loss": 0.4296,
2012
+ "step": 286
2013
+ },
2014
+ {
2015
+ "epoch": 0.7681498829039812,
2016
+ "grad_norm": 2.7269349098205566,
2017
+ "learning_rate": 7.4347569255710254e-06,
2018
+ "loss": 0.4004,
2019
+ "step": 287
2020
+ },
2021
+ {
2022
+ "epoch": 0.7708263633322181,
2023
+ "grad_norm": 3.779465913772583,
2024
+ "learning_rate": 7.41507646854861e-06,
2025
+ "loss": 0.5674,
2026
+ "step": 288
2027
+ },
2028
+ {
2029
+ "epoch": 0.773502843760455,
2030
+ "grad_norm": 2.7215170860290527,
2031
+ "learning_rate": 7.395347087853349e-06,
2032
+ "loss": 0.3874,
2033
+ "step": 289
2034
+ },
2035
+ {
2036
+ "epoch": 0.7761793241886918,
2037
+ "grad_norm": 4.240821838378906,
2038
+ "learning_rate": 7.375569183155306e-06,
2039
+ "loss": 0.5878,
2040
+ "step": 290
2041
+ },
2042
+ {
2043
+ "epoch": 0.7788558046169287,
2044
+ "grad_norm": 3.43664288520813,
2045
+ "learning_rate": 7.355743155107526e-06,
2046
+ "loss": 0.4482,
2047
+ "step": 291
2048
+ },
2049
+ {
2050
+ "epoch": 0.7815322850451656,
2051
+ "grad_norm": 3.580437660217285,
2052
+ "learning_rate": 7.335869405337919e-06,
2053
+ "loss": 0.5135,
2054
+ "step": 292
2055
+ },
2056
+ {
2057
+ "epoch": 0.7842087654734025,
2058
+ "grad_norm": 3.794210433959961,
2059
+ "learning_rate": 7.3159483364411175e-06,
2060
+ "loss": 0.4546,
2061
+ "step": 293
2062
+ },
2063
+ {
2064
+ "epoch": 0.7868852459016393,
2065
+ "grad_norm": 2.8526501655578613,
2066
+ "learning_rate": 7.295980351970331e-06,
2067
+ "loss": 0.416,
2068
+ "step": 294
2069
+ },
2070
+ {
2071
+ "epoch": 0.7895617263298762,
2072
+ "grad_norm": 3.249462127685547,
2073
+ "learning_rate": 7.275965856429167e-06,
2074
+ "loss": 0.4575,
2075
+ "step": 295
2076
+ },
2077
+ {
2078
+ "epoch": 0.7922382067581131,
2079
+ "grad_norm": 4.065032482147217,
2080
+ "learning_rate": 7.255905255263434e-06,
2081
+ "loss": 0.4073,
2082
+ "step": 296
2083
+ },
2084
+ {
2085
+ "epoch": 0.7949146871863499,
2086
+ "grad_norm": 2.9181621074676514,
2087
+ "learning_rate": 7.235798954852929e-06,
2088
+ "loss": 0.467,
2089
+ "step": 297
2090
+ },
2091
+ {
2092
+ "epoch": 0.7975911676145868,
2093
+ "grad_norm": 3.012425661087036,
2094
+ "learning_rate": 7.2156473625032075e-06,
2095
+ "loss": 0.5308,
2096
+ "step": 298
2097
+ },
2098
+ {
2099
+ "epoch": 0.8002676480428237,
2100
+ "grad_norm": 3.2600486278533936,
2101
+ "learning_rate": 7.195450886437334e-06,
2102
+ "loss": 0.5594,
2103
+ "step": 299
2104
+ },
2105
+ {
2106
+ "epoch": 0.8029441284710606,
2107
+ "grad_norm": 3.5234806537628174,
2108
+ "learning_rate": 7.175209935787605e-06,
2109
+ "loss": 0.4377,
2110
+ "step": 300
2111
+ },
2112
+ {
2113
+ "epoch": 0.8056206088992974,
2114
+ "grad_norm": 3.3013510704040527,
2115
+ "learning_rate": 7.154924920587269e-06,
2116
+ "loss": 0.5039,
2117
+ "step": 301
2118
+ },
2119
+ {
2120
+ "epoch": 0.8082970893275343,
2121
+ "grad_norm": 3.6095550060272217,
2122
+ "learning_rate": 7.134596251762217e-06,
2123
+ "loss": 0.5333,
2124
+ "step": 302
2125
+ },
2126
+ {
2127
+ "epoch": 0.8109735697557712,
2128
+ "grad_norm": 3.533142328262329,
2129
+ "learning_rate": 7.114224341122655e-06,
2130
+ "loss": 0.4358,
2131
+ "step": 303
2132
+ },
2133
+ {
2134
+ "epoch": 0.813650050184008,
2135
+ "grad_norm": 3.41983699798584,
2136
+ "learning_rate": 7.093809601354769e-06,
2137
+ "loss": 0.5369,
2138
+ "step": 304
2139
+ },
2140
+ {
2141
+ "epoch": 0.8163265306122449,
2142
+ "grad_norm": 2.5269010066986084,
2143
+ "learning_rate": 7.073352446012357e-06,
2144
+ "loss": 0.4014,
2145
+ "step": 305
2146
+ },
2147
+ {
2148
+ "epoch": 0.8190030110404818,
2149
+ "grad_norm": 3.3515594005584717,
2150
+ "learning_rate": 7.052853289508458e-06,
2151
+ "loss": 0.4681,
2152
+ "step": 306
2153
+ },
2154
+ {
2155
+ "epoch": 0.8216794914687187,
2156
+ "grad_norm": 3.920032501220703,
2157
+ "learning_rate": 7.03231254710695e-06,
2158
+ "loss": 0.5876,
2159
+ "step": 307
2160
+ },
2161
+ {
2162
+ "epoch": 0.8243559718969555,
2163
+ "grad_norm": 2.734036684036255,
2164
+ "learning_rate": 7.0117306349141485e-06,
2165
+ "loss": 0.4294,
2166
+ "step": 308
2167
+ },
2168
+ {
2169
+ "epoch": 0.8270324523251924,
2170
+ "grad_norm": 2.5630767345428467,
2171
+ "learning_rate": 6.991107969870363e-06,
2172
+ "loss": 0.3869,
2173
+ "step": 309
2174
+ },
2175
+ {
2176
+ "epoch": 0.8297089327534293,
2177
+ "grad_norm": 2.897214889526367,
2178
+ "learning_rate": 6.970444969741462e-06,
2179
+ "loss": 0.4756,
2180
+ "step": 310
2181
+ },
2182
+ {
2183
+ "epoch": 0.8323854131816661,
2184
+ "grad_norm": 3.645385980606079,
2185
+ "learning_rate": 6.949742053110408e-06,
2186
+ "loss": 0.465,
2187
+ "step": 311
2188
+ },
2189
+ {
2190
+ "epoch": 0.835061893609903,
2191
+ "grad_norm": 2.9441306591033936,
2192
+ "learning_rate": 6.928999639368773e-06,
2193
+ "loss": 0.4803,
2194
+ "step": 312
2195
+ },
2196
+ {
2197
+ "epoch": 0.8377383740381399,
2198
+ "grad_norm": 3.1111350059509277,
2199
+ "learning_rate": 6.908218148708248e-06,
2200
+ "loss": 0.4119,
2201
+ "step": 313
2202
+ },
2203
+ {
2204
+ "epoch": 0.8404148544663768,
2205
+ "grad_norm": 2.8598246574401855,
2206
+ "learning_rate": 6.887398002112129e-06,
2207
+ "loss": 0.5,
2208
+ "step": 314
2209
+ },
2210
+ {
2211
+ "epoch": 0.8430913348946136,
2212
+ "grad_norm": 2.703573703765869,
2213
+ "learning_rate": 6.866539621346786e-06,
2214
+ "loss": 0.3387,
2215
+ "step": 315
2216
+ },
2217
+ {
2218
+ "epoch": 0.8457678153228505,
2219
+ "grad_norm": 2.924374580383301,
2220
+ "learning_rate": 6.845643428953127e-06,
2221
+ "loss": 0.4498,
2222
+ "step": 316
2223
+ },
2224
+ {
2225
+ "epoch": 0.8484442957510874,
2226
+ "grad_norm": 3.2768142223358154,
2227
+ "learning_rate": 6.824709848238028e-06,
2228
+ "loss": 0.4514,
2229
+ "step": 317
2230
+ },
2231
+ {
2232
+ "epoch": 0.8511207761793241,
2233
+ "grad_norm": 3.422415256500244,
2234
+ "learning_rate": 6.8037393032657665e-06,
2235
+ "loss": 0.4806,
2236
+ "step": 318
2237
+ },
2238
+ {
2239
+ "epoch": 0.853797256607561,
2240
+ "grad_norm": 3.6226656436920166,
2241
+ "learning_rate": 6.782732218849425e-06,
2242
+ "loss": 0.431,
2243
+ "step": 319
2244
+ },
2245
+ {
2246
+ "epoch": 0.8564737370357979,
2247
+ "grad_norm": 3.154444932937622,
2248
+ "learning_rate": 6.761689020542288e-06,
2249
+ "loss": 0.4059,
2250
+ "step": 320
2251
+ },
2252
+ {
2253
+ "epoch": 0.8591502174640347,
2254
+ "grad_norm": 3.5488181114196777,
2255
+ "learning_rate": 6.740610134629224e-06,
2256
+ "loss": 0.5335,
2257
+ "step": 321
2258
+ },
2259
+ {
2260
+ "epoch": 0.8618266978922716,
2261
+ "grad_norm": 3.483675718307495,
2262
+ "learning_rate": 6.719495988118043e-06,
2263
+ "loss": 0.5676,
2264
+ "step": 322
2265
+ },
2266
+ {
2267
+ "epoch": 0.8645031783205085,
2268
+ "grad_norm": 3.83905291557312,
2269
+ "learning_rate": 6.698347008730854e-06,
2270
+ "loss": 0.5997,
2271
+ "step": 323
2272
+ },
2273
+ {
2274
+ "epoch": 0.8671796587487454,
2275
+ "grad_norm": 3.451472759246826,
2276
+ "learning_rate": 6.677163624895393e-06,
2277
+ "loss": 0.4285,
2278
+ "step": 324
2279
+ },
2280
+ {
2281
+ "epoch": 0.8698561391769822,
2282
+ "grad_norm": 3.9657328128814697,
2283
+ "learning_rate": 6.6559462657363525e-06,
2284
+ "loss": 0.5505,
2285
+ "step": 325
2286
+ },
2287
+ {
2288
+ "epoch": 0.8725326196052191,
2289
+ "grad_norm": 3.474752902984619,
2290
+ "learning_rate": 6.634695361066679e-06,
2291
+ "loss": 0.577,
2292
+ "step": 326
2293
+ },
2294
+ {
2295
+ "epoch": 0.875209100033456,
2296
+ "grad_norm": 3.0896995067596436,
2297
+ "learning_rate": 6.613411341378872e-06,
2298
+ "loss": 0.4052,
2299
+ "step": 327
2300
+ },
2301
+ {
2302
+ "epoch": 0.8778855804616928,
2303
+ "grad_norm": 2.541867256164551,
2304
+ "learning_rate": 6.592094637836266e-06,
2305
+ "loss": 0.3609,
2306
+ "step": 328
2307
+ },
2308
+ {
2309
+ "epoch": 0.8805620608899297,
2310
+ "grad_norm": 3.222644329071045,
2311
+ "learning_rate": 6.570745682264288e-06,
2312
+ "loss": 0.4346,
2313
+ "step": 329
2314
+ },
2315
+ {
2316
+ "epoch": 0.8832385413181666,
2317
+ "grad_norm": 2.9932363033294678,
2318
+ "learning_rate": 6.549364907141713e-06,
2319
+ "loss": 0.3862,
2320
+ "step": 330
2321
+ },
2322
+ {
2323
+ "epoch": 0.8859150217464035,
2324
+ "grad_norm": 2.610246181488037,
2325
+ "learning_rate": 6.527952745591911e-06,
2326
+ "loss": 0.3075,
2327
+ "step": 331
2328
+ },
2329
+ {
2330
+ "epoch": 0.8885915021746403,
2331
+ "grad_norm": 3.310206174850464,
2332
+ "learning_rate": 6.506509631374056e-06,
2333
+ "loss": 0.4288,
2334
+ "step": 332
2335
+ },
2336
+ {
2337
+ "epoch": 0.8912679826028772,
2338
+ "grad_norm": 3.0393242835998535,
2339
+ "learning_rate": 6.485035998874356e-06,
2340
+ "loss": 0.5224,
2341
+ "step": 333
2342
+ },
2343
+ {
2344
+ "epoch": 0.8939444630311141,
2345
+ "grad_norm": 3.029529571533203,
2346
+ "learning_rate": 6.4635322830972465e-06,
2347
+ "loss": 0.5228,
2348
+ "step": 334
2349
+ },
2350
+ {
2351
+ "epoch": 0.896620943459351,
2352
+ "grad_norm": 2.9739904403686523,
2353
+ "learning_rate": 6.441998919656575e-06,
2354
+ "loss": 0.4464,
2355
+ "step": 335
2356
+ },
2357
+ {
2358
+ "epoch": 0.8992974238875878,
2359
+ "grad_norm": 3.4808287620544434,
2360
+ "learning_rate": 6.420436344766781e-06,
2361
+ "loss": 0.4787,
2362
+ "step": 336
2363
+ },
2364
+ {
2365
+ "epoch": 0.9019739043158247,
2366
+ "grad_norm": 3.4027199745178223,
2367
+ "learning_rate": 6.398844995234057e-06,
2368
+ "loss": 0.4964,
2369
+ "step": 337
2370
+ },
2371
+ {
2372
+ "epoch": 0.9046503847440616,
2373
+ "grad_norm": 3.5456416606903076,
2374
+ "learning_rate": 6.377225308447503e-06,
2375
+ "loss": 0.5533,
2376
+ "step": 338
2377
+ },
2378
+ {
2379
+ "epoch": 0.9073268651722984,
2380
+ "grad_norm": 3.416761875152588,
2381
+ "learning_rate": 6.355577722370264e-06,
2382
+ "loss": 0.4679,
2383
+ "step": 339
2384
+ },
2385
+ {
2386
+ "epoch": 0.9100033456005353,
2387
+ "grad_norm": 3.634335994720459,
2388
+ "learning_rate": 6.333902675530657e-06,
2389
+ "loss": 0.4683,
2390
+ "step": 340
2391
+ },
2392
+ {
2393
+ "epoch": 0.9126798260287722,
2394
+ "grad_norm": 2.5649242401123047,
2395
+ "learning_rate": 6.312200607013287e-06,
2396
+ "loss": 0.3649,
2397
+ "step": 341
2398
+ },
2399
+ {
2400
+ "epoch": 0.915356306457009,
2401
+ "grad_norm": 3.457108736038208,
2402
+ "learning_rate": 6.2904719564501545e-06,
2403
+ "loss": 0.559,
2404
+ "step": 342
2405
+ },
2406
+ {
2407
+ "epoch": 0.9180327868852459,
2408
+ "grad_norm": 3.2710628509521484,
2409
+ "learning_rate": 6.268717164011751e-06,
2410
+ "loss": 0.4526,
2411
+ "step": 343
2412
+ },
2413
+ {
2414
+ "epoch": 0.9207092673134828,
2415
+ "grad_norm": 2.584167003631592,
2416
+ "learning_rate": 6.246936670398136e-06,
2417
+ "loss": 0.3223,
2418
+ "step": 344
2419
+ },
2420
+ {
2421
+ "epoch": 0.9233857477417197,
2422
+ "grad_norm": 3.2143352031707764,
2423
+ "learning_rate": 6.225130916830017e-06,
2424
+ "loss": 0.3996,
2425
+ "step": 345
2426
+ },
2427
+ {
2428
+ "epoch": 0.9260622281699565,
2429
+ "grad_norm": 2.900538206100464,
2430
+ "learning_rate": 6.203300345039804e-06,
2431
+ "loss": 0.4241,
2432
+ "step": 346
2433
+ },
2434
+ {
2435
+ "epoch": 0.9287387085981934,
2436
+ "grad_norm": 2.566956043243408,
2437
+ "learning_rate": 6.181445397262671e-06,
2438
+ "loss": 0.3754,
2439
+ "step": 347
2440
+ },
2441
+ {
2442
+ "epoch": 0.9314151890264303,
2443
+ "grad_norm": 3.051973819732666,
2444
+ "learning_rate": 6.159566516227582e-06,
2445
+ "loss": 0.464,
2446
+ "step": 348
2447
+ },
2448
+ {
2449
+ "epoch": 0.9340916694546672,
2450
+ "grad_norm": 2.648303985595703,
2451
+ "learning_rate": 6.137664145148339e-06,
2452
+ "loss": 0.3669,
2453
+ "step": 349
2454
+ },
2455
+ {
2456
+ "epoch": 0.936768149882904,
2457
+ "grad_norm": 3.3887152671813965,
2458
+ "learning_rate": 6.115738727714593e-06,
2459
+ "loss": 0.4347,
2460
+ "step": 350
2461
+ },
2462
+ {
2463
+ "epoch": 0.9394446303111409,
2464
+ "grad_norm": 2.8967878818511963,
2465
+ "learning_rate": 6.093790708082861e-06,
2466
+ "loss": 0.4126,
2467
+ "step": 351
2468
+ },
2469
+ {
2470
+ "epoch": 0.9421211107393778,
2471
+ "grad_norm": 3.301456928253174,
2472
+ "learning_rate": 6.071820530867524e-06,
2473
+ "loss": 0.4932,
2474
+ "step": 352
2475
+ },
2476
+ {
2477
+ "epoch": 0.9447975911676146,
2478
+ "grad_norm": 2.467073678970337,
2479
+ "learning_rate": 6.0498286411318255e-06,
2480
+ "loss": 0.3714,
2481
+ "step": 353
2482
+ },
2483
+ {
2484
+ "epoch": 0.9474740715958515,
2485
+ "grad_norm": 2.737948417663574,
2486
+ "learning_rate": 6.027815484378848e-06,
2487
+ "loss": 0.3832,
2488
+ "step": 354
2489
+ },
2490
+ {
2491
+ "epoch": 0.9501505520240883,
2492
+ "grad_norm": 3.305079221725464,
2493
+ "learning_rate": 6.005781506542498e-06,
2494
+ "loss": 0.3813,
2495
+ "step": 355
2496
+ },
2497
+ {
2498
+ "epoch": 0.9528270324523251,
2499
+ "grad_norm": 3.1508371829986572,
2500
+ "learning_rate": 5.983727153978467e-06,
2501
+ "loss": 0.5005,
2502
+ "step": 356
2503
+ },
2504
+ {
2505
+ "epoch": 0.955503512880562,
2506
+ "grad_norm": 3.4938628673553467,
2507
+ "learning_rate": 5.961652873455186e-06,
2508
+ "loss": 0.4586,
2509
+ "step": 357
2510
+ },
2511
+ {
2512
+ "epoch": 0.9581799933087989,
2513
+ "grad_norm": 2.7031075954437256,
2514
+ "learning_rate": 5.939559112144781e-06,
2515
+ "loss": 0.3995,
2516
+ "step": 358
2517
+ },
2518
+ {
2519
+ "epoch": 0.9608564737370358,
2520
+ "grad_norm": 3.1249899864196777,
2521
+ "learning_rate": 5.917446317614012e-06,
2522
+ "loss": 0.4738,
2523
+ "step": 359
2524
+ },
2525
+ {
2526
+ "epoch": 0.9635329541652726,
2527
+ "grad_norm": 3.1881392002105713,
2528
+ "learning_rate": 5.895314937815206e-06,
2529
+ "loss": 0.4234,
2530
+ "step": 360
2531
+ },
2532
+ {
2533
+ "epoch": 0.9662094345935095,
2534
+ "grad_norm": 3.020379066467285,
2535
+ "learning_rate": 5.873165421077186e-06,
2536
+ "loss": 0.347,
2537
+ "step": 361
2538
+ },
2539
+ {
2540
+ "epoch": 0.9688859150217464,
2541
+ "grad_norm": 2.737119436264038,
2542
+ "learning_rate": 5.850998216096181e-06,
2543
+ "loss": 0.3805,
2544
+ "step": 362
2545
+ },
2546
+ {
2547
+ "epoch": 0.9715623954499832,
2548
+ "grad_norm": 3.5535435676574707,
2549
+ "learning_rate": 5.828813771926746e-06,
2550
+ "loss": 0.5056,
2551
+ "step": 363
2552
+ },
2553
+ {
2554
+ "epoch": 0.9742388758782201,
2555
+ "grad_norm": 3.3006703853607178,
2556
+ "learning_rate": 5.806612537972658e-06,
2557
+ "loss": 0.3342,
2558
+ "step": 364
2559
+ },
2560
+ {
2561
+ "epoch": 0.976915356306457,
2562
+ "grad_norm": 3.176393508911133,
2563
+ "learning_rate": 5.784394963977815e-06,
2564
+ "loss": 0.4834,
2565
+ "step": 365
2566
+ },
2567
+ {
2568
+ "epoch": 0.9795918367346939,
2569
+ "grad_norm": 3.3865625858306885,
2570
+ "learning_rate": 5.762161500017128e-06,
2571
+ "loss": 0.4573,
2572
+ "step": 366
2573
+ },
2574
+ {
2575
+ "epoch": 0.9822683171629307,
2576
+ "grad_norm": 2.942143678665161,
2577
+ "learning_rate": 5.739912596487396e-06,
2578
+ "loss": 0.4826,
2579
+ "step": 367
2580
+ },
2581
+ {
2582
+ "epoch": 0.9849447975911676,
2583
+ "grad_norm": 2.834310531616211,
2584
+ "learning_rate": 5.717648704098191e-06,
2585
+ "loss": 0.444,
2586
+ "step": 368
2587
+ },
2588
+ {
2589
+ "epoch": 0.9876212780194045,
2590
+ "grad_norm": 2.967898368835449,
2591
+ "learning_rate": 5.6953702738627215e-06,
2592
+ "loss": 0.3811,
2593
+ "step": 369
2594
+ },
2595
+ {
2596
+ "epoch": 0.9902977584476413,
2597
+ "grad_norm": 2.998699426651001,
2598
+ "learning_rate": 5.6730777570887e-06,
2599
+ "loss": 0.4694,
2600
+ "step": 370
2601
+ },
2602
+ {
2603
+ "epoch": 0.9929742388758782,
2604
+ "grad_norm": 3.34025502204895,
2605
+ "learning_rate": 5.6507716053691916e-06,
2606
+ "loss": 0.5136,
2607
+ "step": 371
2608
+ },
2609
+ {
2610
+ "epoch": 0.9956507193041151,
2611
+ "grad_norm": 2.5834507942199707,
2612
+ "learning_rate": 5.628452270573483e-06,
2613
+ "loss": 0.3702,
2614
+ "step": 372
2615
+ },
2616
+ {
2617
+ "epoch": 0.998327199732352,
2618
+ "grad_norm": 2.72577166557312,
2619
+ "learning_rate": 5.6061202048379125e-06,
2620
+ "loss": 0.3757,
2621
+ "step": 373
2622
+ },
2623
+ {
2624
+ "epoch": 1.0,
2625
+ "grad_norm": 3.7107036113739014,
2626
+ "learning_rate": 5.583775860556717e-06,
2627
+ "loss": 0.3578,
2628
+ "step": 374
2629
+ },
2630
+ {
2631
+ "epoch": 1.0026764804282369,
2632
+ "grad_norm": 2.568798542022705,
2633
+ "learning_rate": 5.561419690372869e-06,
2634
+ "loss": 0.1397,
2635
+ "step": 375
2636
+ },
2637
+ {
2638
+ "epoch": 1.0053529608564737,
2639
+ "grad_norm": 1.9347107410430908,
2640
+ "learning_rate": 5.539052147168903e-06,
2641
+ "loss": 0.2252,
2642
+ "step": 376
2643
+ },
2644
+ {
2645
+ "epoch": 1.0080294412847106,
2646
+ "grad_norm": 2.0152370929718018,
2647
+ "learning_rate": 5.516673684057747e-06,
2648
+ "loss": 0.2206,
2649
+ "step": 377
2650
+ },
2651
+ {
2652
+ "epoch": 1.0107059217129475,
2653
+ "grad_norm": 2.0257060527801514,
2654
+ "learning_rate": 5.494284754373538e-06,
2655
+ "loss": 0.1776,
2656
+ "step": 378
2657
+ },
2658
+ {
2659
+ "epoch": 1.0133824021411844,
2660
+ "grad_norm": 2.107743263244629,
2661
+ "learning_rate": 5.471885811662442e-06,
2662
+ "loss": 0.1799,
2663
+ "step": 379
2664
+ },
2665
+ {
2666
+ "epoch": 1.0160588825694212,
2667
+ "grad_norm": 2.221683979034424,
2668
+ "learning_rate": 5.449477309673462e-06,
2669
+ "loss": 0.1642,
2670
+ "step": 380
2671
+ },
2672
+ {
2673
+ "epoch": 1.018735362997658,
2674
+ "grad_norm": 2.8024303913116455,
2675
+ "learning_rate": 5.427059702349255e-06,
2676
+ "loss": 0.1712,
2677
+ "step": 381
2678
+ },
2679
+ {
2680
+ "epoch": 1.021411843425895,
2681
+ "grad_norm": 2.7197983264923096,
2682
+ "learning_rate": 5.4046334438169245e-06,
2683
+ "loss": 0.2057,
2684
+ "step": 382
2685
+ },
2686
+ {
2687
+ "epoch": 1.0240883238541318,
2688
+ "grad_norm": 1.90606689453125,
2689
+ "learning_rate": 5.382198988378829e-06,
2690
+ "loss": 0.1793,
2691
+ "step": 383
2692
+ },
2693
+ {
2694
+ "epoch": 1.0267648042823687,
2695
+ "grad_norm": 2.782726764678955,
2696
+ "learning_rate": 5.359756790503376e-06,
2697
+ "loss": 0.2036,
2698
+ "step": 384
2699
+ },
2700
+ {
2701
+ "epoch": 1.0294412847106056,
2702
+ "grad_norm": 2.3498716354370117,
2703
+ "learning_rate": 5.337307304815817e-06,
2704
+ "loss": 0.2134,
2705
+ "step": 385
2706
+ },
2707
+ {
2708
+ "epoch": 1.0321177651388425,
2709
+ "grad_norm": 3.6276447772979736,
2710
+ "learning_rate": 5.31485098608904e-06,
2711
+ "loss": 0.1908,
2712
+ "step": 386
2713
+ },
2714
+ {
2715
+ "epoch": 1.0347942455670793,
2716
+ "grad_norm": 3.2593178749084473,
2717
+ "learning_rate": 5.292388289234349e-06,
2718
+ "loss": 0.2326,
2719
+ "step": 387
2720
+ },
2721
+ {
2722
+ "epoch": 1.0374707259953162,
2723
+ "grad_norm": 3.1643850803375244,
2724
+ "learning_rate": 5.2699196692922546e-06,
2725
+ "loss": 0.2165,
2726
+ "step": 388
2727
+ },
2728
+ {
2729
+ "epoch": 1.040147206423553,
2730
+ "grad_norm": 2.73394513130188,
2731
+ "learning_rate": 5.247445581423257e-06,
2732
+ "loss": 0.1906,
2733
+ "step": 389
2734
+ },
2735
+ {
2736
+ "epoch": 1.04282368685179,
2737
+ "grad_norm": 3.0761959552764893,
2738
+ "learning_rate": 5.224966480898624e-06,
2739
+ "loss": 0.1903,
2740
+ "step": 390
2741
+ },
2742
+ {
2743
+ "epoch": 1.0455001672800268,
2744
+ "grad_norm": 3.4978442192077637,
2745
+ "learning_rate": 5.202482823091165e-06,
2746
+ "loss": 0.211,
2747
+ "step": 391
2748
+ },
2749
+ {
2750
+ "epoch": 1.0481766477082637,
2751
+ "grad_norm": 2.406022548675537,
2752
+ "learning_rate": 5.179995063466011e-06,
2753
+ "loss": 0.2095,
2754
+ "step": 392
2755
+ },
2756
+ {
2757
+ "epoch": 1.0508531281365006,
2758
+ "grad_norm": 3.4254419803619385,
2759
+ "learning_rate": 5.157503657571386e-06,
2760
+ "loss": 0.2339,
2761
+ "step": 393
2762
+ },
2763
+ {
2764
+ "epoch": 1.0535296085647374,
2765
+ "grad_norm": 3.1183996200561523,
2766
+ "learning_rate": 5.1350090610293765e-06,
2767
+ "loss": 0.1881,
2768
+ "step": 394
2769
+ },
2770
+ {
2771
+ "epoch": 1.0562060889929743,
2772
+ "grad_norm": 2.5185012817382812,
2773
+ "learning_rate": 5.112511729526708e-06,
2774
+ "loss": 0.1261,
2775
+ "step": 395
2776
+ },
2777
+ {
2778
+ "epoch": 1.0588825694212112,
2779
+ "grad_norm": 2.3276853561401367,
2780
+ "learning_rate": 5.090012118805505e-06,
2781
+ "loss": 0.1423,
2782
+ "step": 396
2783
+ },
2784
+ {
2785
+ "epoch": 1.061559049849448,
2786
+ "grad_norm": 2.8159985542297363,
2787
+ "learning_rate": 5.067510684654069e-06,
2788
+ "loss": 0.1786,
2789
+ "step": 397
2790
+ },
2791
+ {
2792
+ "epoch": 1.064235530277685,
2793
+ "grad_norm": 2.6511971950531006,
2794
+ "learning_rate": 5.0450078828976326e-06,
2795
+ "loss": 0.2184,
2796
+ "step": 398
2797
+ },
2798
+ {
2799
+ "epoch": 1.0669120107059218,
2800
+ "grad_norm": 2.8638663291931152,
2801
+ "learning_rate": 5.02250416938914e-06,
2802
+ "loss": 0.1745,
2803
+ "step": 399
2804
+ },
2805
+ {
2806
+ "epoch": 1.0695884911341587,
2807
+ "grad_norm": 2.6509993076324463,
2808
+ "learning_rate": 5e-06,
2809
+ "loss": 0.1678,
2810
+ "step": 400
2811
+ },
2812
+ {
2813
+ "epoch": 1.0722649715623955,
2814
+ "grad_norm": 2.952925443649292,
2815
+ "learning_rate": 4.977495830610862e-06,
2816
+ "loss": 0.1931,
2817
+ "step": 401
2818
+ },
2819
+ {
2820
+ "epoch": 1.0749414519906324,
2821
+ "grad_norm": 3.211611270904541,
2822
+ "learning_rate": 4.954992117102369e-06,
2823
+ "loss": 0.2434,
2824
+ "step": 402
2825
+ },
2826
+ {
2827
+ "epoch": 1.0776179324188693,
2828
+ "grad_norm": 2.5486137866973877,
2829
+ "learning_rate": 4.932489315345933e-06,
2830
+ "loss": 0.1647,
2831
+ "step": 403
2832
+ },
2833
+ {
2834
+ "epoch": 1.0802944128471061,
2835
+ "grad_norm": 2.394988775253296,
2836
+ "learning_rate": 4.9099878811944965e-06,
2837
+ "loss": 0.1641,
2838
+ "step": 404
2839
+ },
2840
+ {
2841
+ "epoch": 1.082970893275343,
2842
+ "grad_norm": 2.4356367588043213,
2843
+ "learning_rate": 4.887488270473294e-06,
2844
+ "loss": 0.1394,
2845
+ "step": 405
2846
+ },
2847
+ {
2848
+ "epoch": 1.0856473737035799,
2849
+ "grad_norm": 3.3309195041656494,
2850
+ "learning_rate": 4.864990938970624e-06,
2851
+ "loss": 0.2721,
2852
+ "step": 406
2853
+ },
2854
+ {
2855
+ "epoch": 1.0883238541318168,
2856
+ "grad_norm": 2.490863561630249,
2857
+ "learning_rate": 4.842496342428616e-06,
2858
+ "loss": 0.1652,
2859
+ "step": 407
2860
+ },
2861
+ {
2862
+ "epoch": 1.0910003345600536,
2863
+ "grad_norm": 2.8039042949676514,
2864
+ "learning_rate": 4.8200049365339905e-06,
2865
+ "loss": 0.1588,
2866
+ "step": 408
2867
+ },
2868
+ {
2869
+ "epoch": 1.0936768149882905,
2870
+ "grad_norm": 2.8676836490631104,
2871
+ "learning_rate": 4.7975171769088366e-06,
2872
+ "loss": 0.1829,
2873
+ "step": 409
2874
+ },
2875
+ {
2876
+ "epoch": 1.0963532954165274,
2877
+ "grad_norm": 2.506415843963623,
2878
+ "learning_rate": 4.775033519101378e-06,
2879
+ "loss": 0.1836,
2880
+ "step": 410
2881
+ },
2882
+ {
2883
+ "epoch": 1.0990297758447642,
2884
+ "grad_norm": 2.051229238510132,
2885
+ "learning_rate": 4.752554418576744e-06,
2886
+ "loss": 0.1317,
2887
+ "step": 411
2888
+ },
2889
+ {
2890
+ "epoch": 1.1017062562730011,
2891
+ "grad_norm": 2.053317070007324,
2892
+ "learning_rate": 4.730080330707748e-06,
2893
+ "loss": 0.1531,
2894
+ "step": 412
2895
+ },
2896
+ {
2897
+ "epoch": 1.104382736701238,
2898
+ "grad_norm": 2.306492805480957,
2899
+ "learning_rate": 4.707611710765654e-06,
2900
+ "loss": 0.1874,
2901
+ "step": 413
2902
+ },
2903
+ {
2904
+ "epoch": 1.1070592171294746,
2905
+ "grad_norm": 2.638895273208618,
2906
+ "learning_rate": 4.685149013910962e-06,
2907
+ "loss": 0.1927,
2908
+ "step": 414
2909
+ },
2910
+ {
2911
+ "epoch": 1.1097356975577115,
2912
+ "grad_norm": 2.5232789516448975,
2913
+ "learning_rate": 4.662692695184184e-06,
2914
+ "loss": 0.1728,
2915
+ "step": 415
2916
+ },
2917
+ {
2918
+ "epoch": 1.1124121779859484,
2919
+ "grad_norm": 2.99980092048645,
2920
+ "learning_rate": 4.640243209496627e-06,
2921
+ "loss": 0.1639,
2922
+ "step": 416
2923
+ },
2924
+ {
2925
+ "epoch": 1.1150886584141853,
2926
+ "grad_norm": 2.612894058227539,
2927
+ "learning_rate": 4.617801011621175e-06,
2928
+ "loss": 0.2109,
2929
+ "step": 417
2930
+ },
2931
+ {
2932
+ "epoch": 1.1177651388424221,
2933
+ "grad_norm": 2.540539264678955,
2934
+ "learning_rate": 4.595366556183079e-06,
2935
+ "loss": 0.1653,
2936
+ "step": 418
2937
+ },
2938
+ {
2939
+ "epoch": 1.120441619270659,
2940
+ "grad_norm": 2.789428949356079,
2941
+ "learning_rate": 4.572940297650747e-06,
2942
+ "loss": 0.1752,
2943
+ "step": 419
2944
+ },
2945
+ {
2946
+ "epoch": 1.1231180996988959,
2947
+ "grad_norm": 2.719423770904541,
2948
+ "learning_rate": 4.550522690326538e-06,
2949
+ "loss": 0.199,
2950
+ "step": 420
2951
+ },
2952
+ {
2953
+ "epoch": 1.1257945801271327,
2954
+ "grad_norm": 2.775261402130127,
2955
+ "learning_rate": 4.528114188337559e-06,
2956
+ "loss": 0.2183,
2957
+ "step": 421
2958
+ },
2959
+ {
2960
+ "epoch": 1.1284710605553696,
2961
+ "grad_norm": 2.4771995544433594,
2962
+ "learning_rate": 4.505715245626462e-06,
2963
+ "loss": 0.1334,
2964
+ "step": 422
2965
+ },
2966
+ {
2967
+ "epoch": 1.1311475409836065,
2968
+ "grad_norm": 2.7767140865325928,
2969
+ "learning_rate": 4.483326315942253e-06,
2970
+ "loss": 0.1777,
2971
+ "step": 423
2972
+ },
2973
+ {
2974
+ "epoch": 1.1338240214118434,
2975
+ "grad_norm": 2.205811023712158,
2976
+ "learning_rate": 4.460947852831097e-06,
2977
+ "loss": 0.1757,
2978
+ "step": 424
2979
+ },
2980
+ {
2981
+ "epoch": 1.1365005018400802,
2982
+ "grad_norm": 2.6441826820373535,
2983
+ "learning_rate": 4.438580309627132e-06,
2984
+ "loss": 0.1278,
2985
+ "step": 425
2986
+ },
2987
+ {
2988
+ "epoch": 1.139176982268317,
2989
+ "grad_norm": 2.503948211669922,
2990
+ "learning_rate": 4.4162241394432834e-06,
2991
+ "loss": 0.1089,
2992
+ "step": 426
2993
+ },
2994
+ {
2995
+ "epoch": 1.141853462696554,
2996
+ "grad_norm": 2.324817180633545,
2997
+ "learning_rate": 4.393879795162088e-06,
2998
+ "loss": 0.1368,
2999
+ "step": 427
3000
+ },
3001
+ {
3002
+ "epoch": 1.1445299431247908,
3003
+ "grad_norm": 2.718444585800171,
3004
+ "learning_rate": 4.371547729426517e-06,
3005
+ "loss": 0.2321,
3006
+ "step": 428
3007
+ },
3008
+ {
3009
+ "epoch": 1.1472064235530277,
3010
+ "grad_norm": 2.5034608840942383,
3011
+ "learning_rate": 4.349228394630808e-06,
3012
+ "loss": 0.1682,
3013
+ "step": 429
3014
+ },
3015
+ {
3016
+ "epoch": 1.1498829039812646,
3017
+ "grad_norm": 2.102431297302246,
3018
+ "learning_rate": 4.326922242911302e-06,
3019
+ "loss": 0.1457,
3020
+ "step": 430
3021
+ },
3022
+ {
3023
+ "epoch": 1.1525593844095015,
3024
+ "grad_norm": 2.500199317932129,
3025
+ "learning_rate": 4.304629726137279e-06,
3026
+ "loss": 0.1755,
3027
+ "step": 431
3028
+ },
3029
+ {
3030
+ "epoch": 1.1552358648377383,
3031
+ "grad_norm": 2.7158353328704834,
3032
+ "learning_rate": 4.28235129590181e-06,
3033
+ "loss": 0.2289,
3034
+ "step": 432
3035
+ },
3036
+ {
3037
+ "epoch": 1.1579123452659752,
3038
+ "grad_norm": 2.2673025131225586,
3039
+ "learning_rate": 4.260087403512605e-06,
3040
+ "loss": 0.1396,
3041
+ "step": 433
3042
+ },
3043
+ {
3044
+ "epoch": 1.160588825694212,
3045
+ "grad_norm": 2.1406142711639404,
3046
+ "learning_rate": 4.237838499982874e-06,
3047
+ "loss": 0.1609,
3048
+ "step": 434
3049
+ },
3050
+ {
3051
+ "epoch": 1.163265306122449,
3052
+ "grad_norm": 2.2974045276641846,
3053
+ "learning_rate": 4.2156050360221855e-06,
3054
+ "loss": 0.1582,
3055
+ "step": 435
3056
+ },
3057
+ {
3058
+ "epoch": 1.1659417865506858,
3059
+ "grad_norm": 2.915152072906494,
3060
+ "learning_rate": 4.193387462027343e-06,
3061
+ "loss": 0.1921,
3062
+ "step": 436
3063
+ },
3064
+ {
3065
+ "epoch": 1.1686182669789227,
3066
+ "grad_norm": 2.5356252193450928,
3067
+ "learning_rate": 4.171186228073256e-06,
3068
+ "loss": 0.1505,
3069
+ "step": 437
3070
+ },
3071
+ {
3072
+ "epoch": 1.1712947474071596,
3073
+ "grad_norm": 1.8980985879898071,
3074
+ "learning_rate": 4.14900178390382e-06,
3075
+ "loss": 0.1434,
3076
+ "step": 438
3077
+ },
3078
+ {
3079
+ "epoch": 1.1739712278353964,
3080
+ "grad_norm": 2.4500439167022705,
3081
+ "learning_rate": 4.126834578922816e-06,
3082
+ "loss": 0.1954,
3083
+ "step": 439
3084
+ },
3085
+ {
3086
+ "epoch": 1.1766477082636333,
3087
+ "grad_norm": 3.3117830753326416,
3088
+ "learning_rate": 4.104685062184795e-06,
3089
+ "loss": 0.2066,
3090
+ "step": 440
3091
+ },
3092
+ {
3093
+ "epoch": 1.1793241886918702,
3094
+ "grad_norm": 2.445661783218384,
3095
+ "learning_rate": 4.0825536823859895e-06,
3096
+ "loss": 0.1466,
3097
+ "step": 441
3098
+ },
3099
+ {
3100
+ "epoch": 1.182000669120107,
3101
+ "grad_norm": 2.45849609375,
3102
+ "learning_rate": 4.06044088785522e-06,
3103
+ "loss": 0.1569,
3104
+ "step": 442
3105
+ },
3106
+ {
3107
+ "epoch": 1.184677149548344,
3108
+ "grad_norm": 2.892700672149658,
3109
+ "learning_rate": 4.038347126544816e-06,
3110
+ "loss": 0.2013,
3111
+ "step": 443
3112
+ },
3113
+ {
3114
+ "epoch": 1.1873536299765808,
3115
+ "grad_norm": 2.0897271633148193,
3116
+ "learning_rate": 4.016272846021534e-06,
3117
+ "loss": 0.1392,
3118
+ "step": 444
3119
+ },
3120
+ {
3121
+ "epoch": 1.1900301104048177,
3122
+ "grad_norm": 3.540703773498535,
3123
+ "learning_rate": 3.994218493457503e-06,
3124
+ "loss": 0.2104,
3125
+ "step": 445
3126
+ },
3127
+ {
3128
+ "epoch": 1.1927065908330545,
3129
+ "grad_norm": 2.646878480911255,
3130
+ "learning_rate": 3.9721845156211535e-06,
3131
+ "loss": 0.147,
3132
+ "step": 446
3133
+ },
3134
+ {
3135
+ "epoch": 1.1953830712612914,
3136
+ "grad_norm": 2.406614065170288,
3137
+ "learning_rate": 3.950171358868177e-06,
3138
+ "loss": 0.1899,
3139
+ "step": 447
3140
+ },
3141
+ {
3142
+ "epoch": 1.1980595516895283,
3143
+ "grad_norm": 3.0973734855651855,
3144
+ "learning_rate": 3.928179469132477e-06,
3145
+ "loss": 0.1661,
3146
+ "step": 448
3147
+ },
3148
+ {
3149
+ "epoch": 1.2007360321177651,
3150
+ "grad_norm": 2.9602293968200684,
3151
+ "learning_rate": 3.906209291917141e-06,
3152
+ "loss": 0.1846,
3153
+ "step": 449
3154
+ },
3155
+ {
3156
+ "epoch": 1.203412512546002,
3157
+ "grad_norm": 2.1617279052734375,
3158
+ "learning_rate": 3.884261272285409e-06,
3159
+ "loss": 0.1459,
3160
+ "step": 450
3161
+ },
3162
+ {
3163
+ "epoch": 1.2060889929742389,
3164
+ "grad_norm": 2.4841747283935547,
3165
+ "learning_rate": 3.862335854851664e-06,
3166
+ "loss": 0.1684,
3167
+ "step": 451
3168
+ },
3169
+ {
3170
+ "epoch": 1.2087654734024758,
3171
+ "grad_norm": 2.8648061752319336,
3172
+ "learning_rate": 3.8404334837724205e-06,
3173
+ "loss": 0.1734,
3174
+ "step": 452
3175
+ },
3176
+ {
3177
+ "epoch": 1.2114419538307126,
3178
+ "grad_norm": 2.0528404712677,
3179
+ "learning_rate": 3.8185546027373325e-06,
3180
+ "loss": 0.1459,
3181
+ "step": 453
3182
+ },
3183
+ {
3184
+ "epoch": 1.2141184342589495,
3185
+ "grad_norm": 2.765840530395508,
3186
+ "learning_rate": 3.7966996549601968e-06,
3187
+ "loss": 0.2011,
3188
+ "step": 454
3189
+ },
3190
+ {
3191
+ "epoch": 1.2167949146871864,
3192
+ "grad_norm": 2.8985862731933594,
3193
+ "learning_rate": 3.7748690831699858e-06,
3194
+ "loss": 0.1739,
3195
+ "step": 455
3196
+ },
3197
+ {
3198
+ "epoch": 1.2194713951154232,
3199
+ "grad_norm": 2.5325710773468018,
3200
+ "learning_rate": 3.7530633296018664e-06,
3201
+ "loss": 0.1731,
3202
+ "step": 456
3203
+ },
3204
+ {
3205
+ "epoch": 1.2221478755436601,
3206
+ "grad_norm": 2.9402241706848145,
3207
+ "learning_rate": 3.731282835988252e-06,
3208
+ "loss": 0.2061,
3209
+ "step": 457
3210
+ },
3211
+ {
3212
+ "epoch": 1.224824355971897,
3213
+ "grad_norm": 2.370866537094116,
3214
+ "learning_rate": 3.7095280435498476e-06,
3215
+ "loss": 0.1216,
3216
+ "step": 458
3217
+ },
3218
+ {
3219
+ "epoch": 1.2275008364001339,
3220
+ "grad_norm": 2.7907772064208984,
3221
+ "learning_rate": 3.6877993929867146e-06,
3222
+ "loss": 0.2426,
3223
+ "step": 459
3224
+ },
3225
+ {
3226
+ "epoch": 1.2301773168283707,
3227
+ "grad_norm": 3.558084011077881,
3228
+ "learning_rate": 3.6660973244693443e-06,
3229
+ "loss": 0.2267,
3230
+ "step": 460
3231
+ },
3232
+ {
3233
+ "epoch": 1.2328537972566076,
3234
+ "grad_norm": 2.1622250080108643,
3235
+ "learning_rate": 3.6444222776297356e-06,
3236
+ "loss": 0.1365,
3237
+ "step": 461
3238
+ },
3239
+ {
3240
+ "epoch": 1.2355302776848445,
3241
+ "grad_norm": 2.218204975128174,
3242
+ "learning_rate": 3.6227746915524964e-06,
3243
+ "loss": 0.1506,
3244
+ "step": 462
3245
+ },
3246
+ {
3247
+ "epoch": 1.2382067581130813,
3248
+ "grad_norm": 2.30646014213562,
3249
+ "learning_rate": 3.601155004765943e-06,
3250
+ "loss": 0.1739,
3251
+ "step": 463
3252
+ },
3253
+ {
3254
+ "epoch": 1.2408832385413182,
3255
+ "grad_norm": 2.6808791160583496,
3256
+ "learning_rate": 3.5795636552332203e-06,
3257
+ "loss": 0.2019,
3258
+ "step": 464
3259
+ },
3260
+ {
3261
+ "epoch": 1.243559718969555,
3262
+ "grad_norm": 2.717808485031128,
3263
+ "learning_rate": 3.5580010803434254e-06,
3264
+ "loss": 0.1894,
3265
+ "step": 465
3266
+ },
3267
+ {
3268
+ "epoch": 1.246236199397792,
3269
+ "grad_norm": 2.5723161697387695,
3270
+ "learning_rate": 3.536467716902754e-06,
3271
+ "loss": 0.1845,
3272
+ "step": 466
3273
+ },
3274
+ {
3275
+ "epoch": 1.2489126798260288,
3276
+ "grad_norm": 2.140838861465454,
3277
+ "learning_rate": 3.5149640011256438e-06,
3278
+ "loss": 0.1652,
3279
+ "step": 467
3280
+ },
3281
+ {
3282
+ "epoch": 1.2515891602542657,
3283
+ "grad_norm": 2.549647569656372,
3284
+ "learning_rate": 3.4934903686259445e-06,
3285
+ "loss": 0.2027,
3286
+ "step": 468
3287
+ },
3288
+ {
3289
+ "epoch": 1.2542656406825026,
3290
+ "grad_norm": 2.420671224594116,
3291
+ "learning_rate": 3.472047254408091e-06,
3292
+ "loss": 0.2089,
3293
+ "step": 469
3294
+ },
3295
+ {
3296
+ "epoch": 1.2569421211107394,
3297
+ "grad_norm": 2.4578332901000977,
3298
+ "learning_rate": 3.4506350928582878e-06,
3299
+ "loss": 0.164,
3300
+ "step": 470
3301
+ },
3302
+ {
3303
+ "epoch": 1.2596186015389763,
3304
+ "grad_norm": 2.741800308227539,
3305
+ "learning_rate": 3.429254317735714e-06,
3306
+ "loss": 0.172,
3307
+ "step": 471
3308
+ },
3309
+ {
3310
+ "epoch": 1.2622950819672132,
3311
+ "grad_norm": 2.6256773471832275,
3312
+ "learning_rate": 3.4079053621637346e-06,
3313
+ "loss": 0.214,
3314
+ "step": 472
3315
+ },
3316
+ {
3317
+ "epoch": 1.26497156239545,
3318
+ "grad_norm": 2.5597918033599854,
3319
+ "learning_rate": 3.3865886586211285e-06,
3320
+ "loss": 0.1673,
3321
+ "step": 473
3322
+ },
3323
+ {
3324
+ "epoch": 1.267648042823687,
3325
+ "grad_norm": 2.458280086517334,
3326
+ "learning_rate": 3.365304638933322e-06,
3327
+ "loss": 0.1795,
3328
+ "step": 474
3329
+ },
3330
+ {
3331
+ "epoch": 1.2703245232519238,
3332
+ "grad_norm": 2.1865360736846924,
3333
+ "learning_rate": 3.3440537342636483e-06,
3334
+ "loss": 0.1492,
3335
+ "step": 475
3336
+ },
3337
+ {
3338
+ "epoch": 1.2730010036801607,
3339
+ "grad_norm": 2.6727495193481445,
3340
+ "learning_rate": 3.322836375104608e-06,
3341
+ "loss": 0.1857,
3342
+ "step": 476
3343
+ },
3344
+ {
3345
+ "epoch": 1.2756774841083973,
3346
+ "grad_norm": 2.2446489334106445,
3347
+ "learning_rate": 3.3016529912691476e-06,
3348
+ "loss": 0.1487,
3349
+ "step": 477
3350
+ },
3351
+ {
3352
+ "epoch": 1.2783539645366342,
3353
+ "grad_norm": 2.2084949016571045,
3354
+ "learning_rate": 3.2805040118819574e-06,
3355
+ "loss": 0.1687,
3356
+ "step": 478
3357
+ },
3358
+ {
3359
+ "epoch": 1.281030444964871,
3360
+ "grad_norm": 2.6252238750457764,
3361
+ "learning_rate": 3.2593898653707773e-06,
3362
+ "loss": 0.2246,
3363
+ "step": 479
3364
+ },
3365
+ {
3366
+ "epoch": 1.283706925393108,
3367
+ "grad_norm": 3.0610384941101074,
3368
+ "learning_rate": 3.238310979457713e-06,
3369
+ "loss": 0.1692,
3370
+ "step": 480
3371
+ },
3372
+ {
3373
+ "epoch": 1.2863834058213448,
3374
+ "grad_norm": 2.619529962539673,
3375
+ "learning_rate": 3.2172677811505766e-06,
3376
+ "loss": 0.2134,
3377
+ "step": 481
3378
+ },
3379
+ {
3380
+ "epoch": 1.2890598862495817,
3381
+ "grad_norm": 2.1486215591430664,
3382
+ "learning_rate": 3.1962606967342356e-06,
3383
+ "loss": 0.1493,
3384
+ "step": 482
3385
+ },
3386
+ {
3387
+ "epoch": 1.2917363666778185,
3388
+ "grad_norm": 2.3791916370391846,
3389
+ "learning_rate": 3.1752901517619733e-06,
3390
+ "loss": 0.1632,
3391
+ "step": 483
3392
+ },
3393
+ {
3394
+ "epoch": 1.2944128471060554,
3395
+ "grad_norm": 2.50266695022583,
3396
+ "learning_rate": 3.1543565710468743e-06,
3397
+ "loss": 0.1722,
3398
+ "step": 484
3399
+ },
3400
+ {
3401
+ "epoch": 1.2970893275342923,
3402
+ "grad_norm": 2.1674370765686035,
3403
+ "learning_rate": 3.1334603786532147e-06,
3404
+ "loss": 0.1361,
3405
+ "step": 485
3406
+ },
3407
+ {
3408
+ "epoch": 1.2997658079625292,
3409
+ "grad_norm": 2.1245274543762207,
3410
+ "learning_rate": 3.112601997887873e-06,
3411
+ "loss": 0.1298,
3412
+ "step": 486
3413
+ },
3414
+ {
3415
+ "epoch": 1.302442288390766,
3416
+ "grad_norm": 2.3078453540802,
3417
+ "learning_rate": 3.091781851291753e-06,
3418
+ "loss": 0.1755,
3419
+ "step": 487
3420
+ },
3421
+ {
3422
+ "epoch": 1.305118768819003,
3423
+ "grad_norm": 1.8488824367523193,
3424
+ "learning_rate": 3.0710003606312292e-06,
3425
+ "loss": 0.1329,
3426
+ "step": 488
3427
+ },
3428
+ {
3429
+ "epoch": 1.3077952492472398,
3430
+ "grad_norm": 2.116867780685425,
3431
+ "learning_rate": 3.050257946889594e-06,
3432
+ "loss": 0.1499,
3433
+ "step": 489
3434
+ },
3435
+ {
3436
+ "epoch": 1.3104717296754766,
3437
+ "grad_norm": 2.7279255390167236,
3438
+ "learning_rate": 3.02955503025854e-06,
3439
+ "loss": 0.1451,
3440
+ "step": 490
3441
+ },
3442
+ {
3443
+ "epoch": 1.3131482101037135,
3444
+ "grad_norm": 2.121842622756958,
3445
+ "learning_rate": 3.00889203012964e-06,
3446
+ "loss": 0.1373,
3447
+ "step": 491
3448
+ },
3449
+ {
3450
+ "epoch": 1.3158246905319504,
3451
+ "grad_norm": 2.4118058681488037,
3452
+ "learning_rate": 2.988269365085854e-06,
3453
+ "loss": 0.1469,
3454
+ "step": 492
3455
+ },
3456
+ {
3457
+ "epoch": 1.3185011709601873,
3458
+ "grad_norm": 3.3426828384399414,
3459
+ "learning_rate": 2.967687452893051e-06,
3460
+ "loss": 0.2075,
3461
+ "step": 493
3462
+ },
3463
+ {
3464
+ "epoch": 1.3211776513884241,
3465
+ "grad_norm": 2.4673588275909424,
3466
+ "learning_rate": 2.947146710491545e-06,
3467
+ "loss": 0.1379,
3468
+ "step": 494
3469
+ },
3470
+ {
3471
+ "epoch": 1.323854131816661,
3472
+ "grad_norm": 2.403437614440918,
3473
+ "learning_rate": 2.9266475539876447e-06,
3474
+ "loss": 0.127,
3475
+ "step": 495
3476
+ },
3477
+ {
3478
+ "epoch": 1.3265306122448979,
3479
+ "grad_norm": 2.6860616207122803,
3480
+ "learning_rate": 2.9061903986452323e-06,
3481
+ "loss": 0.171,
3482
+ "step": 496
3483
+ },
3484
+ {
3485
+ "epoch": 1.3292070926731347,
3486
+ "grad_norm": 2.758408546447754,
3487
+ "learning_rate": 2.8857756588773457e-06,
3488
+ "loss": 0.1609,
3489
+ "step": 497
3490
+ },
3491
+ {
3492
+ "epoch": 1.3318835731013716,
3493
+ "grad_norm": 2.4655535221099854,
3494
+ "learning_rate": 2.865403748237784e-06,
3495
+ "loss": 0.1769,
3496
+ "step": 498
3497
+ },
3498
+ {
3499
+ "epoch": 1.3345600535296085,
3500
+ "grad_norm": 2.894625186920166,
3501
+ "learning_rate": 2.845075079412731e-06,
3502
+ "loss": 0.1772,
3503
+ "step": 499
3504
+ },
3505
+ {
3506
+ "epoch": 1.3372365339578454,
3507
+ "grad_norm": 2.792318105697632,
3508
+ "learning_rate": 2.824790064212396e-06,
3509
+ "loss": 0.1386,
3510
+ "step": 500
3511
+ }
3512
+ ],
3513
+ "logging_steps": 1,
3514
+ "max_steps": 748,
3515
+ "num_input_tokens_seen": 0,
3516
+ "num_train_epochs": 2,
3517
+ "save_steps": 100,
3518
+ "stateful_callbacks": {
3519
+ "TrainerControl": {
3520
+ "args": {
3521
+ "should_epoch_stop": false,
3522
+ "should_evaluate": false,
3523
+ "should_log": false,
3524
+ "should_save": true,
3525
+ "should_training_stop": false
3526
+ },
3527
+ "attributes": {}
3528
+ }
3529
+ },
3530
+ "total_flos": 5.505182926540636e+17,
3531
+ "train_batch_size": 1,
3532
+ "trial_name": null,
3533
+ "trial_params": null
3534
+ }
checkpoint-600/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
+ }
checkpoint-600/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'Please reason step by step, and put your final answer within \\boxed{}.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nPlease reason step by step, and put your final answer within \\boxed{}.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
checkpoint-600/config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen2ForCausalLM"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "dtype": "float32",
7
+ "eos_token_id": 151645,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 3584,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 18944,
12
+ "layer_types": [
13
+ "full_attention",
14
+ "full_attention",
15
+ "full_attention",
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention"
41
+ ],
42
+ "max_position_embeddings": 4096,
43
+ "max_window_layers": 28,
44
+ "model_type": "qwen2",
45
+ "num_attention_heads": 28,
46
+ "num_hidden_layers": 28,
47
+ "num_key_value_heads": 4,
48
+ "pad_token_id": 151643,
49
+ "rms_norm_eps": 1e-06,
50
+ "rope_scaling": null,
51
+ "rope_theta": 10000,
52
+ "sliding_window": null,
53
+ "tie_word_embeddings": false,
54
+ "transformers_version": "4.57.1",
55
+ "use_cache": false,
56
+ "use_mrope": false,
57
+ "use_sliding_window": false,
58
+ "vocab_size": 152064
59
+ }
checkpoint-600/generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "eos_token_id": [
3
+ 151645,
4
+ 151643
5
+ ],
6
+ "max_new_tokens": 2048,
7
+ "pad_token_id": 151643,
8
+ "transformers_version": "4.57.1"
9
+ }
checkpoint-600/model.safetensors.index.json ADDED
@@ -0,0 +1,347 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_parameters": 7615616512,
4
+ "total_size": 30462466048
5
+ },
6
+ "weight_map": {
7
+ "lm_head.weight": "model-00007-of-00007.safetensors",
8
+ "model.embed_tokens.weight": "model-00001-of-00007.safetensors",
9
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00007.safetensors",
10
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
11
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
12
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
13
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
14
+ "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
15
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
16
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
17
+ "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
18
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
19
+ "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
20
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
21
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00007.safetensors",
22
+ "model.layers.1.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
23
+ "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
24
+ "model.layers.1.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
25
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
26
+ "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
27
+ "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
28
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
29
+ "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
30
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
31
+ "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
32
+ "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
33
+ "model.layers.10.input_layernorm.weight": "model-00003-of-00007.safetensors",
34
+ "model.layers.10.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
35
+ "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
36
+ "model.layers.10.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
37
+ "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
38
+ "model.layers.10.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
39
+ "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
40
+ "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
41
+ "model.layers.10.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
42
+ "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
43
+ "model.layers.10.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
44
+ "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
45
+ "model.layers.11.input_layernorm.weight": "model-00003-of-00007.safetensors",
46
+ "model.layers.11.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
47
+ "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
48
+ "model.layers.11.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
49
+ "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
50
+ "model.layers.11.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
51
+ "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
52
+ "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
53
+ "model.layers.11.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
54
+ "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
55
+ "model.layers.11.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
56
+ "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
57
+ "model.layers.12.input_layernorm.weight": "model-00003-of-00007.safetensors",
58
+ "model.layers.12.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
59
+ "model.layers.12.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
60
+ "model.layers.12.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
61
+ "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
62
+ "model.layers.12.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
63
+ "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
64
+ "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
65
+ "model.layers.12.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
66
+ "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
67
+ "model.layers.12.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
68
+ "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
69
+ "model.layers.13.input_layernorm.weight": "model-00004-of-00007.safetensors",
70
+ "model.layers.13.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
71
+ "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
72
+ "model.layers.13.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
73
+ "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
74
+ "model.layers.13.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
75
+ "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
76
+ "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
77
+ "model.layers.13.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
78
+ "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
79
+ "model.layers.13.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
80
+ "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
81
+ "model.layers.14.input_layernorm.weight": "model-00004-of-00007.safetensors",
82
+ "model.layers.14.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
83
+ "model.layers.14.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
84
+ "model.layers.14.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
85
+ "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
86
+ "model.layers.14.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
87
+ "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
88
+ "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
89
+ "model.layers.14.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
90
+ "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
91
+ "model.layers.14.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
92
+ "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
93
+ "model.layers.15.input_layernorm.weight": "model-00004-of-00007.safetensors",
94
+ "model.layers.15.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
95
+ "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
96
+ "model.layers.15.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
97
+ "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
98
+ "model.layers.15.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
99
+ "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
100
+ "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
101
+ "model.layers.15.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
102
+ "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
103
+ "model.layers.15.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
104
+ "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
105
+ "model.layers.16.input_layernorm.weight": "model-00004-of-00007.safetensors",
106
+ "model.layers.16.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
107
+ "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
108
+ "model.layers.16.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
109
+ "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
110
+ "model.layers.16.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
111
+ "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
112
+ "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
113
+ "model.layers.16.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
114
+ "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
115
+ "model.layers.16.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
116
+ "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
117
+ "model.layers.17.input_layernorm.weight": "model-00004-of-00007.safetensors",
118
+ "model.layers.17.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
119
+ "model.layers.17.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
120
+ "model.layers.17.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
121
+ "model.layers.17.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
122
+ "model.layers.17.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
123
+ "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
124
+ "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
125
+ "model.layers.17.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
126
+ "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
127
+ "model.layers.17.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
128
+ "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
129
+ "model.layers.18.input_layernorm.weight": "model-00005-of-00007.safetensors",
130
+ "model.layers.18.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
131
+ "model.layers.18.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
132
+ "model.layers.18.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
133
+ "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
134
+ "model.layers.18.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
135
+ "model.layers.18.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
136
+ "model.layers.18.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
137
+ "model.layers.18.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
138
+ "model.layers.18.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
139
+ "model.layers.18.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
140
+ "model.layers.18.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
141
+ "model.layers.19.input_layernorm.weight": "model-00005-of-00007.safetensors",
142
+ "model.layers.19.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
143
+ "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
144
+ "model.layers.19.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
145
+ "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
146
+ "model.layers.19.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
147
+ "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
148
+ "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
149
+ "model.layers.19.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
150
+ "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
151
+ "model.layers.19.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
152
+ "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
153
+ "model.layers.2.input_layernorm.weight": "model-00001-of-00007.safetensors",
154
+ "model.layers.2.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
155
+ "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
156
+ "model.layers.2.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
157
+ "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
158
+ "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
159
+ "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
160
+ "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
161
+ "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
162
+ "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
163
+ "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
164
+ "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
165
+ "model.layers.20.input_layernorm.weight": "model-00005-of-00007.safetensors",
166
+ "model.layers.20.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
167
+ "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
168
+ "model.layers.20.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
169
+ "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
170
+ "model.layers.20.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
171
+ "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
172
+ "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
173
+ "model.layers.20.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
174
+ "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
175
+ "model.layers.20.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
176
+ "model.layers.20.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
177
+ "model.layers.21.input_layernorm.weight": "model-00005-of-00007.safetensors",
178
+ "model.layers.21.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
179
+ "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
180
+ "model.layers.21.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
181
+ "model.layers.21.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
182
+ "model.layers.21.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
183
+ "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
184
+ "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
185
+ "model.layers.21.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
186
+ "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
187
+ "model.layers.21.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
188
+ "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
189
+ "model.layers.22.input_layernorm.weight": "model-00005-of-00007.safetensors",
190
+ "model.layers.22.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
191
+ "model.layers.22.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
192
+ "model.layers.22.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
193
+ "model.layers.22.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
194
+ "model.layers.22.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
195
+ "model.layers.22.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
196
+ "model.layers.22.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
197
+ "model.layers.22.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
198
+ "model.layers.22.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
199
+ "model.layers.22.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
200
+ "model.layers.22.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
201
+ "model.layers.23.input_layernorm.weight": "model-00005-of-00007.safetensors",
202
+ "model.layers.23.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
203
+ "model.layers.23.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
204
+ "model.layers.23.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
205
+ "model.layers.23.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
206
+ "model.layers.23.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
207
+ "model.layers.23.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
208
+ "model.layers.23.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
209
+ "model.layers.23.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
210
+ "model.layers.23.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
211
+ "model.layers.23.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
212
+ "model.layers.23.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
213
+ "model.layers.24.input_layernorm.weight": "model-00006-of-00007.safetensors",
214
+ "model.layers.24.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
215
+ "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
216
+ "model.layers.24.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
217
+ "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
218
+ "model.layers.24.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
219
+ "model.layers.24.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
220
+ "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
221
+ "model.layers.24.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
222
+ "model.layers.24.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
223
+ "model.layers.24.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
224
+ "model.layers.24.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
225
+ "model.layers.25.input_layernorm.weight": "model-00006-of-00007.safetensors",
226
+ "model.layers.25.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
227
+ "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
228
+ "model.layers.25.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
229
+ "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
230
+ "model.layers.25.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
231
+ "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
232
+ "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
233
+ "model.layers.25.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
234
+ "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
235
+ "model.layers.25.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
236
+ "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
237
+ "model.layers.26.input_layernorm.weight": "model-00006-of-00007.safetensors",
238
+ "model.layers.26.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
239
+ "model.layers.26.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
240
+ "model.layers.26.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
241
+ "model.layers.26.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
242
+ "model.layers.26.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
243
+ "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
244
+ "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
245
+ "model.layers.26.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
246
+ "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
247
+ "model.layers.26.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
248
+ "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
249
+ "model.layers.27.input_layernorm.weight": "model-00006-of-00007.safetensors",
250
+ "model.layers.27.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
251
+ "model.layers.27.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
252
+ "model.layers.27.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
253
+ "model.layers.27.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
254
+ "model.layers.27.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
255
+ "model.layers.27.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
256
+ "model.layers.27.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
257
+ "model.layers.27.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
258
+ "model.layers.27.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
259
+ "model.layers.27.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
260
+ "model.layers.27.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
261
+ "model.layers.3.input_layernorm.weight": "model-00002-of-00007.safetensors",
262
+ "model.layers.3.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
263
+ "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
264
+ "model.layers.3.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
265
+ "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
266
+ "model.layers.3.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
267
+ "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
268
+ "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
269
+ "model.layers.3.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
270
+ "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
271
+ "model.layers.3.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
272
+ "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
273
+ "model.layers.4.input_layernorm.weight": "model-00002-of-00007.safetensors",
274
+ "model.layers.4.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
275
+ "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
276
+ "model.layers.4.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
277
+ "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
278
+ "model.layers.4.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
279
+ "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
280
+ "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
281
+ "model.layers.4.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
282
+ "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
283
+ "model.layers.4.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
284
+ "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
285
+ "model.layers.5.input_layernorm.weight": "model-00002-of-00007.safetensors",
286
+ "model.layers.5.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
287
+ "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
288
+ "model.layers.5.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
289
+ "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
290
+ "model.layers.5.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
291
+ "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
292
+ "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
293
+ "model.layers.5.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
294
+ "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
295
+ "model.layers.5.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
296
+ "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
297
+ "model.layers.6.input_layernorm.weight": "model-00002-of-00007.safetensors",
298
+ "model.layers.6.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
299
+ "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
300
+ "model.layers.6.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
301
+ "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
302
+ "model.layers.6.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
303
+ "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
304
+ "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
305
+ "model.layers.6.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
306
+ "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
307
+ "model.layers.6.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
308
+ "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
309
+ "model.layers.7.input_layernorm.weight": "model-00002-of-00007.safetensors",
310
+ "model.layers.7.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
311
+ "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
312
+ "model.layers.7.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
313
+ "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
314
+ "model.layers.7.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
315
+ "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
316
+ "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
317
+ "model.layers.7.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
318
+ "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
319
+ "model.layers.7.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
320
+ "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
321
+ "model.layers.8.input_layernorm.weight": "model-00003-of-00007.safetensors",
322
+ "model.layers.8.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
323
+ "model.layers.8.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
324
+ "model.layers.8.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
325
+ "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
326
+ "model.layers.8.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
327
+ "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
328
+ "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
329
+ "model.layers.8.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
330
+ "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
331
+ "model.layers.8.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
332
+ "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
333
+ "model.layers.9.input_layernorm.weight": "model-00003-of-00007.safetensors",
334
+ "model.layers.9.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
335
+ "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
336
+ "model.layers.9.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
337
+ "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
338
+ "model.layers.9.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
339
+ "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
340
+ "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
341
+ "model.layers.9.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
342
+ "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
343
+ "model.layers.9.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
344
+ "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
345
+ "model.norm.weight": "model-00006-of-00007.safetensors"
346
+ }
347
+ }
checkpoint-700/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5664fd5079c69dc8edcc429ded884f07d45c771f27a5a65fe1ad751348f9e1de
3
+ size 15429
checkpoint-700/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a3f17c4e252e5b7e99f3573c3bca992070f9ef436dd578ee9e88c333a94cef0
3
+ size 15429
checkpoint-700/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
3
+ size 11421896
checkpoint-748/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4188048c882e5b595f5e2cce07e8c73ea45afa11bc95f5627f2d6191ccaa5e40
3
+ size 15429
checkpoint-748/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5664fd5079c69dc8edcc429ded884f07d45c771f27a5a65fe1ad751348f9e1de
3
+ size 15429
checkpoint-748/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a3f17c4e252e5b7e99f3573c3bca992070f9ef436dd578ee9e88c333a94cef0
3
+ size 15429