Hugging Face's logo

336labs
/
VisionEncoder-to-MLLM-ModelZoo

Safetensors
vision-encoder-evaluation
multimodal
visual-tokenizer-benchmark
checkpoints
Model card Files
xet
 Community
lyunee commited on
Commit
7b74a84
·
verified ·
1 Parent(s): ab5c40b

Upload continuous evaluation checkpoints, batch 10

Browse files
Files changed (45) hide show
  1. .gitattributes +2 -0
  2. continuous/qwen25/pe_core_s16_384/finetune/trainer_state.json +1856 -0
  3. continuous/qwen25/pe_core_s16_384/finetune/training_args.bin +3 -0
  4. continuous/qwen25/pe_core_s16_384/finetune/vocab.json +0 -0
  5. continuous/qwen25/pe_core_s16_384/pretrain/config.json +87 -0
  6. continuous/qwen25/pe_core_s16_384/pretrain/mm_projector.bin +3 -0
  7. continuous/qwen25/pe_core_s16_384/pretrain/trainer_state.json +421 -0
  8. continuous/qwen25/pe_core_t16_384/finetune/added_tokens.json +24 -0
  9. continuous/qwen25/pe_core_t16_384/finetune/chat_template.jinja +54 -0
  10. continuous/qwen25/pe_core_t16_384/finetune/config.json +87 -0
  11. continuous/qwen25/pe_core_t16_384/finetune/generation_config.json +13 -0
  12. continuous/qwen25/pe_core_t16_384/finetune/merges.txt +0 -0
  13. continuous/qwen25/pe_core_t16_384/finetune/model.safetensors +3 -0
  14. continuous/qwen25/pe_core_t16_384/finetune/special_tokens_map.json +31 -0
  15. continuous/qwen25/pe_core_t16_384/finetune/tokenizer.json +3 -0
  16. continuous/qwen25/pe_core_t16_384/finetune/tokenizer_config.json +208 -0
  17. continuous/qwen25/pe_core_t16_384/finetune/trainer_state.json +1856 -0
  18. continuous/qwen25/pe_core_t16_384/finetune/training_args.bin +3 -0
  19. continuous/qwen25/pe_core_t16_384/finetune/vocab.json +0 -0
  20. continuous/qwen25/pe_core_t16_384/pretrain/config.json +87 -0
  21. continuous/qwen25/pe_core_t16_384/pretrain/mm_projector.bin +3 -0
  22. continuous/qwen25/pe_core_t16_384/pretrain/trainer_state.json +421 -0
  23. continuous/qwen25/pe_lang_g14_448/finetune/added_tokens.json +24 -0
  24. continuous/qwen25/pe_lang_g14_448/finetune/chat_template.jinja +54 -0
  25. continuous/qwen25/pe_lang_g14_448/finetune/config.json +87 -0
  26. continuous/qwen25/pe_lang_g14_448/finetune/generation_config.json +13 -0
  27. continuous/qwen25/pe_lang_g14_448/finetune/merges.txt +0 -0
  28. continuous/qwen25/pe_lang_g14_448/finetune/model-00001-of-00002.safetensors +3 -0
  29. continuous/qwen25/pe_lang_g14_448/finetune/model-00002-of-00002.safetensors +3 -0
  30. continuous/qwen25/pe_lang_g14_448/finetune/model.safetensors.index.json +0 -0
  31. continuous/qwen25/pe_lang_g14_448/finetune/special_tokens_map.json +31 -0
  32. continuous/qwen25/pe_lang_g14_448/finetune/tokenizer.json +3 -0
  33. continuous/qwen25/pe_lang_g14_448/finetune/tokenizer_config.json +208 -0
  34. continuous/qwen25/pe_lang_g14_448/finetune/trainer_state.json +1856 -0
  35. continuous/qwen25/pe_lang_g14_448/finetune/training_args.bin +3 -0
  36. continuous/qwen25/pe_lang_g14_448/finetune/vocab.json +0 -0
  37. continuous/qwen25/pe_lang_g14_448/pretrain/config.json +87 -0
  38. continuous/qwen25/pe_lang_g14_448/pretrain/mm_projector.bin +3 -0
  39. continuous/qwen25/pe_lang_g14_448/pretrain/trainer_state.json +421 -0
  40. continuous/qwen25/pe_lang_g14_448_tiling/finetune/added_tokens.json +24 -0
  41. continuous/qwen25/pe_lang_g14_448_tiling/finetune/chat_template.jinja +54 -0
  42. continuous/qwen25/pe_lang_g14_448_tiling/finetune/config.json +87 -0
  43. continuous/qwen25/pe_lang_g14_448_tiling/finetune/generation_config.json +13 -0
  44. continuous/qwen25/pe_lang_g14_448_tiling/finetune/merges.txt +0 -0
  45. continuous/qwen25/pe_lang_g14_448_tiling/finetune/model-00001-of-00002.safetensors +3 -0
.gitattributes CHANGED
@@ -79,3 +79,5 @@ continuous/qwen25/pe_core_b16_224/finetune/tokenizer.json filter=lfs diff=lfs me
79
  continuous/qwen25/pe_core_g14_448/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
80
  continuous/qwen25/pe_core_l14_336/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
81
  continuous/qwen25/pe_core_s16_384/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
 
79
  continuous/qwen25/pe_core_g14_448/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
80
  continuous/qwen25/pe_core_l14_336/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
81
  continuous/qwen25/pe_core_s16_384/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
82
+ continuous/qwen25/pe_core_t16_384/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
83
+ continuous/qwen25/pe_lang_g14_448/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
continuous/qwen25/pe_core_s16_384/finetune/trainer_state.json ADDED
@@ -0,0 +1,1856 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.9999037443449803,
6
+ "eval_steps": 500,
7
+ "global_step": 2597,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.003850226200789296,
14
+ "grad_norm": 14.070768356323242,
15
+ "learning_rate": 1.153846153846154e-06,
16
+ "loss": 1.9244,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.007700452401578592,
21
+ "grad_norm": 11.925518989562988,
22
+ "learning_rate": 2.435897435897436e-06,
23
+ "loss": 1.6633,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.01155067860236789,
28
+ "grad_norm": 3.785520076751709,
29
+ "learning_rate": 3.7179487179487184e-06,
30
+ "loss": 1.1741,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.015400904803157185,
35
+ "grad_norm": 0.7478566765785217,
36
+ "learning_rate": 5e-06,
37
+ "loss": 0.9642,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.019251131003946483,
42
+ "grad_norm": 0.35097622871398926,
43
+ "learning_rate": 6.282051282051282e-06,
44
+ "loss": 0.9242,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.02310135720473578,
49
+ "grad_norm": 0.2869117558002472,
50
+ "learning_rate": 7.564102564102564e-06,
51
+ "loss": 0.8801,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.026951583405525073,
56
+ "grad_norm": 0.2318483144044876,
57
+ "learning_rate": 8.846153846153847e-06,
58
+ "loss": 0.8549,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.03080180960631437,
63
+ "grad_norm": 0.24267660081386566,
64
+ "learning_rate": 9.999996111488723e-06,
65
+ "loss": 0.8644,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.034652035807103666,
70
+ "grad_norm": 0.21493682265281677,
71
+ "learning_rate": 9.999529497453782e-06,
72
+ "loss": 0.8453,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.038502262007892966,
77
+ "grad_norm": 0.22352737188339233,
78
+ "learning_rate": 9.998285264324193e-06,
79
+ "loss": 0.826,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.04235248820868226,
84
+ "grad_norm": 0.20293062925338745,
85
+ "learning_rate": 9.996263605626052e-06,
86
+ "loss": 0.8147,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.04620271440947156,
91
+ "grad_norm": 0.2265503704547882,
92
+ "learning_rate": 9.99346483580503e-06,
93
+ "loss": 0.8266,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.05005294061026085,
98
+ "grad_norm": 0.20966871082782745,
99
+ "learning_rate": 9.989889390177467e-06,
100
+ "loss": 0.8186,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.053903166811050146,
105
+ "grad_norm": 0.24433185160160065,
106
+ "learning_rate": 9.985537824862647e-06,
107
+ "loss": 0.8191,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.057753393011839446,
112
+ "grad_norm": 0.24447955191135406,
113
+ "learning_rate": 9.980410816696317e-06,
114
+ "loss": 0.8014,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.06160361921262874,
119
+ "grad_norm": 0.2235342562198639,
120
+ "learning_rate": 9.97450916312541e-06,
121
+ "loss": 0.8085,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.06545384541341803,
126
+ "grad_norm": 0.1999809592962265,
127
+ "learning_rate": 9.967833782084004e-06,
128
+ "loss": 0.8096,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.06930407161420733,
133
+ "grad_norm": 0.23577070236206055,
134
+ "learning_rate": 9.960385711850554e-06,
135
+ "loss": 0.8362,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.07315429781499663,
140
+ "grad_norm": 0.22884425520896912,
141
+ "learning_rate": 9.952166110886394e-06,
142
+ "loss": 0.827,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.07700452401578593,
147
+ "grad_norm": 0.25477662682533264,
148
+ "learning_rate": 9.943176257655567e-06,
149
+ "loss": 0.8213,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.08085475021657522,
154
+ "grad_norm": 0.21394552290439606,
155
+ "learning_rate": 9.933417550425952e-06,
156
+ "loss": 0.8101,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.08470497641736452,
161
+ "grad_norm": 0.2264709621667862,
162
+ "learning_rate": 9.9228915070518e-06,
163
+ "loss": 0.799,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.08855520261815382,
168
+ "grad_norm": 0.2280338704586029,
169
+ "learning_rate": 9.911599764737632e-06,
170
+ "loss": 0.8018,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.09240542881894312,
175
+ "grad_norm": 0.2230648547410965,
176
+ "learning_rate": 9.899544079783606e-06,
177
+ "loss": 0.7962,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.0962556550197324,
182
+ "grad_norm": 0.23267169296741486,
183
+ "learning_rate": 9.886726327312338e-06,
184
+ "loss": 0.8108,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.1001058812205217,
189
+ "grad_norm": 0.21577537059783936,
190
+ "learning_rate": 9.87314850097724e-06,
191
+ "loss": 0.8105,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.103956107421311,
196
+ "grad_norm": 0.21776197850704193,
197
+ "learning_rate": 9.858812712652452e-06,
198
+ "loss": 0.7988,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.10780633362210029,
203
+ "grad_norm": 0.2081999033689499,
204
+ "learning_rate": 9.843721192104341e-06,
205
+ "loss": 0.7987,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.11165655982288959,
210
+ "grad_norm": 0.22058522701263428,
211
+ "learning_rate": 9.827876286644698e-06,
212
+ "loss": 0.8184,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.11550678602367889,
217
+ "grad_norm": 0.20375195145606995,
218
+ "learning_rate": 9.811280460765643e-06,
219
+ "loss": 0.7785,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.11935701222446819,
224
+ "grad_norm": 0.22356650233268738,
225
+ "learning_rate": 9.793936295756292e-06,
226
+ "loss": 0.7891,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.12320723842525748,
231
+ "grad_norm": 0.22649717330932617,
232
+ "learning_rate": 9.775846489301279e-06,
233
+ "loss": 0.7932,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.12705746462604678,
238
+ "grad_norm": 0.224069282412529,
239
+ "learning_rate": 9.757013855061157e-06,
240
+ "loss": 0.7911,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.13090769082683607,
245
+ "grad_norm": 0.22156226634979248,
246
+ "learning_rate": 9.737441322234758e-06,
247
+ "loss": 0.8046,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.13475791702762538,
252
+ "grad_norm": 0.226561039686203,
253
+ "learning_rate": 9.717131935103605e-06,
254
+ "loss": 0.8024,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.13860814322841467,
259
+ "grad_norm": 0.22070609033107758,
260
+ "learning_rate": 9.696088852558398e-06,
261
+ "loss": 0.7889,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.14245836942920398,
266
+ "grad_norm": 0.21713665127754211,
267
+ "learning_rate": 9.674315347607686e-06,
268
+ "loss": 0.814,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.14630859562999327,
273
+ "grad_norm": 0.22033236920833588,
274
+ "learning_rate": 9.651814806868796e-06,
275
+ "loss": 0.7795,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.15015882183078255,
280
+ "grad_norm": 0.2194281667470932,
281
+ "learning_rate": 9.62859073004107e-06,
282
+ "loss": 0.7839,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.15400904803157187,
287
+ "grad_norm": 0.20464260876178741,
288
+ "learning_rate": 9.604646729361547e-06,
289
+ "loss": 0.7903,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.15785927423236115,
294
+ "grad_norm": 0.20834913849830627,
295
+ "learning_rate": 9.579986529043099e-06,
296
+ "loss": 0.7822,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.16170950043315044,
301
+ "grad_norm": 0.2194782942533493,
302
+ "learning_rate": 9.554613964695189e-06,
303
+ "loss": 0.7939,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.16555972663393975,
308
+ "grad_norm": 0.21577735245227814,
309
+ "learning_rate": 9.528532982727281e-06,
310
+ "loss": 0.7855,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.16940995283472904,
315
+ "grad_norm": 0.2090895175933838,
316
+ "learning_rate": 9.501747639735018e-06,
317
+ "loss": 0.7842,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.17326017903551832,
322
+ "grad_norm": 0.2177504003047943,
323
+ "learning_rate": 9.474262101869272e-06,
324
+ "loss": 0.7724,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.17711040523630764,
329
+ "grad_norm": 0.22241517901420593,
330
+ "learning_rate": 9.446080644188141e-06,
331
+ "loss": 0.7773,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.18096063143709692,
336
+ "grad_norm": 0.2119297832250595,
337
+ "learning_rate": 9.417207649992017e-06,
338
+ "loss": 0.7853,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.18481085763788624,
343
+ "grad_norm": 0.2144523411989212,
344
+ "learning_rate": 9.387647610141816e-06,
345
+ "loss": 0.7851,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.18866108383867553,
350
+ "grad_norm": 0.20313963294029236,
351
+ "learning_rate": 9.35740512236047e-06,
352
+ "loss": 0.7706,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.1925113100394648,
357
+ "grad_norm": 0.1999216079711914,
358
+ "learning_rate": 9.326484890517809e-06,
359
+ "loss": 0.7799,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.19636153624025413,
364
+ "grad_norm": 0.2177458554506302,
365
+ "learning_rate": 9.294891723898926e-06,
366
+ "loss": 0.771,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.2002117624410434,
371
+ "grad_norm": 0.22673913836479187,
372
+ "learning_rate": 9.262630536456153e-06,
373
+ "loss": 0.8007,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.2040619886418327,
378
+ "grad_norm": 0.2234310358762741,
379
+ "learning_rate": 9.229706346044749e-06,
380
+ "loss": 0.7741,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.207912214842622,
385
+ "grad_norm": 0.2314017415046692,
386
+ "learning_rate": 9.196124273642422e-06,
387
+ "loss": 0.7851,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.2117624410434113,
392
+ "grad_norm": 0.2147892415523529,
393
+ "learning_rate": 9.161889542552842e-06,
394
+ "loss": 0.7877,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.21561266724420058,
399
+ "grad_norm": 0.21380889415740967,
400
+ "learning_rate": 9.127007477593193e-06,
401
+ "loss": 0.7905,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.2194628934449899,
406
+ "grad_norm": 0.2433636337518692,
407
+ "learning_rate": 9.091483504265974e-06,
408
+ "loss": 0.8016,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.22331311964577918,
413
+ "grad_norm": 0.23178638517856598,
414
+ "learning_rate": 9.055323147915116e-06,
415
+ "loss": 0.779,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.2271633458465685,
420
+ "grad_norm": 0.21859340369701385,
421
+ "learning_rate": 9.018532032866585e-06,
422
+ "loss": 0.7903,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.23101357204735778,
427
+ "grad_norm": 0.2236909717321396,
428
+ "learning_rate": 8.981115881553579e-06,
429
+ "loss": 0.7809,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.23486379824814707,
434
+ "grad_norm": 0.24130985140800476,
435
+ "learning_rate": 8.943080513626476e-06,
436
+ "loss": 0.8005,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.23871402444893638,
441
+ "grad_norm": 0.2166345715522766,
442
+ "learning_rate": 8.904431845047641e-06,
443
+ "loss": 0.7848,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.24256425064972567,
448
+ "grad_norm": 0.2301849126815796,
449
+ "learning_rate": 8.86517588717128e-06,
450
+ "loss": 0.8007,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.24641447685051496,
455
+ "grad_norm": 0.21026968955993652,
456
+ "learning_rate": 8.82531874580844e-06,
457
+ "loss": 0.7907,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.25026470305130427,
462
+ "grad_norm": 0.2013477087020874,
463
+ "learning_rate": 8.784866620277314e-06,
464
+ "loss": 0.7486,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.25411492925209356,
469
+ "grad_norm": 0.23873844742774963,
470
+ "learning_rate": 8.743825802439018e-06,
471
+ "loss": 0.7817,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.25796515545288284,
476
+ "grad_norm": 0.23840580880641937,
477
+ "learning_rate": 8.702202675718956e-06,
478
+ "loss": 0.7877,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.26181538165367213,
483
+ "grad_norm": 0.21497559547424316,
484
+ "learning_rate": 8.660003714113959e-06,
485
+ "loss": 0.7578,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.26566560785446147,
490
+ "grad_norm": 0.2166726142168045,
491
+ "learning_rate": 8.617235481185323e-06,
492
+ "loss": 0.7795,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.26951583405525076,
497
+ "grad_norm": 0.22696036100387573,
498
+ "learning_rate": 8.573904629037931e-06,
499
+ "loss": 0.7846,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.27336606025604004,
504
+ "grad_norm": 0.21610699594020844,
505
+ "learning_rate": 8.530017897285585e-06,
506
+ "loss": 0.7665,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.27721628645682933,
511
+ "grad_norm": 0.199606791138649,
512
+ "learning_rate": 8.485582112002746e-06,
513
+ "loss": 0.771,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.2810665126576186,
518
+ "grad_norm": 0.21623998880386353,
519
+ "learning_rate": 8.440604184662811e-06,
520
+ "loss": 0.7742,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.28491673885840796,
525
+ "grad_norm": 0.22314266860485077,
526
+ "learning_rate": 8.395091111063118e-06,
527
+ "loss": 0.7644,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.28876696505919724,
532
+ "grad_norm": 0.2068866789340973,
533
+ "learning_rate": 8.349049970236822e-06,
534
+ "loss": 0.7623,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.29261719125998653,
539
+ "grad_norm": 0.22543509304523468,
540
+ "learning_rate": 8.302487923351842e-06,
541
+ "loss": 0.754,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.2964674174607758,
546
+ "grad_norm": 0.2135101854801178,
547
+ "learning_rate": 8.255412212597021e-06,
548
+ "loss": 0.7716,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.3003176436615651,
553
+ "grad_norm": 0.21102476119995117,
554
+ "learning_rate": 8.207830160055688e-06,
555
+ "loss": 0.7679,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.3041678698623544,
560
+ "grad_norm": 0.22333043813705444,
561
+ "learning_rate": 8.159749166566786e-06,
562
+ "loss": 0.7542,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.30801809606314373,
567
+ "grad_norm": 0.21361437439918518,
568
+ "learning_rate": 8.11117671057377e-06,
569
+ "loss": 0.7794,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.311868322263933,
574
+ "grad_norm": 0.2327384054660797,
575
+ "learning_rate": 8.062120346961404e-06,
576
+ "loss": 0.7719,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.3157185484647223,
581
+ "grad_norm": 0.2164904922246933,
582
+ "learning_rate": 8.012587705880705e-06,
583
+ "loss": 0.7661,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.3195687746655116,
588
+ "grad_norm": 0.21246978640556335,
589
+ "learning_rate": 7.962586491562146e-06,
590
+ "loss": 0.7513,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.3234190008663009,
595
+ "grad_norm": 0.23735634982585907,
596
+ "learning_rate": 7.912124481117357e-06,
597
+ "loss": 0.7546,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.3272692270670902,
602
+ "grad_norm": 0.21253885328769684,
603
+ "learning_rate": 7.861209523329494e-06,
604
+ "loss": 0.7677,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.3311194532678795,
609
+ "grad_norm": 0.22427675127983093,
610
+ "learning_rate": 7.809849537432432e-06,
611
+ "loss": 0.751,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.3349696794686688,
616
+ "grad_norm": 0.21635830402374268,
617
+ "learning_rate": 7.758052511879035e-06,
618
+ "loss": 0.7604,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.3388199056694581,
623
+ "grad_norm": 0.20576179027557373,
624
+ "learning_rate": 7.705826503098642e-06,
625
+ "loss": 0.7544,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.34267013187024736,
630
+ "grad_norm": 0.21176539361476898,
631
+ "learning_rate": 7.653179634243979e-06,
632
+ "loss": 0.7552,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.34652035807103665,
637
+ "grad_norm": 0.2002309262752533,
638
+ "learning_rate": 7.600120093927696e-06,
639
+ "loss": 0.7575,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.350370584271826,
644
+ "grad_norm": 0.2256026417016983,
645
+ "learning_rate": 7.546656134948729e-06,
646
+ "loss": 0.7421,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.3542208104726153,
651
+ "grad_norm": 0.21585462987422943,
652
+ "learning_rate": 7.492796073008661e-06,
653
+ "loss": 0.7456,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.35807103667340456,
658
+ "grad_norm": 0.23257692158222198,
659
+ "learning_rate": 7.438548285418326e-06,
660
+ "loss": 0.7338,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.36192126287419385,
665
+ "grad_norm": 0.23778419196605682,
666
+ "learning_rate": 7.3839212097948e-06,
667
+ "loss": 0.766,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.36577148907498314,
672
+ "grad_norm": 0.22468209266662598,
673
+ "learning_rate": 7.328923342749039e-06,
674
+ "loss": 0.7807,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.3696217152757725,
679
+ "grad_norm": 0.22629082202911377,
680
+ "learning_rate": 7.273563238564319e-06,
681
+ "loss": 0.7658,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.37347194147656176,
686
+ "grad_norm": 0.22021794319152832,
687
+ "learning_rate": 7.217849507865724e-06,
688
+ "loss": 0.7444,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.37732216767735105,
693
+ "grad_norm": 0.21375489234924316,
694
+ "learning_rate": 7.161790816280858e-06,
695
+ "loss": 0.7522,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.38117239387814034,
700
+ "grad_norm": 0.23369838297367096,
701
+ "learning_rate": 7.1053958830920125e-06,
702
+ "loss": 0.7322,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.3850226200789296,
707
+ "grad_norm": 0.21405458450317383,
708
+ "learning_rate": 7.0486734798799715e-06,
709
+ "loss": 0.7502,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.3888728462797189,
714
+ "grad_norm": 0.21508653461933136,
715
+ "learning_rate": 6.991632429159708e-06,
716
+ "loss": 0.7744,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.39272307248050825,
721
+ "grad_norm": 0.2402423918247223,
722
+ "learning_rate": 6.934281603008132e-06,
723
+ "loss": 0.7658,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.39657329868129754,
728
+ "grad_norm": 0.2130342423915863,
729
+ "learning_rate": 6.876629921684143e-06,
730
+ "loss": 0.7637,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.4004235248820868,
735
+ "grad_norm": 0.22236108779907227,
736
+ "learning_rate": 6.818686352241199e-06,
737
+ "loss": 0.7538,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.4042737510828761,
742
+ "grad_norm": 0.20041681826114655,
743
+ "learning_rate": 6.760459907132573e-06,
744
+ "loss": 0.7498,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.4081239772836654,
749
+ "grad_norm": 0.20651090145111084,
750
+ "learning_rate": 6.701959642809589e-06,
751
+ "loss": 0.755,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.41197420348445474,
756
+ "grad_norm": 0.2179126888513565,
757
+ "learning_rate": 6.643194658312987e-06,
758
+ "loss": 0.7431,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.415824429685244,
763
+ "grad_norm": 0.2104136198759079,
764
+ "learning_rate": 6.584174093857676e-06,
765
+ "loss": 0.7694,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.4196746558860333,
770
+ "grad_norm": 0.23837727308273315,
771
+ "learning_rate": 6.5249071294110735e-06,
772
+ "loss": 0.7525,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.4235248820868226,
777
+ "grad_norm": 0.22131626307964325,
778
+ "learning_rate": 6.465402983265274e-06,
779
+ "loss": 0.7669,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.4273751082876119,
784
+ "grad_norm": 0.2218618392944336,
785
+ "learning_rate": 6.405670910603243e-06,
786
+ "loss": 0.7619,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.43122533448840117,
791
+ "grad_norm": 0.24281419813632965,
792
+ "learning_rate": 6.345720202059291e-06,
793
+ "loss": 0.7587,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.4350755606891905,
798
+ "grad_norm": 0.2185698300600052,
799
+ "learning_rate": 6.285560182274008e-06,
800
+ "loss": 0.735,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.4389257868899798,
805
+ "grad_norm": 0.21317046880722046,
806
+ "learning_rate": 6.225200208443936e-06,
807
+ "loss": 0.758,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.4427760130907691,
812
+ "grad_norm": 0.2099648267030716,
813
+ "learning_rate": 6.16464966886616e-06,
814
+ "loss": 0.7472,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.44662623929155837,
819
+ "grad_norm": 0.2584485411643982,
820
+ "learning_rate": 6.103917981478065e-06,
821
+ "loss": 0.7729,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.45047646549234766,
826
+ "grad_norm": 0.22130228579044342,
827
+ "learning_rate": 6.043014592392489e-06,
828
+ "loss": 0.7815,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.454326691693137,
833
+ "grad_norm": 0.20336058735847473,
834
+ "learning_rate": 5.981948974428485e-06,
835
+ "loss": 0.7554,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.4581769178939263,
840
+ "grad_norm": 0.21857856214046478,
841
+ "learning_rate": 5.920730625637934e-06,
842
+ "loss": 0.7335,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.46202714409471557,
847
+ "grad_norm": 0.22480957210063934,
848
+ "learning_rate": 5.8593690678282355e-06,
849
+ "loss": 0.7479,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.46587737029550486,
854
+ "grad_norm": 0.22406768798828125,
855
+ "learning_rate": 5.797873845081301e-06,
856
+ "loss": 0.7508,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.46972759649629414,
861
+ "grad_norm": 0.2213744819164276,
862
+ "learning_rate": 5.736254522269075e-06,
863
+ "loss": 0.7662,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.47357782269708343,
868
+ "grad_norm": 0.21722674369812012,
869
+ "learning_rate": 5.6745206835658386e-06,
870
+ "loss": 0.7696,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.47742804889787277,
875
+ "grad_norm": 0.20620597898960114,
876
+ "learning_rate": 5.612681930957495e-06,
877
+ "loss": 0.7386,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.48127827509866206,
882
+ "grad_norm": 0.2059183120727539,
883
+ "learning_rate": 5.550747882748095e-06,
884
+ "loss": 0.735,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.48512850129945134,
889
+ "grad_norm": 0.2326713651418686,
890
+ "learning_rate": 5.4887281720638195e-06,
891
+ "loss": 0.748,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.48897872750024063,
896
+ "grad_norm": 0.2409675121307373,
897
+ "learning_rate": 5.42663244535466e-06,
898
+ "loss": 0.7276,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.4928289537010299,
903
+ "grad_norm": 0.22057171165943146,
904
+ "learning_rate": 5.3644703608940205e-06,
905
+ "loss": 0.7383,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.49667917990181926,
910
+ "grad_norm": 0.22365929186344147,
911
+ "learning_rate": 5.302251587276491e-06,
912
+ "loss": 0.7486,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.5005294061026085,
917
+ "grad_norm": 0.21189197897911072,
918
+ "learning_rate": 5.2399858019140005e-06,
919
+ "loss": 0.7487,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.5043796323033978,
924
+ "grad_norm": 0.2064068466424942,
925
+ "learning_rate": 5.177682689530615e-06,
926
+ "loss": 0.762,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.5082298585041871,
931
+ "grad_norm": 0.22045668959617615,
932
+ "learning_rate": 5.11535194065619e-06,
933
+ "loss": 0.7314,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.5120800847049765,
938
+ "grad_norm": 0.2449105829000473,
939
+ "learning_rate": 5.053003250119113e-06,
940
+ "loss": 0.7399,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.5159303109057657,
945
+ "grad_norm": 0.2230972796678543,
946
+ "learning_rate": 4.9906463155383925e-06,
947
+ "loss": 0.7485,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.519780537106555,
952
+ "grad_norm": 0.21870507299900055,
953
+ "learning_rate": 4.9282908358153046e-06,
954
+ "loss": 0.7432,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.5236307633073443,
959
+ "grad_norm": 0.2480202168226242,
960
+ "learning_rate": 4.8659465096248345e-06,
961
+ "loss": 0.7574,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.5274809895081336,
966
+ "grad_norm": 0.2236793488264084,
967
+ "learning_rate": 4.803623033907169e-06,
968
+ "loss": 0.753,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.5313312157089229,
973
+ "grad_norm": 0.21204204857349396,
974
+ "learning_rate": 4.741330102359441e-06,
975
+ "loss": 0.7622,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.5351814419097122,
980
+ "grad_norm": 0.23196139931678772,
981
+ "learning_rate": 4.679077403927996e-06,
982
+ "loss": 0.7409,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.5390316681105015,
987
+ "grad_norm": 0.21143215894699097,
988
+ "learning_rate": 4.616874621301373e-06,
989
+ "loss": 0.7425,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.5428818943112907,
994
+ "grad_norm": 0.22586460411548615,
995
+ "learning_rate": 4.554731429404293e-06,
996
+ "loss": 0.7576,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.5467321205120801,
1001
+ "grad_norm": 0.2326935976743698,
1002
+ "learning_rate": 4.492657493892822e-06,
1003
+ "loss": 0.7392,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.5505823467128694,
1008
+ "grad_norm": 0.2281971126794815,
1009
+ "learning_rate": 4.430662469650999e-06,
1010
+ "loss": 0.7559,
1011
+ "step": 1430
1012
+ },
1013
+ {
1014
+ "epoch": 0.5544325729136587,
1015
+ "grad_norm": 0.22184833884239197,
1016
+ "learning_rate": 4.368755999289118e-06,
1017
+ "loss": 0.7609,
1018
+ "step": 1440
1019
+ },
1020
+ {
1021
+ "epoch": 0.558282799114448,
1022
+ "grad_norm": 0.22476889193058014,
1023
+ "learning_rate": 4.306947711643949e-06,
1024
+ "loss": 0.7389,
1025
+ "step": 1450
1026
+ },
1027
+ {
1028
+ "epoch": 0.5621330253152372,
1029
+ "grad_norm": 0.22329770028591156,
1030
+ "learning_rate": 4.245247220281068e-06,
1031
+ "loss": 0.7471,
1032
+ "step": 1460
1033
+ },
1034
+ {
1035
+ "epoch": 0.5659832515160266,
1036
+ "grad_norm": 0.2164798229932785,
1037
+ "learning_rate": 4.183664121999587e-06,
1038
+ "loss": 0.7412,
1039
+ "step": 1470
1040
+ },
1041
+ {
1042
+ "epoch": 0.5698334777168159,
1043
+ "grad_norm": 0.23423472046852112,
1044
+ "learning_rate": 4.122207995339474e-06,
1045
+ "loss": 0.7383,
1046
+ "step": 1480
1047
+ },
1048
+ {
1049
+ "epoch": 0.5736837039176051,
1050
+ "grad_norm": 0.21122677624225616,
1051
+ "learning_rate": 4.060888399091734e-06,
1052
+ "loss": 0.7453,
1053
+ "step": 1490
1054
+ },
1055
+ {
1056
+ "epoch": 0.5775339301183945,
1057
+ "grad_norm": 0.20782630145549774,
1058
+ "learning_rate": 3.999714870811639e-06,
1059
+ "loss": 0.7571,
1060
+ "step": 1500
1061
+ },
1062
+ {
1063
+ "epoch": 0.5813841563191837,
1064
+ "grad_norm": 0.21436728537082672,
1065
+ "learning_rate": 3.938696925335275e-06,
1066
+ "loss": 0.7483,
1067
+ "step": 1510
1068
+ },
1069
+ {
1070
+ "epoch": 0.5852343825199731,
1071
+ "grad_norm": 0.2140621542930603,
1072
+ "learning_rate": 3.87784405329962e-06,
1073
+ "loss": 0.7566,
1074
+ "step": 1520
1075
+ },
1076
+ {
1077
+ "epoch": 0.5890846087207623,
1078
+ "grad_norm": 0.21937020123004913,
1079
+ "learning_rate": 3.817165719666382e-06,
1080
+ "loss": 0.7556,
1081
+ "step": 1530
1082
+ },
1083
+ {
1084
+ "epoch": 0.5929348349215516,
1085
+ "grad_norm": 0.20964060723781586,
1086
+ "learning_rate": 3.756671362249841e-06,
1087
+ "loss": 0.7641,
1088
+ "step": 1540
1089
+ },
1090
+ {
1091
+ "epoch": 0.596785061122341,
1092
+ "grad_norm": 0.21703605353832245,
1093
+ "learning_rate": 3.6963703902488907e-06,
1094
+ "loss": 0.7486,
1095
+ "step": 1550
1096
+ },
1097
+ {
1098
+ "epoch": 0.6006352873231302,
1099
+ "grad_norm": 0.2124706506729126,
1100
+ "learning_rate": 3.6362721827835655e-06,
1101
+ "loss": 0.7601,
1102
+ "step": 1560
1103
+ },
1104
+ {
1105
+ "epoch": 0.6044855135239195,
1106
+ "grad_norm": 0.24413390457630157,
1107
+ "learning_rate": 3.576386087436212e-06,
1108
+ "loss": 0.7603,
1109
+ "step": 1570
1110
+ },
1111
+ {
1112
+ "epoch": 0.6083357397247088,
1113
+ "grad_norm": 0.23105426132678986,
1114
+ "learning_rate": 3.5167214187975913e-06,
1115
+ "loss": 0.7497,
1116
+ "step": 1580
1117
+ },
1118
+ {
1119
+ "epoch": 0.6121859659254981,
1120
+ "grad_norm": 0.19321472942829132,
1121
+ "learning_rate": 3.457287457018086e-06,
1122
+ "loss": 0.752,
1123
+ "step": 1590
1124
+ },
1125
+ {
1126
+ "epoch": 0.6160361921262875,
1127
+ "grad_norm": 0.21062351763248444,
1128
+ "learning_rate": 3.3980934463643003e-06,
1129
+ "loss": 0.7416,
1130
+ "step": 1600
1131
+ },
1132
+ {
1133
+ "epoch": 0.6198864183270767,
1134
+ "grad_norm": 0.20000532269477844,
1135
+ "learning_rate": 3.3391485937812053e-06,
1136
+ "loss": 0.7223,
1137
+ "step": 1610
1138
+ },
1139
+ {
1140
+ "epoch": 0.623736644527866,
1141
+ "grad_norm": 0.21732376515865326,
1142
+ "learning_rate": 3.280462067460115e-06,
1143
+ "loss": 0.7218,
1144
+ "step": 1620
1145
+ },
1146
+ {
1147
+ "epoch": 0.6275868707286553,
1148
+ "grad_norm": 0.21343010663986206,
1149
+ "learning_rate": 3.222042995412669e-06,
1150
+ "loss": 0.7536,
1151
+ "step": 1630
1152
+ },
1153
+ {
1154
+ "epoch": 0.6314370969294446,
1155
+ "grad_norm": 0.19068543612957,
1156
+ "learning_rate": 3.163900464051089e-06,
1157
+ "loss": 0.7407,
1158
+ "step": 1640
1159
+ },
1160
+ {
1161
+ "epoch": 0.635287323130234,
1162
+ "grad_norm": 0.22441427409648895,
1163
+ "learning_rate": 3.1060435167748826e-06,
1164
+ "loss": 0.7522,
1165
+ "step": 1650
1166
+ },
1167
+ {
1168
+ "epoch": 0.6391375493310232,
1169
+ "grad_norm": 0.2167486995458603,
1170
+ "learning_rate": 3.048481152564253e-06,
1171
+ "loss": 0.7334,
1172
+ "step": 1660
1173
+ },
1174
+ {
1175
+ "epoch": 0.6429877755318125,
1176
+ "grad_norm": 0.2284378707408905,
1177
+ "learning_rate": 2.9912223245804017e-06,
1178
+ "loss": 0.755,
1179
+ "step": 1670
1180
+ },
1181
+ {
1182
+ "epoch": 0.6468380017326018,
1183
+ "grad_norm": 0.22393666207790375,
1184
+ "learning_rate": 2.934275938772981e-06,
1185
+ "loss": 0.7475,
1186
+ "step": 1680
1187
+ },
1188
+ {
1189
+ "epoch": 0.6506882279333911,
1190
+ "grad_norm": 0.2215895652770996,
1191
+ "learning_rate": 2.8776508524948676e-06,
1192
+ "loss": 0.7383,
1193
+ "step": 1690
1194
+ },
1195
+ {
1196
+ "epoch": 0.6545384541341804,
1197
+ "grad_norm": 0.193879634141922,
1198
+ "learning_rate": 2.821355873124503e-06,
1199
+ "loss": 0.7361,
1200
+ "step": 1700
1201
+ },
1202
+ {
1203
+ "epoch": 0.6583886803349697,
1204
+ "grad_norm": 0.2186749428510666,
1205
+ "learning_rate": 2.7653997566960144e-06,
1206
+ "loss": 0.7515,
1207
+ "step": 1710
1208
+ },
1209
+ {
1210
+ "epoch": 0.662238906535759,
1211
+ "grad_norm": 0.2197483628988266,
1212
+ "learning_rate": 2.709791206537309e-06,
1213
+ "loss": 0.7406,
1214
+ "step": 1720
1215
+ },
1216
+ {
1217
+ "epoch": 0.6660891327365482,
1218
+ "grad_norm": 0.22159917652606964,
1219
+ "learning_rate": 2.6545388719163747e-06,
1220
+ "loss": 0.7427,
1221
+ "step": 1730
1222
+ },
1223
+ {
1224
+ "epoch": 0.6699393589373376,
1225
+ "grad_norm": 0.2196548879146576,
1226
+ "learning_rate": 2.599651346695979e-06,
1227
+ "loss": 0.7441,
1228
+ "step": 1740
1229
+ },
1230
+ {
1231
+ "epoch": 0.6737895851381268,
1232
+ "grad_norm": 0.20640288293361664,
1233
+ "learning_rate": 2.545137167997007e-06,
1234
+ "loss": 0.7432,
1235
+ "step": 1750
1236
+ },
1237
+ {
1238
+ "epoch": 0.6776398113389162,
1239
+ "grad_norm": 0.20737870037555695,
1240
+ "learning_rate": 2.4910048148705972e-06,
1241
+ "loss": 0.7322,
1242
+ "step": 1760
1243
+ },
1244
+ {
1245
+ "epoch": 0.6814900375397055,
1246
+ "grad_norm": 0.22075513005256653,
1247
+ "learning_rate": 2.4372627069793327e-06,
1248
+ "loss": 0.7754,
1249
+ "step": 1770
1250
+ },
1251
+ {
1252
+ "epoch": 0.6853402637404947,
1253
+ "grad_norm": 0.23360034823417664,
1254
+ "learning_rate": 2.3839192032876486e-06,
1255
+ "loss": 0.7605,
1256
+ "step": 1780
1257
+ },
1258
+ {
1259
+ "epoch": 0.6891904899412841,
1260
+ "grad_norm": 0.21007204055786133,
1261
+ "learning_rate": 2.3309826007617047e-06,
1262
+ "loss": 0.7703,
1263
+ "step": 1790
1264
+ },
1265
+ {
1266
+ "epoch": 0.6930407161420733,
1267
+ "grad_norm": 0.2110135555267334,
1268
+ "learning_rate": 2.278461133078881e-06,
1269
+ "loss": 0.7327,
1270
+ "step": 1800
1271
+ },
1272
+ {
1273
+ "epoch": 0.6968909423428626,
1274
+ "grad_norm": 0.22355303168296814,
1275
+ "learning_rate": 2.2263629693471258e-06,
1276
+ "loss": 0.7257,
1277
+ "step": 1810
1278
+ },
1279
+ {
1280
+ "epoch": 0.700741168543652,
1281
+ "grad_norm": 0.21064426004886627,
1282
+ "learning_rate": 2.1746962128343414e-06,
1283
+ "loss": 0.7472,
1284
+ "step": 1820
1285
+ },
1286
+ {
1287
+ "epoch": 0.7045913947444412,
1288
+ "grad_norm": 0.20727697014808655,
1289
+ "learning_rate": 2.1234688997080183e-06,
1290
+ "loss": 0.7383,
1291
+ "step": 1830
1292
+ },
1293
+ {
1294
+ "epoch": 0.7084416209452306,
1295
+ "grad_norm": 0.22939011454582214,
1296
+ "learning_rate": 2.0726889977853053e-06,
1297
+ "loss": 0.7438,
1298
+ "step": 1840
1299
+ },
1300
+ {
1301
+ "epoch": 0.7122918471460198,
1302
+ "grad_norm": 0.2254185825586319,
1303
+ "learning_rate": 2.022364405293703e-06,
1304
+ "loss": 0.7551,
1305
+ "step": 1850
1306
+ },
1307
+ {
1308
+ "epoch": 0.7161420733468091,
1309
+ "grad_norm": 0.21334202587604523,
1310
+ "learning_rate": 1.9725029496425814e-06,
1311
+ "loss": 0.7497,
1312
+ "step": 1860
1313
+ },
1314
+ {
1315
+ "epoch": 0.7199922995475985,
1316
+ "grad_norm": 0.19662483036518097,
1317
+ "learning_rate": 1.923112386205727e-06,
1318
+ "loss": 0.7444,
1319
+ "step": 1870
1320
+ },
1321
+ {
1322
+ "epoch": 0.7238425257483877,
1323
+ "grad_norm": 0.20636425912380219,
1324
+ "learning_rate": 1.874200397115079e-06,
1325
+ "loss": 0.7535,
1326
+ "step": 1880
1327
+ },
1328
+ {
1329
+ "epoch": 0.727692751949177,
1330
+ "grad_norm": 0.23089876770973206,
1331
+ "learning_rate": 1.825774590065857e-06,
1332
+ "loss": 0.7361,
1333
+ "step": 1890
1334
+ },
1335
+ {
1336
+ "epoch": 0.7315429781499663,
1337
+ "grad_norm": 0.22025808691978455,
1338
+ "learning_rate": 1.7778424971332813e-06,
1339
+ "loss": 0.7462,
1340
+ "step": 1900
1341
+ },
1342
+ {
1343
+ "epoch": 0.7353932043507556,
1344
+ "grad_norm": 0.20464438199996948,
1345
+ "learning_rate": 1.7304115736010407e-06,
1346
+ "loss": 0.7425,
1347
+ "step": 1910
1348
+ },
1349
+ {
1350
+ "epoch": 0.739243430551545,
1351
+ "grad_norm": 0.21225477755069733,
1352
+ "learning_rate": 1.6834891968017108e-06,
1353
+ "loss": 0.7391,
1354
+ "step": 1920
1355
+ },
1356
+ {
1357
+ "epoch": 0.7430936567523342,
1358
+ "grad_norm": 0.2213049978017807,
1359
+ "learning_rate": 1.6370826649692884e-06,
1360
+ "loss": 0.7512,
1361
+ "step": 1930
1362
+ },
1363
+ {
1364
+ "epoch": 0.7469438829531235,
1365
+ "grad_norm": 0.22514623403549194,
1366
+ "learning_rate": 1.5911991961040463e-06,
1367
+ "loss": 0.7612,
1368
+ "step": 1940
1369
+ },
1370
+ {
1371
+ "epoch": 0.7507941091539128,
1372
+ "grad_norm": 0.218727245926857,
1373
+ "learning_rate": 1.545845926849851e-06,
1374
+ "loss": 0.7482,
1375
+ "step": 1950
1376
+ },
1377
+ {
1378
+ "epoch": 0.7546443353547021,
1379
+ "grad_norm": 0.23346175253391266,
1380
+ "learning_rate": 1.5010299113841397e-06,
1381
+ "loss": 0.7646,
1382
+ "step": 1960
1383
+ },
1384
+ {
1385
+ "epoch": 0.7584945615554913,
1386
+ "grad_norm": 0.2150743007659912,
1387
+ "learning_rate": 1.4567581203207238e-06,
1388
+ "loss": 0.7544,
1389
+ "step": 1970
1390
+ },
1391
+ {
1392
+ "epoch": 0.7623447877562807,
1393
+ "grad_norm": 0.23959948122501373,
1394
+ "learning_rate": 1.413037439625593e-06,
1395
+ "loss": 0.7539,
1396
+ "step": 1980
1397
+ },
1398
+ {
1399
+ "epoch": 0.76619501395707,
1400
+ "grad_norm": 0.2288261204957962,
1401
+ "learning_rate": 1.3698746695458847e-06,
1402
+ "loss": 0.747,
1403
+ "step": 1990
1404
+ },
1405
+ {
1406
+ "epoch": 0.7700452401578592,
1407
+ "grad_norm": 0.21261751651763916,
1408
+ "learning_rate": 1.3272765235521745e-06,
1409
+ "loss": 0.7397,
1410
+ "step": 2000
1411
+ },
1412
+ {
1413
+ "epoch": 0.7738954663586486,
1414
+ "grad_norm": 0.21261800825595856,
1415
+ "learning_rate": 1.2852496272942877e-06,
1416
+ "loss": 0.7491,
1417
+ "step": 2010
1418
+ },
1419
+ {
1420
+ "epoch": 0.7777456925594378,
1421
+ "grad_norm": 0.20133650302886963,
1422
+ "learning_rate": 1.2438005175707396e-06,
1423
+ "loss": 0.7339,
1424
+ "step": 2020
1425
+ },
1426
+ {
1427
+ "epoch": 0.7815959187602272,
1428
+ "grad_norm": 0.21505649387836456,
1429
+ "learning_rate": 1.2029356413120297e-06,
1430
+ "loss": 0.7196,
1431
+ "step": 2030
1432
+ },
1433
+ {
1434
+ "epoch": 0.7854461449610165,
1435
+ "grad_norm": 0.21872040629386902,
1436
+ "learning_rate": 1.1626613545778786e-06,
1437
+ "loss": 0.7368,
1438
+ "step": 2040
1439
+ },
1440
+ {
1441
+ "epoch": 0.7892963711618057,
1442
+ "grad_norm": 0.20398880541324615,
1443
+ "learning_rate": 1.1229839215686294e-06,
1444
+ "loss": 0.7404,
1445
+ "step": 2050
1446
+ },
1447
+ {
1448
+ "epoch": 0.7931465973625951,
1449
+ "grad_norm": 0.19927100837230682,
1450
+ "learning_rate": 1.0839095136509188e-06,
1451
+ "loss": 0.7237,
1452
+ "step": 2060
1453
+ },
1454
+ {
1455
+ "epoch": 0.7969968235633843,
1456
+ "grad_norm": 0.2335076928138733,
1457
+ "learning_rate": 1.045444208397791e-06,
1458
+ "loss": 0.7464,
1459
+ "step": 2070
1460
+ },
1461
+ {
1462
+ "epoch": 0.8008470497641736,
1463
+ "grad_norm": 0.20840466022491455,
1464
+ "learning_rate": 1.007593988643402e-06,
1465
+ "loss": 0.7309,
1466
+ "step": 2080
1467
+ },
1468
+ {
1469
+ "epoch": 0.804697275964963,
1470
+ "grad_norm": 0.20391108095645905,
1471
+ "learning_rate": 9.703647415524613e-07,
1472
+ "loss": 0.7523,
1473
+ "step": 2090
1474
+ },
1475
+ {
1476
+ "epoch": 0.8085475021657522,
1477
+ "grad_norm": 0.22041288018226624,
1478
+ "learning_rate": 9.337622577045485e-07,
1479
+ "loss": 0.7402,
1480
+ "step": 2100
1481
+ },
1482
+ {
1483
+ "epoch": 0.8123977283665416,
1484
+ "grad_norm": 0.21443821489810944,
1485
+ "learning_rate": 8.977922301934588e-07,
1486
+ "loss": 0.7523,
1487
+ "step": 2110
1488
+ },
1489
+ {
1490
+ "epoch": 0.8162479545673308,
1491
+ "grad_norm": 0.21505364775657654,
1492
+ "learning_rate": 8.624602537417009e-07,
1493
+ "loss": 0.7521,
1494
+ "step": 2120
1495
+ },
1496
+ {
1497
+ "epoch": 0.8200981807681201,
1498
+ "grad_norm": 0.23077352344989777,
1499
+ "learning_rate": 8.277718238303123e-07,
1500
+ "loss": 0.737,
1501
+ "step": 2130
1502
+ },
1503
+ {
1504
+ "epoch": 0.8239484069689095,
1505
+ "grad_norm": 0.21526844799518585,
1506
+ "learning_rate": 7.937323358440935e-07,
1507
+ "loss": 0.7414,
1508
+ "step": 2140
1509
+ },
1510
+ {
1511
+ "epoch": 0.8277986331696987,
1512
+ "grad_norm": 0.23636110126972198,
1513
+ "learning_rate": 7.603470842324223e-07,
1514
+ "loss": 0.7319,
1515
+ "step": 2150
1516
+ },
1517
+ {
1518
+ "epoch": 0.831648859370488,
1519
+ "grad_norm": 0.22079512476921082,
1520
+ "learning_rate": 7.276212616857575e-07,
1521
+ "loss": 0.752,
1522
+ "step": 2160
1523
+ },
1524
+ {
1525
+ "epoch": 0.8354990855712773,
1526
+ "grad_norm": 0.21660324931144714,
1527
+ "learning_rate": 6.955599583279837e-07,
1528
+ "loss": 0.7502,
1529
+ "step": 2170
1530
+ },
1531
+ {
1532
+ "epoch": 0.8393493117720666,
1533
+ "grad_norm": 0.20978274941444397,
1534
+ "learning_rate": 6.641681609246981e-07,
1535
+ "loss": 0.7431,
1536
+ "step": 2180
1537
+ },
1538
+ {
1539
+ "epoch": 0.8431995379728559,
1540
+ "grad_norm": 0.21244993805885315,
1541
+ "learning_rate": 6.334507521075711e-07,
1542
+ "loss": 0.7491,
1543
+ "step": 2190
1544
+ },
1545
+ {
1546
+ "epoch": 0.8470497641736452,
1547
+ "grad_norm": 0.22180497646331787,
1548
+ "learning_rate": 6.034125096149179e-07,
1549
+ "loss": 0.7454,
1550
+ "step": 2200
1551
+ },
1552
+ {
1553
+ "epoch": 0.8508999903744345,
1554
+ "grad_norm": 0.21428000926971436,
1555
+ "learning_rate": 5.740581055485722e-07,
1556
+ "loss": 0.7143,
1557
+ "step": 2210
1558
+ },
1559
+ {
1560
+ "epoch": 0.8547502165752238,
1561
+ "grad_norm": 0.20056380331516266,
1562
+ "learning_rate": 5.453921056471961e-07,
1563
+ "loss": 0.7482,
1564
+ "step": 2220
1565
+ },
1566
+ {
1567
+ "epoch": 0.8586004427760131,
1568
+ "grad_norm": 0.21499356627464294,
1569
+ "learning_rate": 5.174189685761288e-07,
1570
+ "loss": 0.7451,
1571
+ "step": 2230
1572
+ },
1573
+ {
1574
+ "epoch": 0.8624506689768023,
1575
+ "grad_norm": 0.20503699779510498,
1576
+ "learning_rate": 4.901430452338979e-07,
1577
+ "loss": 0.7562,
1578
+ "step": 2240
1579
+ },
1580
+ {
1581
+ "epoch": 0.8663008951775917,
1582
+ "grad_norm": 0.20474731922149658,
1583
+ "learning_rate": 4.6356857807548286e-07,
1584
+ "loss": 0.7394,
1585
+ "step": 2250
1586
+ },
1587
+ {
1588
+ "epoch": 0.870151121378381,
1589
+ "grad_norm": 0.21160638332366943,
1590
+ "learning_rate": 4.376997004524541e-07,
1591
+ "loss": 0.729,
1592
+ "step": 2260
1593
+ },
1594
+ {
1595
+ "epoch": 0.8740013475791703,
1596
+ "grad_norm": 0.21231776475906372,
1597
+ "learning_rate": 4.12540435970073e-07,
1598
+ "loss": 0.735,
1599
+ "step": 2270
1600
+ },
1601
+ {
1602
+ "epoch": 0.8778515737799596,
1603
+ "grad_norm": 0.22009672224521637,
1604
+ "learning_rate": 3.880946978614703e-07,
1605
+ "loss": 0.7488,
1606
+ "step": 2280
1607
+ },
1608
+ {
1609
+ "epoch": 0.8817017999807488,
1610
+ "grad_norm": 0.21584372222423553,
1611
+ "learning_rate": 3.643662883789878e-07,
1612
+ "loss": 0.7502,
1613
+ "step": 2290
1614
+ },
1615
+ {
1616
+ "epoch": 0.8855520261815382,
1617
+ "grad_norm": 0.1994863897562027,
1618
+ "learning_rate": 3.41358898202781e-07,
1619
+ "loss": 0.7299,
1620
+ "step": 2300
1621
+ },
1622
+ {
1623
+ "epoch": 0.8894022523823275,
1624
+ "grad_norm": 0.2026340663433075,
1625
+ "learning_rate": 3.190761058667741e-07,
1626
+ "loss": 0.7412,
1627
+ "step": 2310
1628
+ },
1629
+ {
1630
+ "epoch": 0.8932524785831167,
1631
+ "grad_norm": 0.2083987593650818,
1632
+ "learning_rate": 2.9752137720206463e-07,
1633
+ "loss": 0.7454,
1634
+ "step": 2320
1635
+ },
1636
+ {
1637
+ "epoch": 0.8971027047839061,
1638
+ "grad_norm": 0.20857802033424377,
1639
+ "learning_rate": 2.7669806479785067e-07,
1640
+ "loss": 0.7509,
1641
+ "step": 2330
1642
+ },
1643
+ {
1644
+ "epoch": 0.9009529309846953,
1645
+ "grad_norm": 0.21632817387580872,
1646
+ "learning_rate": 2.566094074799763e-07,
1647
+ "loss": 0.7424,
1648
+ "step": 2340
1649
+ },
1650
+ {
1651
+ "epoch": 0.9048031571854847,
1652
+ "grad_norm": 0.21334072947502136,
1653
+ "learning_rate": 2.3725852980716557e-07,
1654
+ "loss": 0.7301,
1655
+ "step": 2350
1656
+ },
1657
+ {
1658
+ "epoch": 0.908653383386274,
1659
+ "grad_norm": 0.20961429178714752,
1660
+ "learning_rate": 2.1864844158503983e-07,
1661
+ "loss": 0.7374,
1662
+ "step": 2360
1663
+ },
1664
+ {
1665
+ "epoch": 0.9125036095870632,
1666
+ "grad_norm": 0.23114798963069916,
1667
+ "learning_rate": 2.007820373979702e-07,
1668
+ "loss": 0.733,
1669
+ "step": 2370
1670
+ },
1671
+ {
1672
+ "epoch": 0.9163538357878526,
1673
+ "grad_norm": 0.21847453713417053,
1674
+ "learning_rate": 1.8366209615886532e-07,
1675
+ "loss": 0.7431,
1676
+ "step": 2380
1677
+ },
1678
+ {
1679
+ "epoch": 0.9202040619886418,
1680
+ "grad_norm": 0.21445241570472717,
1681
+ "learning_rate": 1.6729128067693613e-07,
1682
+ "loss": 0.757,
1683
+ "step": 2390
1684
+ },
1685
+ {
1686
+ "epoch": 0.9240542881894311,
1687
+ "grad_norm": 0.2274923324584961,
1688
+ "learning_rate": 1.5167213724353426e-07,
1689
+ "loss": 0.7544,
1690
+ "step": 2400
1691
+ },
1692
+ {
1693
+ "epoch": 0.9279045143902205,
1694
+ "grad_norm": 0.21417398750782013,
1695
+ "learning_rate": 1.3680709523610058e-07,
1696
+ "loss": 0.733,
1697
+ "step": 2410
1698
+ },
1699
+ {
1700
+ "epoch": 0.9317547405910097,
1701
+ "grad_norm": 0.21389243006706238,
1702
+ "learning_rate": 1.2269846674030418e-07,
1703
+ "loss": 0.7592,
1704
+ "step": 2420
1705
+ },
1706
+ {
1707
+ "epoch": 0.935604966791799,
1708
+ "grad_norm": 0.2238440066576004,
1709
+ "learning_rate": 1.0934844619042595e-07,
1710
+ "loss": 0.754,
1711
+ "step": 2430
1712
+ },
1713
+ {
1714
+ "epoch": 0.9394551929925883,
1715
+ "grad_norm": 0.21876788139343262,
1716
+ "learning_rate": 9.675911002803961e-08,
1717
+ "loss": 0.7416,
1718
+ "step": 2440
1719
+ },
1720
+ {
1721
+ "epoch": 0.9433054191933776,
1722
+ "grad_norm": 0.23330867290496826,
1723
+ "learning_rate": 8.493241637904381e-08,
1724
+ "loss": 0.738,
1725
+ "step": 2450
1726
+ },
1727
+ {
1728
+ "epoch": 0.9471556453941669,
1729
+ "grad_norm": 0.21204282343387604,
1730
+ "learning_rate": 7.387020474909689e-08,
1731
+ "loss": 0.7606,
1732
+ "step": 2460
1733
+ },
1734
+ {
1735
+ "epoch": 0.9510058715949562,
1736
+ "grad_norm": 0.23525182902812958,
1737
+ "learning_rate": 6.357419573750856e-08,
1738
+ "loss": 0.7549,
1739
+ "step": 2470
1740
+ },
1741
+ {
1742
+ "epoch": 0.9548560977957455,
1743
+ "grad_norm": 0.21425315737724304,
1744
+ "learning_rate": 5.4045990769613365e-08,
1745
+ "loss": 0.736,
1746
+ "step": 2480
1747
+ },
1748
+ {
1749
+ "epoch": 0.9587063239965348,
1750
+ "grad_norm": 0.21619154512882233,
1751
+ "learning_rate": 4.5287071847694434e-08,
1752
+ "loss": 0.7415,
1753
+ "step": 2490
1754
+ },
1755
+ {
1756
+ "epoch": 0.9625565501973241,
1757
+ "grad_norm": 0.21400199830532074,
1758
+ "learning_rate": 3.7298801320467817e-08,
1759
+ "loss": 0.7405,
1760
+ "step": 2500
1761
+ },
1762
+ {
1763
+ "epoch": 0.9664067763981133,
1764
+ "grad_norm": 0.21438342332839966,
1765
+ "learning_rate": 3.0082421671192576e-08,
1766
+ "loss": 0.7326,
1767
+ "step": 2510
1768
+ },
1769
+ {
1770
+ "epoch": 0.9702570025989027,
1771
+ "grad_norm": 0.21002815663814545,
1772
+ "learning_rate": 2.3639055324410355e-08,
1773
+ "loss": 0.7536,
1774
+ "step": 2520
1775
+ },
1776
+ {
1777
+ "epoch": 0.974107228799692,
1778
+ "grad_norm": 0.21058446168899536,
1779
+ "learning_rate": 1.7969704471369477e-08,
1780
+ "loss": 0.7382,
1781
+ "step": 2530
1782
+ },
1783
+ {
1784
+ "epoch": 0.9779574550004813,
1785
+ "grad_norm": 0.2074694186449051,
1786
+ "learning_rate": 1.3075250914142413e-08,
1787
+ "loss": 0.7401,
1788
+ "step": 2540
1789
+ },
1790
+ {
1791
+ "epoch": 0.9818076812012706,
1792
+ "grad_norm": 0.20972532033920288,
1793
+ "learning_rate": 8.956455928476049e-09,
1794
+ "loss": 0.75,
1795
+ "step": 2550
1796
+ },
1797
+ {
1798
+ "epoch": 0.9856579074020598,
1799
+ "grad_norm": 0.20573227107524872,
1800
+ "learning_rate": 5.6139601453786275e-09,
1801
+ "loss": 0.74,
1802
+ "step": 2560
1803
+ },
1804
+ {
1805
+ "epoch": 0.9895081336028492,
1806
+ "grad_norm": 0.21348406374454498,
1807
+ "learning_rate": 3.0482834514805602e-09,
1808
+ "loss": 0.7314,
1809
+ "step": 2570
1810
+ },
1811
+ {
1812
+ "epoch": 0.9933583598036385,
1813
+ "grad_norm": 0.20883937180042267,
1814
+ "learning_rate": 1.2598249081713366e-09,
1815
+ "loss": 0.7208,
1816
+ "step": 2580
1817
+ },
1818
+ {
1819
+ "epoch": 0.9972085860044277,
1820
+ "grad_norm": 0.22162315249443054,
1821
+ "learning_rate": 2.488626895302826e-10,
1822
+ "loss": 0.7557,
1823
+ "step": 2590
1824
+ },
1825
+ {
1826
+ "epoch": 0.9999037443449803,
1827
+ "step": 2597,
1828
+ "total_flos": 3.3602472712444314e+18,
1829
+ "train_loss": 0.7722201674178237,
1830
+ "train_runtime": 11503.296,
1831
+ "train_samples_per_second": 57.801,
1832
+ "train_steps_per_second": 0.226
1833
+ }
1834
+ ],
1835
+ "logging_steps": 10,
1836
+ "max_steps": 2597,
1837
+ "num_input_tokens_seen": 418855352,
1838
+ "num_train_epochs": 1,
1839
+ "save_steps": 500,
1840
+ "stateful_callbacks": {
1841
+ "TrainerControl": {
1842
+ "args": {
1843
+ "should_epoch_stop": false,
1844
+ "should_evaluate": false,
1845
+ "should_log": false,
1846
+ "should_save": true,
1847
+ "should_training_stop": true
1848
+ },
1849
+ "attributes": {}
1850
+ }
1851
+ },
1852
+ "total_flos": 3.3602472712444314e+18,
1853
+ "train_batch_size": 8,
1854
+ "trial_name": null,
1855
+ "trial_params": null
1856
+ }
continuous/qwen25/pe_core_s16_384/finetune/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b1ad86b398366f6d3af5a524ef86c09c392dd65e86fd9ec4d383830796614f6
3
+ size 6264
continuous/qwen25/pe_core_s16_384/finetune/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
continuous/qwen25/pe_core_s16_384/pretrain/config.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_faster_video": false,
3
+ "add_time_instruction": false,
4
+ "architectures": [
5
+ "Qwen2ForCausalLM"
6
+ ],
7
+ "attention_dropout": 0.0,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151645,
10
+ "faster_token_stride": 10,
11
+ "force_sample": false,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 1536,
14
+ "image_aspect_ratio": "square",
15
+ "image_crop_resolution": null,
16
+ "image_grid_pinpoints": null,
17
+ "image_split_resolution": null,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 8960,
20
+ "layer_types": [
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
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention"
49
+ ],
50
+ "max_position_embeddings": 32768,
51
+ "max_window_layers": 21,
52
+ "mm_hidden_size": 384,
53
+ "mm_newline_position": "grid",
54
+ "mm_patch_merge_type": "flat",
55
+ "mm_projector_lr": null,
56
+ "mm_projector_type": "mlp2x_gelu",
57
+ "mm_resampler_type": null,
58
+ "mm_spatial_pool_mode": "bilinear",
59
+ "mm_spatial_pool_stride": null,
60
+ "mm_tunable_parts": "mm_mlp_adapter",
61
+ "mm_use_im_patch_token": false,
62
+ "mm_use_im_start_end": false,
63
+ "mm_vision_select_feature": "patch",
64
+ "mm_vision_select_layer": -1,
65
+ "mm_vision_tower": "vtb_ssl:pe",
66
+ "mm_vision_tower_lr": null,
67
+ "model_type": "qwen2",
68
+ "num_attention_heads": 12,
69
+ "num_hidden_layers": 28,
70
+ "num_key_value_heads": 2,
71
+ "pad_token_id": 151643,
72
+ "pos_skipping_range": 4096,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_scaling": null,
75
+ "rope_theta": 1000000.0,
76
+ "sliding_window": null,
77
+ "tie_word_embeddings": true,
78
+ "tokenizer_model_max_length": 8192,
79
+ "tokenizer_padding_side": "right",
80
+ "transformers_version": "4.57.6",
81
+ "use_cache": true,
82
+ "use_mm_proj": true,
83
+ "use_pos_skipping": false,
84
+ "use_sliding_window": false,
85
+ "vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/pe/PE-Core-S16-384/PE-Core-S16-384.pt",
86
+ "vocab_size": 151936
87
+ }
continuous/qwen25/pe_core_s16_384/pretrain/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87c66cf6654da598928cdec9913998d770223fc7b90a9e4c717a91f73deaa0b5
3
+ size 11810872
continuous/qwen25/pe_core_s16_384/pretrain/trainer_state.json ADDED
@@ -0,0 +1,421 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 545,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.01834862385321101,
14
+ "grad_norm": 0.4759298264980316,
15
+ "learning_rate": 0.00032142857142857147,
16
+ "loss": 4.7262,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.03669724770642202,
21
+ "grad_norm": 0.25505897402763367,
22
+ "learning_rate": 0.0006785714285714287,
23
+ "loss": 3.672,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.05504587155963303,
28
+ "grad_norm": 0.11046168208122253,
29
+ "learning_rate": 0.0009999907688176173,
30
+ "loss": 3.3699,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.07339449541284404,
35
+ "grad_norm": 0.09747314453125,
36
+ "learning_rate": 0.0009988834393115766,
37
+ "loss": 3.2899,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.09174311926605505,
42
+ "grad_norm": 0.06942719966173172,
43
+ "learning_rate": 0.0009959345572940772,
44
+ "loss": 3.2602,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.11009174311926606,
49
+ "grad_norm": 0.05991967394948006,
50
+ "learning_rate": 0.0009911550081155983,
51
+ "loss": 3.2298,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.12844036697247707,
56
+ "grad_norm": 0.08429647237062454,
57
+ "learning_rate": 0.0009845624347566433,
58
+ "loss": 3.2006,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.14678899082568808,
63
+ "grad_norm": 0.05233258008956909,
64
+ "learning_rate": 0.0009761811727013547,
65
+ "loss": 3.1768,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.1651376146788991,
70
+ "grad_norm": 0.06384216248989105,
71
+ "learning_rate": 0.0009660421601067666,
72
+ "loss": 3.1447,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.1834862385321101,
77
+ "grad_norm": 0.07861265540122986,
78
+ "learning_rate": 0.0009541828235992926,
79
+ "loss": 3.1258,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.2018348623853211,
84
+ "grad_norm": 0.06452345848083496,
85
+ "learning_rate": 0.0009406469401200151,
86
+ "loss": 3.0959,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.22018348623853212,
91
+ "grad_norm": 0.06264937669038773,
92
+ "learning_rate": 0.0009254844753287492,
93
+ "loss": 3.08,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.23853211009174313,
98
+ "grad_norm": 0.08487759530544281,
99
+ "learning_rate": 0.0009087513991633924,
100
+ "loss": 3.0472,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.25688073394495414,
105
+ "grad_norm": 0.06901400536298752,
106
+ "learning_rate": 0.0008905094792353917,
107
+ "loss": 3.0318,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.27522935779816515,
112
+ "grad_norm": 0.08387160301208496,
113
+ "learning_rate": 0.0008708260528239789,
114
+ "loss": 3.0181,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.29357798165137616,
119
+ "grad_norm": 0.08248110115528107,
120
+ "learning_rate": 0.0008497737783108237,
121
+ "loss": 2.9976,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.3119266055045872,
126
+ "grad_norm": 0.07094696164131165,
127
+ "learning_rate": 0.0008274303669726426,
128
+ "loss": 2.9871,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.3302752293577982,
133
+ "grad_norm": 0.08606205880641937,
134
+ "learning_rate": 0.0008038782961218137,
135
+ "loss": 2.9662,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.3486238532110092,
140
+ "grad_norm": 0.09440477192401886,
141
+ "learning_rate": 0.000779204504653894,
142
+ "loss": 2.9557,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.3669724770642202,
147
+ "grad_norm": 0.07502669841051102,
148
+ "learning_rate": 0.0007535000721258791,
149
+ "loss": 2.927,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.3853211009174312,
154
+ "grad_norm": 0.07418882846832275,
155
+ "learning_rate": 0.00072685988254984,
156
+ "loss": 2.9199,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.4036697247706422,
161
+ "grad_norm": 0.07589250802993774,
162
+ "learning_rate": 0.0006993822741429906,
163
+ "loss": 2.9277,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.42201834862385323,
168
+ "grad_norm": 0.06174904853105545,
169
+ "learning_rate": 0.0006711686763270818,
170
+ "loss": 2.9168,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.44036697247706424,
175
+ "grad_norm": 0.0767219215631485,
176
+ "learning_rate": 0.0006423232353170831,
177
+ "loss": 2.9,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.45871559633027525,
182
+ "grad_norm": 0.06774676591157913,
183
+ "learning_rate": 0.0006129524296812335,
184
+ "loss": 2.8837,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.47706422018348627,
189
+ "grad_norm": 0.0824272483587265,
190
+ "learning_rate": 0.0005831646772915651,
191
+ "loss": 2.8908,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.4954128440366973,
196
+ "grad_norm": 0.06711604446172714,
197
+ "learning_rate": 0.0005530699351157809,
198
+ "loss": 2.8674,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.5137614678899083,
203
+ "grad_norm": 0.08274008333683014,
204
+ "learning_rate": 0.0005227792933277943,
205
+ "loss": 2.8705,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.5321100917431193,
210
+ "grad_norm": 0.08067454397678375,
211
+ "learning_rate": 0.0004924045652352092,
212
+ "loss": 2.8503,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.5504587155963303,
217
+ "grad_norm": 0.07841738313436508,
218
+ "learning_rate": 0.0004620578745374633,
219
+ "loss": 2.8472,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.5688073394495413,
224
+ "grad_norm": 0.07220886647701263,
225
+ "learning_rate": 0.0004318512414382059,
226
+ "loss": 2.8527,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.5871559633027523,
231
+ "grad_norm": 0.07472312450408936,
232
+ "learning_rate": 0.0004018961691397148,
233
+ "loss": 2.8491,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.6055045871559633,
238
+ "grad_norm": 0.09096528589725494,
239
+ "learning_rate": 0.0003723032322457458,
240
+ "loss": 2.8306,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.6238532110091743,
245
+ "grad_norm": 0.0760083943605423,
246
+ "learning_rate": 0.0003431816685921625,
247
+ "loss": 2.8344,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.6422018348623854,
252
+ "grad_norm": 0.07564907521009445,
253
+ "learning_rate": 0.0003146389760120416,
254
+ "loss": 2.8357,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.6605504587155964,
259
+ "grad_norm": 0.06661021709442139,
260
+ "learning_rate": 0.00028678051552373485,
261
+ "loss": 2.8348,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.6788990825688074,
266
+ "grad_norm": 0.06174774467945099,
267
+ "learning_rate": 0.0002597091224066581,
268
+ "loss": 2.8238,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.6972477064220184,
273
+ "grad_norm": 0.07245096564292908,
274
+ "learning_rate": 0.0002335247266004629,
275
+ "loss": 2.8327,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.7155963302752294,
280
+ "grad_norm": 0.07246304303407669,
281
+ "learning_rate": 0.000208323983828828,
282
+ "loss": 2.8215,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.7339449541284404,
287
+ "grad_norm": 0.08363640308380127,
288
+ "learning_rate": 0.0001841999188095224,
289
+ "loss": 2.814,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.7522935779816514,
294
+ "grad_norm": 0.06631910055875778,
295
+ "learning_rate": 0.00016124158186777676,
296
+ "loss": 2.8129,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.7706422018348624,
301
+ "grad_norm": 0.06086333096027374,
302
+ "learning_rate": 0.0001395337202205211,
303
+ "loss": 2.8069,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.7889908256880734,
308
+ "grad_norm": 0.06265068799257278,
309
+ "learning_rate": 0.00011915646514489292,
310
+ "loss": 2.808,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.8073394495412844,
315
+ "grad_norm": 0.0720076635479927,
316
+ "learning_rate": 0.00010018503618578818,
317
+ "loss": 2.8175,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.8256880733944955,
322
+ "grad_norm": 0.0660257488489151,
323
+ "learning_rate": 8.268946349432583e-05,
324
+ "loss": 2.8132,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.8440366972477065,
329
+ "grad_norm": 0.06526762992143631,
330
+ "learning_rate": 6.673432932217117e-05,
331
+ "loss": 2.7996,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.8623853211009175,
336
+ "grad_norm": 0.060389626771211624,
337
+ "learning_rate": 5.2378529625954684e-05,
338
+ "loss": 2.7923,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.8807339449541285,
343
+ "grad_norm": 0.06408651173114777,
344
+ "learning_rate": 3.9675056661785556e-05,
345
+ "loss": 2.7972,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.8990825688073395,
350
+ "grad_norm": 0.0650276243686676,
351
+ "learning_rate": 2.8670803372379538e-05,
352
+ "loss": 2.8028,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.9174311926605505,
357
+ "grad_norm": 0.06080158054828644,
358
+ "learning_rate": 1.9406390288876586e-05,
359
+ "loss": 2.8112,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.9357798165137615,
364
+ "grad_norm": 0.061907704919576645,
365
+ "learning_rate": 1.191601558631461e-05,
366
+ "loss": 2.7982,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.9541284403669725,
371
+ "grad_norm": 0.05906208977103233,
372
+ "learning_rate": 6.2273288462559305e-06,
373
+ "loss": 2.7972,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.9724770642201835,
378
+ "grad_norm": 0.06016189604997635,
379
+ "learning_rate": 2.361328992552314e-06,
380
+ "loss": 2.7959,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.9908256880733946,
385
+ "grad_norm": 0.06137807294726372,
386
+ "learning_rate": 3.322867770044202e-07,
387
+ "loss": 2.799,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 1.0,
392
+ "step": 545,
393
+ "total_flos": 2.5588851569328128e+17,
394
+ "train_loss": 2.9753420821023644,
395
+ "train_runtime": 5335.2512,
396
+ "train_samples_per_second": 104.611,
397
+ "train_steps_per_second": 0.102
398
+ }
399
+ ],
400
+ "logging_steps": 10,
401
+ "max_steps": 545,
402
+ "num_input_tokens_seen": 31896544,
403
+ "num_train_epochs": 1,
404
+ "save_steps": 500,
405
+ "stateful_callbacks": {
406
+ "TrainerControl": {
407
+ "args": {
408
+ "should_epoch_stop": false,
409
+ "should_evaluate": false,
410
+ "should_log": false,
411
+ "should_save": false,
412
+ "should_training_stop": false
413
+ },
414
+ "attributes": {}
415
+ }
416
+ },
417
+ "total_flos": 2.5588851569328128e+17,
418
+ "train_batch_size": 32,
419
+ "trial_name": null,
420
+ "trial_params": null
421
+ }
continuous/qwen25/pe_core_t16_384/finetune/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
+ }
continuous/qwen25/pe_core_t16_384/finetune/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
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
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\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|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 %}
continuous/qwen25/pe_core_t16_384/finetune/config.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_faster_video": false,
3
+ "add_time_instruction": false,
4
+ "architectures": [
5
+ "LlavaQwenForCausalLM"
6
+ ],
7
+ "attention_dropout": 0.0,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151645,
10
+ "faster_token_stride": 10,
11
+ "force_sample": false,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 1536,
14
+ "image_aspect_ratio": "square",
15
+ "image_crop_resolution": null,
16
+ "image_grid_pinpoints": null,
17
+ "image_split_resolution": null,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 8960,
20
+ "layer_types": [
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
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention"
49
+ ],
50
+ "max_position_embeddings": 32768,
51
+ "max_window_layers": 21,
52
+ "mm_hidden_size": 192,
53
+ "mm_newline_position": "grid",
54
+ "mm_patch_merge_type": "flat",
55
+ "mm_projector_lr": 0.0001,
56
+ "mm_projector_type": "mlp2x_gelu",
57
+ "mm_resampler_type": null,
58
+ "mm_spatial_pool_mode": "bilinear",
59
+ "mm_spatial_pool_stride": null,
60
+ "mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
61
+ "mm_use_im_patch_token": false,
62
+ "mm_use_im_start_end": false,
63
+ "mm_vision_select_feature": "patch",
64
+ "mm_vision_select_layer": -1,
65
+ "mm_vision_tower": "vtb_ssl:pe",
66
+ "mm_vision_tower_lr": null,
67
+ "model_type": "qwen2",
68
+ "num_attention_heads": 12,
69
+ "num_hidden_layers": 28,
70
+ "num_key_value_heads": 2,
71
+ "pad_token_id": 151643,
72
+ "pos_skipping_range": 4096,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_scaling": null,
75
+ "rope_theta": 1000000.0,
76
+ "sliding_window": null,
77
+ "tie_word_embeddings": true,
78
+ "tokenizer_model_max_length": 8192,
79
+ "tokenizer_padding_side": "right",
80
+ "transformers_version": "4.57.6",
81
+ "use_cache": true,
82
+ "use_mm_proj": true,
83
+ "use_pos_skipping": false,
84
+ "use_sliding_window": false,
85
+ "vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/pe/PE-Core-T16-384/PE-Core-T16-384.pt",
86
+ "vocab_size": 151936
87
+ }
continuous/qwen25/pe_core_t16_384/finetune/generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_sample": true,
3
+ "eos_token_id": [
4
+ 151645,
5
+ 151643
6
+ ],
7
+ "pad_token_id": 151643,
8
+ "repetition_penalty": 1.1,
9
+ "temperature": 0.7,
10
+ "top_k": 20,
11
+ "top_p": 0.8,
12
+ "transformers_version": "4.57.6"
13
+ }
continuous/qwen25/pe_core_t16_384/finetune/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
continuous/qwen25/pe_core_t16_384/finetune/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4024900d44fe5e8d966de9e39c2d8f792fee97b0add22541de208372227ea1c8
3
+ size 3577148632
continuous/qwen25/pe_core_t16_384/finetune/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
+ }
continuous/qwen25/pe_core_t16_384/finetune/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
3
+ size 11421896
continuous/qwen25/pe_core_t16_384/finetune/tokenizer_config.json ADDED
@@ -0,0 +1,208 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ }
181
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "clean_up_tokenization_spaces": false,
199
+ "eos_token": "<|im_end|>",
200
+ "errors": "replace",
201
+ "extra_special_tokens": {},
202
+ "model_max_length": 8192,
203
+ "pad_token": "<|endoftext|>",
204
+ "padding_side": "right",
205
+ "split_special_tokens": false,
206
+ "tokenizer_class": "Qwen2Tokenizer",
207
+ "unk_token": null
208
+ }
continuous/qwen25/pe_core_t16_384/finetune/trainer_state.json ADDED
@@ -0,0 +1,1856 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.9999037443449803,
6
+ "eval_steps": 500,
7
+ "global_step": 2597,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.003850226200789296,
14
+ "grad_norm": 17.367647171020508,
15
+ "learning_rate": 1.153846153846154e-06,
16
+ "loss": 1.8273,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.007700452401578592,
21
+ "grad_norm": 10.829907417297363,
22
+ "learning_rate": 2.435897435897436e-06,
23
+ "loss": 1.5299,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.01155067860236789,
28
+ "grad_norm": 3.5354433059692383,
29
+ "learning_rate": 3.7179487179487184e-06,
30
+ "loss": 1.1236,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.015400904803157185,
35
+ "grad_norm": 0.7277745604515076,
36
+ "learning_rate": 5e-06,
37
+ "loss": 0.9601,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.019251131003946483,
42
+ "grad_norm": 0.3940025269985199,
43
+ "learning_rate": 6.282051282051282e-06,
44
+ "loss": 0.9338,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.02310135720473578,
49
+ "grad_norm": 0.2855851352214813,
50
+ "learning_rate": 7.564102564102564e-06,
51
+ "loss": 0.8896,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.026951583405525073,
56
+ "grad_norm": 0.2450128048658371,
57
+ "learning_rate": 8.846153846153847e-06,
58
+ "loss": 0.8642,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.03080180960631437,
63
+ "grad_norm": 0.2445809692144394,
64
+ "learning_rate": 9.999996111488723e-06,
65
+ "loss": 0.8728,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.034652035807103666,
70
+ "grad_norm": 0.22359246015548706,
71
+ "learning_rate": 9.999529497453782e-06,
72
+ "loss": 0.8528,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.038502262007892966,
77
+ "grad_norm": 0.24374406039714813,
78
+ "learning_rate": 9.998285264324193e-06,
79
+ "loss": 0.8342,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.04235248820868226,
84
+ "grad_norm": 0.20783279836177826,
85
+ "learning_rate": 9.996263605626052e-06,
86
+ "loss": 0.8227,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.04620271440947156,
91
+ "grad_norm": 0.23108923435211182,
92
+ "learning_rate": 9.99346483580503e-06,
93
+ "loss": 0.8352,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.05005294061026085,
98
+ "grad_norm": 0.20994922518730164,
99
+ "learning_rate": 9.989889390177467e-06,
100
+ "loss": 0.8267,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.053903166811050146,
105
+ "grad_norm": 0.24246668815612793,
106
+ "learning_rate": 9.985537824862647e-06,
107
+ "loss": 0.8268,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.057753393011839446,
112
+ "grad_norm": 0.2447115182876587,
113
+ "learning_rate": 9.980410816696317e-06,
114
+ "loss": 0.8097,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.06160361921262874,
119
+ "grad_norm": 0.23438964784145355,
120
+ "learning_rate": 9.97450916312541e-06,
121
+ "loss": 0.8165,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.06545384541341803,
126
+ "grad_norm": 0.2084164023399353,
127
+ "learning_rate": 9.967833782084004e-06,
128
+ "loss": 0.8177,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.06930407161420733,
133
+ "grad_norm": 0.24644213914871216,
134
+ "learning_rate": 9.960385711850554e-06,
135
+ "loss": 0.8452,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.07315429781499663,
140
+ "grad_norm": 0.22420117259025574,
141
+ "learning_rate": 9.952166110886394e-06,
142
+ "loss": 0.8354,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.07700452401578593,
147
+ "grad_norm": 0.2497764527797699,
148
+ "learning_rate": 9.943176257655567e-06,
149
+ "loss": 0.8296,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.08085475021657522,
154
+ "grad_norm": 0.2168266326189041,
155
+ "learning_rate": 9.933417550425952e-06,
156
+ "loss": 0.8187,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.08470497641736452,
161
+ "grad_norm": 0.2279394418001175,
162
+ "learning_rate": 9.9228915070518e-06,
163
+ "loss": 0.8085,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.08855520261815382,
168
+ "grad_norm": 0.22323378920555115,
169
+ "learning_rate": 9.911599764737632e-06,
170
+ "loss": 0.8106,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.09240542881894312,
175
+ "grad_norm": 0.22828063368797302,
176
+ "learning_rate": 9.899544079783606e-06,
177
+ "loss": 0.8048,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.0962556550197324,
182
+ "grad_norm": 0.2285298854112625,
183
+ "learning_rate": 9.886726327312338e-06,
184
+ "loss": 0.8189,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.1001058812205217,
189
+ "grad_norm": 0.21486660838127136,
190
+ "learning_rate": 9.87314850097724e-06,
191
+ "loss": 0.8195,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.103956107421311,
196
+ "grad_norm": 0.2172309309244156,
197
+ "learning_rate": 9.858812712652452e-06,
198
+ "loss": 0.8072,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.10780633362210029,
203
+ "grad_norm": 0.2116362303495407,
204
+ "learning_rate": 9.843721192104341e-06,
205
+ "loss": 0.8082,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.11165655982288959,
210
+ "grad_norm": 0.22169508039951324,
211
+ "learning_rate": 9.827876286644698e-06,
212
+ "loss": 0.8277,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.11550678602367889,
217
+ "grad_norm": 0.20665833353996277,
218
+ "learning_rate": 9.811280460765643e-06,
219
+ "loss": 0.7874,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.11935701222446819,
224
+ "grad_norm": 0.22340384125709534,
225
+ "learning_rate": 9.793936295756292e-06,
226
+ "loss": 0.7977,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.12320723842525748,
231
+ "grad_norm": 0.22323718667030334,
232
+ "learning_rate": 9.775846489301279e-06,
233
+ "loss": 0.8021,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.12705746462604678,
238
+ "grad_norm": 0.22597520053386688,
239
+ "learning_rate": 9.757013855061157e-06,
240
+ "loss": 0.8006,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.13090769082683607,
245
+ "grad_norm": 0.22803010046482086,
246
+ "learning_rate": 9.737441322234758e-06,
247
+ "loss": 0.813,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.13475791702762538,
252
+ "grad_norm": 0.23011109232902527,
253
+ "learning_rate": 9.717131935103605e-06,
254
+ "loss": 0.8113,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.13860814322841467,
259
+ "grad_norm": 0.22403956949710846,
260
+ "learning_rate": 9.696088852558398e-06,
261
+ "loss": 0.7978,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.14245836942920398,
266
+ "grad_norm": 0.218888059258461,
267
+ "learning_rate": 9.674315347607686e-06,
268
+ "loss": 0.8231,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.14630859562999327,
273
+ "grad_norm": 0.22372202575206757,
274
+ "learning_rate": 9.651814806868796e-06,
275
+ "loss": 0.7888,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.15015882183078255,
280
+ "grad_norm": 0.23021449148654938,
281
+ "learning_rate": 9.62859073004107e-06,
282
+ "loss": 0.7938,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.15400904803157187,
287
+ "grad_norm": 0.20330512523651123,
288
+ "learning_rate": 9.604646729361547e-06,
289
+ "loss": 0.7999,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.15785927423236115,
294
+ "grad_norm": 0.21020904183387756,
295
+ "learning_rate": 9.579986529043099e-06,
296
+ "loss": 0.7925,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.16170950043315044,
301
+ "grad_norm": 0.22589834034442902,
302
+ "learning_rate": 9.554613964695189e-06,
303
+ "loss": 0.8039,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.16555972663393975,
308
+ "grad_norm": 0.22565020620822906,
309
+ "learning_rate": 9.528532982727281e-06,
310
+ "loss": 0.7951,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.16940995283472904,
315
+ "grad_norm": 0.21097584068775177,
316
+ "learning_rate": 9.501747639735018e-06,
317
+ "loss": 0.794,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.17326017903551832,
322
+ "grad_norm": 0.2193388044834137,
323
+ "learning_rate": 9.474262101869272e-06,
324
+ "loss": 0.7826,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.17711040523630764,
329
+ "grad_norm": 0.22315821051597595,
330
+ "learning_rate": 9.446080644188141e-06,
331
+ "loss": 0.7877,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.18096063143709692,
336
+ "grad_norm": 0.21498656272888184,
337
+ "learning_rate": 9.417207649992017e-06,
338
+ "loss": 0.796,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.18481085763788624,
343
+ "grad_norm": 0.22002875804901123,
344
+ "learning_rate": 9.387647610141816e-06,
345
+ "loss": 0.7955,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.18866108383867553,
350
+ "grad_norm": 0.208359956741333,
351
+ "learning_rate": 9.35740512236047e-06,
352
+ "loss": 0.7812,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.1925113100394648,
357
+ "grad_norm": 0.19891683757305145,
358
+ "learning_rate": 9.326484890517809e-06,
359
+ "loss": 0.79,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.19636153624025413,
364
+ "grad_norm": 0.22781360149383545,
365
+ "learning_rate": 9.294891723898926e-06,
366
+ "loss": 0.7819,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.2002117624410434,
371
+ "grad_norm": 0.23109783232212067,
372
+ "learning_rate": 9.262630536456153e-06,
373
+ "loss": 0.8113,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.2040619886418327,
378
+ "grad_norm": 0.229313924908638,
379
+ "learning_rate": 9.229706346044749e-06,
380
+ "loss": 0.7859,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.207912214842622,
385
+ "grad_norm": 0.230558380484581,
386
+ "learning_rate": 9.196124273642422e-06,
387
+ "loss": 0.7961,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.2117624410434113,
392
+ "grad_norm": 0.21538856625556946,
393
+ "learning_rate": 9.161889542552842e-06,
394
+ "loss": 0.7988,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.21561266724420058,
399
+ "grad_norm": 0.21000443398952484,
400
+ "learning_rate": 9.127007477593193e-06,
401
+ "loss": 0.8035,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.2194628934449899,
406
+ "grad_norm": 0.23782221972942352,
407
+ "learning_rate": 9.091483504265974e-06,
408
+ "loss": 0.8146,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.22331311964577918,
413
+ "grad_norm": 0.22798803448677063,
414
+ "learning_rate": 9.055323147915116e-06,
415
+ "loss": 0.7911,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.2271633458465685,
420
+ "grad_norm": 0.2225879281759262,
421
+ "learning_rate": 9.018532032866585e-06,
422
+ "loss": 0.8034,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.23101357204735778,
427
+ "grad_norm": 0.22372370958328247,
428
+ "learning_rate": 8.981115881553579e-06,
429
+ "loss": 0.7942,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.23486379824814707,
434
+ "grad_norm": 0.2352064996957779,
435
+ "learning_rate": 8.943080513626476e-06,
436
+ "loss": 0.8147,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.23871402444893638,
441
+ "grad_norm": 0.2146414816379547,
442
+ "learning_rate": 8.904431845047641e-06,
443
+ "loss": 0.7984,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.24256425064972567,
448
+ "grad_norm": 0.23191219568252563,
449
+ "learning_rate": 8.86517588717128e-06,
450
+ "loss": 0.8152,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.24641447685051496,
455
+ "grad_norm": 0.21249859035015106,
456
+ "learning_rate": 8.82531874580844e-06,
457
+ "loss": 0.8037,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.25026470305130427,
462
+ "grad_norm": 0.19937728345394135,
463
+ "learning_rate": 8.784866620277314e-06,
464
+ "loss": 0.7624,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.25411492925209356,
469
+ "grad_norm": 0.2418939769268036,
470
+ "learning_rate": 8.743825802439018e-06,
471
+ "loss": 0.799,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.25796515545288284,
476
+ "grad_norm": 0.24642415344715118,
477
+ "learning_rate": 8.702202675718956e-06,
478
+ "loss": 0.8032,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.26181538165367213,
483
+ "grad_norm": 0.21066072583198547,
484
+ "learning_rate": 8.660003714113959e-06,
485
+ "loss": 0.774,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.26566560785446147,
490
+ "grad_norm": 0.22219420969486237,
491
+ "learning_rate": 8.617235481185323e-06,
492
+ "loss": 0.7974,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.26951583405525076,
497
+ "grad_norm": 0.2208748757839203,
498
+ "learning_rate": 8.573904629037931e-06,
499
+ "loss": 0.8021,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.27336606025604004,
504
+ "grad_norm": 0.21711879968643188,
505
+ "learning_rate": 8.530017897285585e-06,
506
+ "loss": 0.7832,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.27721628645682933,
511
+ "grad_norm": 0.2022019773721695,
512
+ "learning_rate": 8.485582112002746e-06,
513
+ "loss": 0.7881,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.2810665126576186,
518
+ "grad_norm": 0.21869993209838867,
519
+ "learning_rate": 8.440604184662811e-06,
520
+ "loss": 0.7922,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.28491673885840796,
525
+ "grad_norm": 0.219015434384346,
526
+ "learning_rate": 8.395091111063118e-06,
527
+ "loss": 0.7833,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.28876696505919724,
532
+ "grad_norm": 0.20134137570858002,
533
+ "learning_rate": 8.349049970236822e-06,
534
+ "loss": 0.7807,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.29261719125998653,
539
+ "grad_norm": 0.21704059839248657,
540
+ "learning_rate": 8.302487923351842e-06,
541
+ "loss": 0.7734,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.2964674174607758,
546
+ "grad_norm": 0.21385399997234344,
547
+ "learning_rate": 8.255412212597021e-06,
548
+ "loss": 0.7913,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.3003176436615651,
553
+ "grad_norm": 0.20547780394554138,
554
+ "learning_rate": 8.207830160055688e-06,
555
+ "loss": 0.7875,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.3041678698623544,
560
+ "grad_norm": 0.22444429993629456,
561
+ "learning_rate": 8.159749166566786e-06,
562
+ "loss": 0.775,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.30801809606314373,
567
+ "grad_norm": 0.21407656371593475,
568
+ "learning_rate": 8.11117671057377e-06,
569
+ "loss": 0.8013,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.311868322263933,
574
+ "grad_norm": 0.22725602984428406,
575
+ "learning_rate": 8.062120346961404e-06,
576
+ "loss": 0.7944,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.3157185484647223,
581
+ "grad_norm": 0.21397405862808228,
582
+ "learning_rate": 8.012587705880705e-06,
583
+ "loss": 0.7893,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.3195687746655116,
588
+ "grad_norm": 0.21557657420635223,
589
+ "learning_rate": 7.962586491562146e-06,
590
+ "loss": 0.7742,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.3234190008663009,
595
+ "grad_norm": 0.23305575549602509,
596
+ "learning_rate": 7.912124481117357e-06,
597
+ "loss": 0.7789,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.3272692270670902,
602
+ "grad_norm": 0.20947976410388947,
603
+ "learning_rate": 7.861209523329494e-06,
604
+ "loss": 0.7882,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.3311194532678795,
609
+ "grad_norm": 0.21610717475414276,
610
+ "learning_rate": 7.809849537432432e-06,
611
+ "loss": 0.7759,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.3349696794686688,
616
+ "grad_norm": 0.20966877043247223,
617
+ "learning_rate": 7.758052511879035e-06,
618
+ "loss": 0.7844,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.3388199056694581,
623
+ "grad_norm": 0.20292435586452484,
624
+ "learning_rate": 7.705826503098642e-06,
625
+ "loss": 0.779,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.34267013187024736,
630
+ "grad_norm": 0.2059665471315384,
631
+ "learning_rate": 7.653179634243979e-06,
632
+ "loss": 0.7787,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.34652035807103665,
637
+ "grad_norm": 0.19499820470809937,
638
+ "learning_rate": 7.600120093927696e-06,
639
+ "loss": 0.7845,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.350370584271826,
644
+ "grad_norm": 0.2227495163679123,
645
+ "learning_rate": 7.546656134948729e-06,
646
+ "loss": 0.7698,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.3542208104726153,
651
+ "grad_norm": 0.22212427854537964,
652
+ "learning_rate": 7.492796073008661e-06,
653
+ "loss": 0.7711,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.35807103667340456,
658
+ "grad_norm": 0.23213179409503937,
659
+ "learning_rate": 7.438548285418326e-06,
660
+ "loss": 0.7597,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.36192126287419385,
665
+ "grad_norm": 0.2381962239742279,
666
+ "learning_rate": 7.3839212097948e-06,
667
+ "loss": 0.7922,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.36577148907498314,
672
+ "grad_norm": 0.22338318824768066,
673
+ "learning_rate": 7.328923342749039e-06,
674
+ "loss": 0.8059,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.3696217152757725,
679
+ "grad_norm": 0.2308112382888794,
680
+ "learning_rate": 7.273563238564319e-06,
681
+ "loss": 0.7922,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.37347194147656176,
686
+ "grad_norm": 0.2151169627904892,
687
+ "learning_rate": 7.217849507865724e-06,
688
+ "loss": 0.7727,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.37732216767735105,
693
+ "grad_norm": 0.21206025779247284,
694
+ "learning_rate": 7.161790816280858e-06,
695
+ "loss": 0.7771,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.38117239387814034,
700
+ "grad_norm": 0.22582051157951355,
701
+ "learning_rate": 7.1053958830920125e-06,
702
+ "loss": 0.7639,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.3850226200789296,
707
+ "grad_norm": 0.21462911367416382,
708
+ "learning_rate": 7.0486734798799715e-06,
709
+ "loss": 0.7766,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.3888728462797189,
714
+ "grad_norm": 0.20677325129508972,
715
+ "learning_rate": 6.991632429159708e-06,
716
+ "loss": 0.8008,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.39272307248050825,
721
+ "grad_norm": 0.23331701755523682,
722
+ "learning_rate": 6.934281603008132e-06,
723
+ "loss": 0.7947,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.39657329868129754,
728
+ "grad_norm": 0.20991840958595276,
729
+ "learning_rate": 6.876629921684143e-06,
730
+ "loss": 0.7921,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.4004235248820868,
735
+ "grad_norm": 0.22223185002803802,
736
+ "learning_rate": 6.818686352241199e-06,
737
+ "loss": 0.7833,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.4042737510828761,
742
+ "grad_norm": 0.19977134466171265,
743
+ "learning_rate": 6.760459907132573e-06,
744
+ "loss": 0.7794,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.4081239772836654,
749
+ "grad_norm": 0.1991460770368576,
750
+ "learning_rate": 6.701959642809589e-06,
751
+ "loss": 0.7859,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.41197420348445474,
756
+ "grad_norm": 0.2155880331993103,
757
+ "learning_rate": 6.643194658312987e-06,
758
+ "loss": 0.7757,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.415824429685244,
763
+ "grad_norm": 0.21074089407920837,
764
+ "learning_rate": 6.584174093857676e-06,
765
+ "loss": 0.8008,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.4196746558860333,
770
+ "grad_norm": 0.24788056313991547,
771
+ "learning_rate": 6.5249071294110735e-06,
772
+ "loss": 0.7825,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.4235248820868226,
777
+ "grad_norm": 0.22166846692562103,
778
+ "learning_rate": 6.465402983265274e-06,
779
+ "loss": 0.7967,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.4273751082876119,
784
+ "grad_norm": 0.2202390879392624,
785
+ "learning_rate": 6.405670910603243e-06,
786
+ "loss": 0.7932,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.43122533448840117,
791
+ "grad_norm": 0.2386539727449417,
792
+ "learning_rate": 6.345720202059291e-06,
793
+ "loss": 0.7888,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.4350755606891905,
798
+ "grad_norm": 0.21722938120365143,
799
+ "learning_rate": 6.285560182274008e-06,
800
+ "loss": 0.7661,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.4389257868899798,
805
+ "grad_norm": 0.2083626538515091,
806
+ "learning_rate": 6.225200208443936e-06,
807
+ "loss": 0.7879,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.4427760130907691,
812
+ "grad_norm": 0.2138058990240097,
813
+ "learning_rate": 6.16464966886616e-06,
814
+ "loss": 0.7802,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.44662623929155837,
819
+ "grad_norm": 0.24705791473388672,
820
+ "learning_rate": 6.103917981478065e-06,
821
+ "loss": 0.8028,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.45047646549234766,
826
+ "grad_norm": 0.21614623069763184,
827
+ "learning_rate": 6.043014592392489e-06,
828
+ "loss": 0.8149,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.454326691693137,
833
+ "grad_norm": 0.2011115849018097,
834
+ "learning_rate": 5.981948974428485e-06,
835
+ "loss": 0.787,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.4581769178939263,
840
+ "grad_norm": 0.22295816242694855,
841
+ "learning_rate": 5.920730625637934e-06,
842
+ "loss": 0.7681,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.46202714409471557,
847
+ "grad_norm": 0.22437895834445953,
848
+ "learning_rate": 5.8593690678282355e-06,
849
+ "loss": 0.7788,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.46587737029550486,
854
+ "grad_norm": 0.22152379155158997,
855
+ "learning_rate": 5.797873845081301e-06,
856
+ "loss": 0.7813,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.46972759649629414,
861
+ "grad_norm": 0.22674143314361572,
862
+ "learning_rate": 5.736254522269075e-06,
863
+ "loss": 0.7979,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.47357782269708343,
868
+ "grad_norm": 0.20974773168563843,
869
+ "learning_rate": 5.6745206835658386e-06,
870
+ "loss": 0.8013,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.47742804889787277,
875
+ "grad_norm": 0.20750001072883606,
876
+ "learning_rate": 5.612681930957495e-06,
877
+ "loss": 0.7699,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.48127827509866206,
882
+ "grad_norm": 0.20418374240398407,
883
+ "learning_rate": 5.550747882748095e-06,
884
+ "loss": 0.7687,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.48512850129945134,
889
+ "grad_norm": 0.2326849102973938,
890
+ "learning_rate": 5.4887281720638195e-06,
891
+ "loss": 0.7769,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.48897872750024063,
896
+ "grad_norm": 0.24448955059051514,
897
+ "learning_rate": 5.42663244535466e-06,
898
+ "loss": 0.762,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.4928289537010299,
903
+ "grad_norm": 0.21627455949783325,
904
+ "learning_rate": 5.3644703608940205e-06,
905
+ "loss": 0.7696,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.49667917990181926,
910
+ "grad_norm": 0.22086074948310852,
911
+ "learning_rate": 5.302251587276491e-06,
912
+ "loss": 0.7806,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.5005294061026085,
917
+ "grad_norm": 0.21091578900814056,
918
+ "learning_rate": 5.2399858019140005e-06,
919
+ "loss": 0.7823,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.5043796323033978,
924
+ "grad_norm": 0.20973587036132812,
925
+ "learning_rate": 5.177682689530615e-06,
926
+ "loss": 0.7927,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.5082298585041871,
931
+ "grad_norm": 0.22317810356616974,
932
+ "learning_rate": 5.11535194065619e-06,
933
+ "loss": 0.7657,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.5120800847049765,
938
+ "grad_norm": 0.2405206561088562,
939
+ "learning_rate": 5.053003250119113e-06,
940
+ "loss": 0.775,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.5159303109057657,
945
+ "grad_norm": 0.22057868540287018,
946
+ "learning_rate": 4.9906463155383925e-06,
947
+ "loss": 0.7794,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.519780537106555,
952
+ "grad_norm": 0.21699489653110504,
953
+ "learning_rate": 4.9282908358153046e-06,
954
+ "loss": 0.7715,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.5236307633073443,
959
+ "grad_norm": 0.2457379698753357,
960
+ "learning_rate": 4.8659465096248345e-06,
961
+ "loss": 0.791,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.5274809895081336,
966
+ "grad_norm": 0.22284463047981262,
967
+ "learning_rate": 4.803623033907169e-06,
968
+ "loss": 0.7899,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.5313312157089229,
973
+ "grad_norm": 0.20529958605766296,
974
+ "learning_rate": 4.741330102359441e-06,
975
+ "loss": 0.7956,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.5351814419097122,
980
+ "grad_norm": 0.22772520780563354,
981
+ "learning_rate": 4.679077403927996e-06,
982
+ "loss": 0.7753,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.5390316681105015,
987
+ "grad_norm": 0.21363946795463562,
988
+ "learning_rate": 4.616874621301373e-06,
989
+ "loss": 0.7754,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.5428818943112907,
994
+ "grad_norm": 0.23127782344818115,
995
+ "learning_rate": 4.554731429404293e-06,
996
+ "loss": 0.7881,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.5467321205120801,
1001
+ "grad_norm": 0.2314860224723816,
1002
+ "learning_rate": 4.492657493892822e-06,
1003
+ "loss": 0.776,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.5505823467128694,
1008
+ "grad_norm": 0.2294529527425766,
1009
+ "learning_rate": 4.430662469650999e-06,
1010
+ "loss": 0.7903,
1011
+ "step": 1430
1012
+ },
1013
+ {
1014
+ "epoch": 0.5544325729136587,
1015
+ "grad_norm": 0.22614791989326477,
1016
+ "learning_rate": 4.368755999289118e-06,
1017
+ "loss": 0.7926,
1018
+ "step": 1440
1019
+ },
1020
+ {
1021
+ "epoch": 0.558282799114448,
1022
+ "grad_norm": 0.23127935826778412,
1023
+ "learning_rate": 4.306947711643949e-06,
1024
+ "loss": 0.7747,
1025
+ "step": 1450
1026
+ },
1027
+ {
1028
+ "epoch": 0.5621330253152372,
1029
+ "grad_norm": 0.22498902678489685,
1030
+ "learning_rate": 4.245247220281068e-06,
1031
+ "loss": 0.7796,
1032
+ "step": 1460
1033
+ },
1034
+ {
1035
+ "epoch": 0.5659832515160266,
1036
+ "grad_norm": 0.21079044044017792,
1037
+ "learning_rate": 4.183664121999587e-06,
1038
+ "loss": 0.7767,
1039
+ "step": 1470
1040
+ },
1041
+ {
1042
+ "epoch": 0.5698334777168159,
1043
+ "grad_norm": 0.23290404677391052,
1044
+ "learning_rate": 4.122207995339474e-06,
1045
+ "loss": 0.7717,
1046
+ "step": 1480
1047
+ },
1048
+ {
1049
+ "epoch": 0.5736837039176051,
1050
+ "grad_norm": 0.21198002994060516,
1051
+ "learning_rate": 4.060888399091734e-06,
1052
+ "loss": 0.7791,
1053
+ "step": 1490
1054
+ },
1055
+ {
1056
+ "epoch": 0.5775339301183945,
1057
+ "grad_norm": 0.20387519896030426,
1058
+ "learning_rate": 3.999714870811639e-06,
1059
+ "loss": 0.7873,
1060
+ "step": 1500
1061
+ },
1062
+ {
1063
+ "epoch": 0.5813841563191837,
1064
+ "grad_norm": 0.21729393303394318,
1065
+ "learning_rate": 3.938696925335275e-06,
1066
+ "loss": 0.7819,
1067
+ "step": 1510
1068
+ },
1069
+ {
1070
+ "epoch": 0.5852343825199731,
1071
+ "grad_norm": 0.21671172976493835,
1072
+ "learning_rate": 3.87784405329962e-06,
1073
+ "loss": 0.7892,
1074
+ "step": 1520
1075
+ },
1076
+ {
1077
+ "epoch": 0.5890846087207623,
1078
+ "grad_norm": 0.21479076147079468,
1079
+ "learning_rate": 3.817165719666382e-06,
1080
+ "loss": 0.7925,
1081
+ "step": 1530
1082
+ },
1083
+ {
1084
+ "epoch": 0.5929348349215516,
1085
+ "grad_norm": 0.2090505212545395,
1086
+ "learning_rate": 3.756671362249841e-06,
1087
+ "loss": 0.7953,
1088
+ "step": 1540
1089
+ },
1090
+ {
1091
+ "epoch": 0.596785061122341,
1092
+ "grad_norm": 0.2151031643152237,
1093
+ "learning_rate": 3.6963703902488907e-06,
1094
+ "loss": 0.7828,
1095
+ "step": 1550
1096
+ },
1097
+ {
1098
+ "epoch": 0.6006352873231302,
1099
+ "grad_norm": 0.2179129421710968,
1100
+ "learning_rate": 3.6362721827835655e-06,
1101
+ "loss": 0.7999,
1102
+ "step": 1560
1103
+ },
1104
+ {
1105
+ "epoch": 0.6044855135239195,
1106
+ "grad_norm": 0.22582603991031647,
1107
+ "learning_rate": 3.576386087436212e-06,
1108
+ "loss": 0.794,
1109
+ "step": 1570
1110
+ },
1111
+ {
1112
+ "epoch": 0.6083357397247088,
1113
+ "grad_norm": 0.23373280465602875,
1114
+ "learning_rate": 3.5167214187975913e-06,
1115
+ "loss": 0.7836,
1116
+ "step": 1580
1117
+ },
1118
+ {
1119
+ "epoch": 0.6121859659254981,
1120
+ "grad_norm": 0.1938573271036148,
1121
+ "learning_rate": 3.457287457018086e-06,
1122
+ "loss": 0.7851,
1123
+ "step": 1590
1124
+ },
1125
+ {
1126
+ "epoch": 0.6160361921262875,
1127
+ "grad_norm": 0.2112901210784912,
1128
+ "learning_rate": 3.3980934463643003e-06,
1129
+ "loss": 0.7742,
1130
+ "step": 1600
1131
+ },
1132
+ {
1133
+ "epoch": 0.6198864183270767,
1134
+ "grad_norm": 0.20416200160980225,
1135
+ "learning_rate": 3.3391485937812053e-06,
1136
+ "loss": 0.7552,
1137
+ "step": 1610
1138
+ },
1139
+ {
1140
+ "epoch": 0.623736644527866,
1141
+ "grad_norm": 0.21709291636943817,
1142
+ "learning_rate": 3.280462067460115e-06,
1143
+ "loss": 0.7553,
1144
+ "step": 1620
1145
+ },
1146
+ {
1147
+ "epoch": 0.6275868707286553,
1148
+ "grad_norm": 0.21496610343456268,
1149
+ "learning_rate": 3.222042995412669e-06,
1150
+ "loss": 0.7869,
1151
+ "step": 1630
1152
+ },
1153
+ {
1154
+ "epoch": 0.6314370969294446,
1155
+ "grad_norm": 0.19041138887405396,
1156
+ "learning_rate": 3.163900464051089e-06,
1157
+ "loss": 0.7758,
1158
+ "step": 1640
1159
+ },
1160
+ {
1161
+ "epoch": 0.635287323130234,
1162
+ "grad_norm": 0.21960784494876862,
1163
+ "learning_rate": 3.1060435167748826e-06,
1164
+ "loss": 0.7848,
1165
+ "step": 1650
1166
+ },
1167
+ {
1168
+ "epoch": 0.6391375493310232,
1169
+ "grad_norm": 0.2221481204032898,
1170
+ "learning_rate": 3.048481152564253e-06,
1171
+ "loss": 0.7655,
1172
+ "step": 1660
1173
+ },
1174
+ {
1175
+ "epoch": 0.6429877755318125,
1176
+ "grad_norm": 0.22567495703697205,
1177
+ "learning_rate": 2.9912223245804017e-06,
1178
+ "loss": 0.7845,
1179
+ "step": 1670
1180
+ },
1181
+ {
1182
+ "epoch": 0.6468380017326018,
1183
+ "grad_norm": 0.2255728840827942,
1184
+ "learning_rate": 2.934275938772981e-06,
1185
+ "loss": 0.7828,
1186
+ "step": 1680
1187
+ },
1188
+ {
1189
+ "epoch": 0.6506882279333911,
1190
+ "grad_norm": 0.2179878205060959,
1191
+ "learning_rate": 2.8776508524948676e-06,
1192
+ "loss": 0.7716,
1193
+ "step": 1690
1194
+ },
1195
+ {
1196
+ "epoch": 0.6545384541341804,
1197
+ "grad_norm": 0.19325846433639526,
1198
+ "learning_rate": 2.821355873124503e-06,
1199
+ "loss": 0.7738,
1200
+ "step": 1700
1201
+ },
1202
+ {
1203
+ "epoch": 0.6583886803349697,
1204
+ "grad_norm": 0.22007331252098083,
1205
+ "learning_rate": 2.7653997566960144e-06,
1206
+ "loss": 0.7837,
1207
+ "step": 1710
1208
+ },
1209
+ {
1210
+ "epoch": 0.662238906535759,
1211
+ "grad_norm": 0.22229479253292084,
1212
+ "learning_rate": 2.709791206537309e-06,
1213
+ "loss": 0.7737,
1214
+ "step": 1720
1215
+ },
1216
+ {
1217
+ "epoch": 0.6660891327365482,
1218
+ "grad_norm": 0.21648991107940674,
1219
+ "learning_rate": 2.6545388719163747e-06,
1220
+ "loss": 0.7801,
1221
+ "step": 1730
1222
+ },
1223
+ {
1224
+ "epoch": 0.6699393589373376,
1225
+ "grad_norm": 0.22088177502155304,
1226
+ "learning_rate": 2.599651346695979e-06,
1227
+ "loss": 0.7797,
1228
+ "step": 1740
1229
+ },
1230
+ {
1231
+ "epoch": 0.6737895851381268,
1232
+ "grad_norm": 0.2064712643623352,
1233
+ "learning_rate": 2.545137167997007e-06,
1234
+ "loss": 0.7771,
1235
+ "step": 1750
1236
+ },
1237
+ {
1238
+ "epoch": 0.6776398113389162,
1239
+ "grad_norm": 0.2050490528345108,
1240
+ "learning_rate": 2.4910048148705972e-06,
1241
+ "loss": 0.7661,
1242
+ "step": 1760
1243
+ },
1244
+ {
1245
+ "epoch": 0.6814900375397055,
1246
+ "grad_norm": 0.2280055284500122,
1247
+ "learning_rate": 2.4372627069793327e-06,
1248
+ "loss": 0.8081,
1249
+ "step": 1770
1250
+ },
1251
+ {
1252
+ "epoch": 0.6853402637404947,
1253
+ "grad_norm": 0.2262575328350067,
1254
+ "learning_rate": 2.3839192032876486e-06,
1255
+ "loss": 0.7946,
1256
+ "step": 1780
1257
+ },
1258
+ {
1259
+ "epoch": 0.6891904899412841,
1260
+ "grad_norm": 0.21428121626377106,
1261
+ "learning_rate": 2.3309826007617047e-06,
1262
+ "loss": 0.8047,
1263
+ "step": 1790
1264
+ },
1265
+ {
1266
+ "epoch": 0.6930407161420733,
1267
+ "grad_norm": 0.21658219397068024,
1268
+ "learning_rate": 2.278461133078881e-06,
1269
+ "loss": 0.7675,
1270
+ "step": 1800
1271
+ },
1272
+ {
1273
+ "epoch": 0.6968909423428626,
1274
+ "grad_norm": 0.230987548828125,
1275
+ "learning_rate": 2.2263629693471258e-06,
1276
+ "loss": 0.7611,
1277
+ "step": 1810
1278
+ },
1279
+ {
1280
+ "epoch": 0.700741168543652,
1281
+ "grad_norm": 0.20359963178634644,
1282
+ "learning_rate": 2.1746962128343414e-06,
1283
+ "loss": 0.7828,
1284
+ "step": 1820
1285
+ },
1286
+ {
1287
+ "epoch": 0.7045913947444412,
1288
+ "grad_norm": 0.2122824639081955,
1289
+ "learning_rate": 2.1234688997080183e-06,
1290
+ "loss": 0.7717,
1291
+ "step": 1830
1292
+ },
1293
+ {
1294
+ "epoch": 0.7084416209452306,
1295
+ "grad_norm": 0.22804367542266846,
1296
+ "learning_rate": 2.0726889977853053e-06,
1297
+ "loss": 0.7788,
1298
+ "step": 1840
1299
+ },
1300
+ {
1301
+ "epoch": 0.7122918471460198,
1302
+ "grad_norm": 0.23184917867183685,
1303
+ "learning_rate": 2.022364405293703e-06,
1304
+ "loss": 0.788,
1305
+ "step": 1850
1306
+ },
1307
+ {
1308
+ "epoch": 0.7161420733468091,
1309
+ "grad_norm": 0.2160610556602478,
1310
+ "learning_rate": 1.9725029496425814e-06,
1311
+ "loss": 0.7832,
1312
+ "step": 1860
1313
+ },
1314
+ {
1315
+ "epoch": 0.7199922995475985,
1316
+ "grad_norm": 0.19942256808280945,
1317
+ "learning_rate": 1.923112386205727e-06,
1318
+ "loss": 0.7792,
1319
+ "step": 1870
1320
+ },
1321
+ {
1322
+ "epoch": 0.7238425257483877,
1323
+ "grad_norm": 0.207593634724617,
1324
+ "learning_rate": 1.874200397115079e-06,
1325
+ "loss": 0.7871,
1326
+ "step": 1880
1327
+ },
1328
+ {
1329
+ "epoch": 0.727692751949177,
1330
+ "grad_norm": 0.22840920090675354,
1331
+ "learning_rate": 1.825774590065857e-06,
1332
+ "loss": 0.7699,
1333
+ "step": 1890
1334
+ },
1335
+ {
1336
+ "epoch": 0.7315429781499663,
1337
+ "grad_norm": 0.22553792595863342,
1338
+ "learning_rate": 1.7778424971332813e-06,
1339
+ "loss": 0.78,
1340
+ "step": 1900
1341
+ },
1342
+ {
1343
+ "epoch": 0.7353932043507556,
1344
+ "grad_norm": 0.2044847011566162,
1345
+ "learning_rate": 1.7304115736010407e-06,
1346
+ "loss": 0.7784,
1347
+ "step": 1910
1348
+ },
1349
+ {
1350
+ "epoch": 0.739243430551545,
1351
+ "grad_norm": 0.21524064242839813,
1352
+ "learning_rate": 1.6834891968017108e-06,
1353
+ "loss": 0.7709,
1354
+ "step": 1920
1355
+ },
1356
+ {
1357
+ "epoch": 0.7430936567523342,
1358
+ "grad_norm": 0.2226194143295288,
1359
+ "learning_rate": 1.6370826649692884e-06,
1360
+ "loss": 0.7893,
1361
+ "step": 1930
1362
+ },
1363
+ {
1364
+ "epoch": 0.7469438829531235,
1365
+ "grad_norm": 0.2284434288740158,
1366
+ "learning_rate": 1.5911991961040463e-06,
1367
+ "loss": 0.7938,
1368
+ "step": 1940
1369
+ },
1370
+ {
1371
+ "epoch": 0.7507941091539128,
1372
+ "grad_norm": 0.22104844450950623,
1373
+ "learning_rate": 1.545845926849851e-06,
1374
+ "loss": 0.7798,
1375
+ "step": 1950
1376
+ },
1377
+ {
1378
+ "epoch": 0.7546443353547021,
1379
+ "grad_norm": 0.23390527069568634,
1380
+ "learning_rate": 1.5010299113841397e-06,
1381
+ "loss": 0.7994,
1382
+ "step": 1960
1383
+ },
1384
+ {
1385
+ "epoch": 0.7584945615554913,
1386
+ "grad_norm": 0.21723578870296478,
1387
+ "learning_rate": 1.4567581203207238e-06,
1388
+ "loss": 0.7874,
1389
+ "step": 1970
1390
+ },
1391
+ {
1392
+ "epoch": 0.7623447877562807,
1393
+ "grad_norm": 0.23087738454341888,
1394
+ "learning_rate": 1.413037439625593e-06,
1395
+ "loss": 0.7896,
1396
+ "step": 1980
1397
+ },
1398
+ {
1399
+ "epoch": 0.76619501395707,
1400
+ "grad_norm": 0.23346710205078125,
1401
+ "learning_rate": 1.3698746695458847e-06,
1402
+ "loss": 0.7793,
1403
+ "step": 1990
1404
+ },
1405
+ {
1406
+ "epoch": 0.7700452401578592,
1407
+ "grad_norm": 0.21459193527698517,
1408
+ "learning_rate": 1.3272765235521745e-06,
1409
+ "loss": 0.7721,
1410
+ "step": 2000
1411
+ },
1412
+ {
1413
+ "epoch": 0.7738954663586486,
1414
+ "grad_norm": 0.2160487323999405,
1415
+ "learning_rate": 1.2852496272942877e-06,
1416
+ "loss": 0.7809,
1417
+ "step": 2010
1418
+ },
1419
+ {
1420
+ "epoch": 0.7777456925594378,
1421
+ "grad_norm": 0.2031269371509552,
1422
+ "learning_rate": 1.2438005175707396e-06,
1423
+ "loss": 0.7681,
1424
+ "step": 2020
1425
+ },
1426
+ {
1427
+ "epoch": 0.7815959187602272,
1428
+ "grad_norm": 0.2131243795156479,
1429
+ "learning_rate": 1.2029356413120297e-06,
1430
+ "loss": 0.7557,
1431
+ "step": 2030
1432
+ },
1433
+ {
1434
+ "epoch": 0.7854461449610165,
1435
+ "grad_norm": 0.21357999742031097,
1436
+ "learning_rate": 1.1626613545778786e-06,
1437
+ "loss": 0.7715,
1438
+ "step": 2040
1439
+ },
1440
+ {
1441
+ "epoch": 0.7892963711618057,
1442
+ "grad_norm": 0.20606112480163574,
1443
+ "learning_rate": 1.1229839215686294e-06,
1444
+ "loss": 0.7749,
1445
+ "step": 2050
1446
+ },
1447
+ {
1448
+ "epoch": 0.7931465973625951,
1449
+ "grad_norm": 0.197964146733284,
1450
+ "learning_rate": 1.0839095136509188e-06,
1451
+ "loss": 0.7608,
1452
+ "step": 2060
1453
+ },
1454
+ {
1455
+ "epoch": 0.7969968235633843,
1456
+ "grad_norm": 0.23236627876758575,
1457
+ "learning_rate": 1.045444208397791e-06,
1458
+ "loss": 0.7836,
1459
+ "step": 2070
1460
+ },
1461
+ {
1462
+ "epoch": 0.8008470497641736,
1463
+ "grad_norm": 0.2074769288301468,
1464
+ "learning_rate": 1.007593988643402e-06,
1465
+ "loss": 0.7645,
1466
+ "step": 2080
1467
+ },
1468
+ {
1469
+ "epoch": 0.804697275964963,
1470
+ "grad_norm": 0.2059885859489441,
1471
+ "learning_rate": 9.703647415524613e-07,
1472
+ "loss": 0.786,
1473
+ "step": 2090
1474
+ },
1475
+ {
1476
+ "epoch": 0.8085475021657522,
1477
+ "grad_norm": 0.21596188843250275,
1478
+ "learning_rate": 9.337622577045485e-07,
1479
+ "loss": 0.7748,
1480
+ "step": 2100
1481
+ },
1482
+ {
1483
+ "epoch": 0.8123977283665416,
1484
+ "grad_norm": 0.21881628036499023,
1485
+ "learning_rate": 8.977922301934588e-07,
1486
+ "loss": 0.7829,
1487
+ "step": 2110
1488
+ },
1489
+ {
1490
+ "epoch": 0.8162479545673308,
1491
+ "grad_norm": 0.21545195579528809,
1492
+ "learning_rate": 8.624602537417009e-07,
1493
+ "loss": 0.7845,
1494
+ "step": 2120
1495
+ },
1496
+ {
1497
+ "epoch": 0.8200981807681201,
1498
+ "grad_norm": 0.23128840327262878,
1499
+ "learning_rate": 8.277718238303123e-07,
1500
+ "loss": 0.7705,
1501
+ "step": 2130
1502
+ },
1503
+ {
1504
+ "epoch": 0.8239484069689095,
1505
+ "grad_norm": 0.21875283122062683,
1506
+ "learning_rate": 7.937323358440935e-07,
1507
+ "loss": 0.7753,
1508
+ "step": 2140
1509
+ },
1510
+ {
1511
+ "epoch": 0.8277986331696987,
1512
+ "grad_norm": 0.23091091215610504,
1513
+ "learning_rate": 7.603470842324223e-07,
1514
+ "loss": 0.7667,
1515
+ "step": 2150
1516
+ },
1517
+ {
1518
+ "epoch": 0.831648859370488,
1519
+ "grad_norm": 0.21720926463603973,
1520
+ "learning_rate": 7.276212616857575e-07,
1521
+ "loss": 0.7846,
1522
+ "step": 2160
1523
+ },
1524
+ {
1525
+ "epoch": 0.8354990855712773,
1526
+ "grad_norm": 0.21350611746311188,
1527
+ "learning_rate": 6.955599583279837e-07,
1528
+ "loss": 0.7871,
1529
+ "step": 2170
1530
+ },
1531
+ {
1532
+ "epoch": 0.8393493117720666,
1533
+ "grad_norm": 0.21267233788967133,
1534
+ "learning_rate": 6.641681609246981e-07,
1535
+ "loss": 0.7754,
1536
+ "step": 2180
1537
+ },
1538
+ {
1539
+ "epoch": 0.8431995379728559,
1540
+ "grad_norm": 0.21710823476314545,
1541
+ "learning_rate": 6.334507521075711e-07,
1542
+ "loss": 0.7859,
1543
+ "step": 2190
1544
+ },
1545
+ {
1546
+ "epoch": 0.8470497641736452,
1547
+ "grad_norm": 0.2263507843017578,
1548
+ "learning_rate": 6.034125096149179e-07,
1549
+ "loss": 0.7787,
1550
+ "step": 2200
1551
+ },
1552
+ {
1553
+ "epoch": 0.8508999903744345,
1554
+ "grad_norm": 0.2152056097984314,
1555
+ "learning_rate": 5.740581055485722e-07,
1556
+ "loss": 0.7488,
1557
+ "step": 2210
1558
+ },
1559
+ {
1560
+ "epoch": 0.8547502165752238,
1561
+ "grad_norm": 0.2016780525445938,
1562
+ "learning_rate": 5.453921056471961e-07,
1563
+ "loss": 0.7825,
1564
+ "step": 2220
1565
+ },
1566
+ {
1567
+ "epoch": 0.8586004427760131,
1568
+ "grad_norm": 0.21628202497959137,
1569
+ "learning_rate": 5.174189685761288e-07,
1570
+ "loss": 0.7778,
1571
+ "step": 2230
1572
+ },
1573
+ {
1574
+ "epoch": 0.8624506689768023,
1575
+ "grad_norm": 0.20691245794296265,
1576
+ "learning_rate": 4.901430452338979e-07,
1577
+ "loss": 0.7876,
1578
+ "step": 2240
1579
+ },
1580
+ {
1581
+ "epoch": 0.8663008951775917,
1582
+ "grad_norm": 0.20736122131347656,
1583
+ "learning_rate": 4.6356857807548286e-07,
1584
+ "loss": 0.7737,
1585
+ "step": 2250
1586
+ },
1587
+ {
1588
+ "epoch": 0.870151121378381,
1589
+ "grad_norm": 0.21868611872196198,
1590
+ "learning_rate": 4.376997004524541e-07,
1591
+ "loss": 0.7629,
1592
+ "step": 2260
1593
+ },
1594
+ {
1595
+ "epoch": 0.8740013475791703,
1596
+ "grad_norm": 0.21716275811195374,
1597
+ "learning_rate": 4.12540435970073e-07,
1598
+ "loss": 0.7724,
1599
+ "step": 2270
1600
+ },
1601
+ {
1602
+ "epoch": 0.8778515737799596,
1603
+ "grad_norm": 0.21693965792655945,
1604
+ "learning_rate": 3.880946978614703e-07,
1605
+ "loss": 0.7797,
1606
+ "step": 2280
1607
+ },
1608
+ {
1609
+ "epoch": 0.8817017999807488,
1610
+ "grad_norm": 0.2148953527212143,
1611
+ "learning_rate": 3.643662883789878e-07,
1612
+ "loss": 0.7865,
1613
+ "step": 2290
1614
+ },
1615
+ {
1616
+ "epoch": 0.8855520261815382,
1617
+ "grad_norm": 0.20053698122501373,
1618
+ "learning_rate": 3.41358898202781e-07,
1619
+ "loss": 0.7646,
1620
+ "step": 2300
1621
+ },
1622
+ {
1623
+ "epoch": 0.8894022523823275,
1624
+ "grad_norm": 0.20031020045280457,
1625
+ "learning_rate": 3.190761058667741e-07,
1626
+ "loss": 0.7774,
1627
+ "step": 2310
1628
+ },
1629
+ {
1630
+ "epoch": 0.8932524785831167,
1631
+ "grad_norm": 0.21005405485630035,
1632
+ "learning_rate": 2.9752137720206463e-07,
1633
+ "loss": 0.7816,
1634
+ "step": 2320
1635
+ },
1636
+ {
1637
+ "epoch": 0.8971027047839061,
1638
+ "grad_norm": 0.21162590384483337,
1639
+ "learning_rate": 2.7669806479785067e-07,
1640
+ "loss": 0.7855,
1641
+ "step": 2330
1642
+ },
1643
+ {
1644
+ "epoch": 0.9009529309846953,
1645
+ "grad_norm": 0.21444492042064667,
1646
+ "learning_rate": 2.566094074799763e-07,
1647
+ "loss": 0.7809,
1648
+ "step": 2340
1649
+ },
1650
+ {
1651
+ "epoch": 0.9048031571854847,
1652
+ "grad_norm": 0.2105267494916916,
1653
+ "learning_rate": 2.3725852980716557e-07,
1654
+ "loss": 0.763,
1655
+ "step": 2350
1656
+ },
1657
+ {
1658
+ "epoch": 0.908653383386274,
1659
+ "grad_norm": 0.21125130355358124,
1660
+ "learning_rate": 2.1864844158503983e-07,
1661
+ "loss": 0.7753,
1662
+ "step": 2360
1663
+ },
1664
+ {
1665
+ "epoch": 0.9125036095870632,
1666
+ "grad_norm": 0.23165112733840942,
1667
+ "learning_rate": 2.007820373979702e-07,
1668
+ "loss": 0.7676,
1669
+ "step": 2370
1670
+ },
1671
+ {
1672
+ "epoch": 0.9163538357878526,
1673
+ "grad_norm": 0.21406236290931702,
1674
+ "learning_rate": 1.8366209615886532e-07,
1675
+ "loss": 0.778,
1676
+ "step": 2380
1677
+ },
1678
+ {
1679
+ "epoch": 0.9202040619886418,
1680
+ "grad_norm": 0.21401266753673553,
1681
+ "learning_rate": 1.6729128067693613e-07,
1682
+ "loss": 0.7858,
1683
+ "step": 2390
1684
+ },
1685
+ {
1686
+ "epoch": 0.9240542881894311,
1687
+ "grad_norm": 0.2325669527053833,
1688
+ "learning_rate": 1.5167213724353426e-07,
1689
+ "loss": 0.7867,
1690
+ "step": 2400
1691
+ },
1692
+ {
1693
+ "epoch": 0.9279045143902205,
1694
+ "grad_norm": 0.2123924344778061,
1695
+ "learning_rate": 1.3680709523610058e-07,
1696
+ "loss": 0.7664,
1697
+ "step": 2410
1698
+ },
1699
+ {
1700
+ "epoch": 0.9317547405910097,
1701
+ "grad_norm": 0.21428336203098297,
1702
+ "learning_rate": 1.2269846674030418e-07,
1703
+ "loss": 0.7928,
1704
+ "step": 2420
1705
+ },
1706
+ {
1707
+ "epoch": 0.935604966791799,
1708
+ "grad_norm": 0.22480536997318268,
1709
+ "learning_rate": 1.0934844619042595e-07,
1710
+ "loss": 0.7859,
1711
+ "step": 2430
1712
+ },
1713
+ {
1714
+ "epoch": 0.9394551929925883,
1715
+ "grad_norm": 0.2250777781009674,
1716
+ "learning_rate": 9.675911002803961e-08,
1717
+ "loss": 0.7768,
1718
+ "step": 2440
1719
+ },
1720
+ {
1721
+ "epoch": 0.9433054191933776,
1722
+ "grad_norm": 0.23111087083816528,
1723
+ "learning_rate": 8.493241637904381e-08,
1724
+ "loss": 0.775,
1725
+ "step": 2450
1726
+ },
1727
+ {
1728
+ "epoch": 0.9471556453941669,
1729
+ "grad_norm": 0.21048076450824738,
1730
+ "learning_rate": 7.387020474909689e-08,
1731
+ "loss": 0.7922,
1732
+ "step": 2460
1733
+ },
1734
+ {
1735
+ "epoch": 0.9510058715949562,
1736
+ "grad_norm": 0.22783935070037842,
1737
+ "learning_rate": 6.357419573750856e-08,
1738
+ "loss": 0.7884,
1739
+ "step": 2470
1740
+ },
1741
+ {
1742
+ "epoch": 0.9548560977957455,
1743
+ "grad_norm": 0.21480537950992584,
1744
+ "learning_rate": 5.4045990769613365e-08,
1745
+ "loss": 0.7702,
1746
+ "step": 2480
1747
+ },
1748
+ {
1749
+ "epoch": 0.9587063239965348,
1750
+ "grad_norm": 0.21127092838287354,
1751
+ "learning_rate": 4.5287071847694434e-08,
1752
+ "loss": 0.7778,
1753
+ "step": 2490
1754
+ },
1755
+ {
1756
+ "epoch": 0.9625565501973241,
1757
+ "grad_norm": 0.21304671466350555,
1758
+ "learning_rate": 3.7298801320467817e-08,
1759
+ "loss": 0.7731,
1760
+ "step": 2500
1761
+ },
1762
+ {
1763
+ "epoch": 0.9664067763981133,
1764
+ "grad_norm": 0.2156616747379303,
1765
+ "learning_rate": 3.0082421671192576e-08,
1766
+ "loss": 0.7651,
1767
+ "step": 2510
1768
+ },
1769
+ {
1770
+ "epoch": 0.9702570025989027,
1771
+ "grad_norm": 0.2140880525112152,
1772
+ "learning_rate": 2.3639055324410355e-08,
1773
+ "loss": 0.7841,
1774
+ "step": 2520
1775
+ },
1776
+ {
1777
+ "epoch": 0.974107228799692,
1778
+ "grad_norm": 0.21568623185157776,
1779
+ "learning_rate": 1.7969704471369477e-08,
1780
+ "loss": 0.7717,
1781
+ "step": 2530
1782
+ },
1783
+ {
1784
+ "epoch": 0.9779574550004813,
1785
+ "grad_norm": 0.20705671608448029,
1786
+ "learning_rate": 1.3075250914142413e-08,
1787
+ "loss": 0.7727,
1788
+ "step": 2540
1789
+ },
1790
+ {
1791
+ "epoch": 0.9818076812012706,
1792
+ "grad_norm": 0.20660637319087982,
1793
+ "learning_rate": 8.956455928476049e-09,
1794
+ "loss": 0.7834,
1795
+ "step": 2550
1796
+ },
1797
+ {
1798
+ "epoch": 0.9856579074020598,
1799
+ "grad_norm": 0.207611545920372,
1800
+ "learning_rate": 5.6139601453786275e-09,
1801
+ "loss": 0.7764,
1802
+ "step": 2560
1803
+ },
1804
+ {
1805
+ "epoch": 0.9895081336028492,
1806
+ "grad_norm": 0.211708202958107,
1807
+ "learning_rate": 3.0482834514805602e-09,
1808
+ "loss": 0.7658,
1809
+ "step": 2570
1810
+ },
1811
+ {
1812
+ "epoch": 0.9933583598036385,
1813
+ "grad_norm": 0.20962680876255035,
1814
+ "learning_rate": 1.2598249081713366e-09,
1815
+ "loss": 0.7577,
1816
+ "step": 2580
1817
+ },
1818
+ {
1819
+ "epoch": 0.9972085860044277,
1820
+ "grad_norm": 0.22254925966262817,
1821
+ "learning_rate": 2.488626895302826e-10,
1822
+ "loss": 0.7887,
1823
+ "step": 2590
1824
+ },
1825
+ {
1826
+ "epoch": 0.9999037443449803,
1827
+ "step": 2597,
1828
+ "total_flos": 3.315169149931487e+18,
1829
+ "train_loss": 0.7969856405423429,
1830
+ "train_runtime": 11512.3311,
1831
+ "train_samples_per_second": 57.756,
1832
+ "train_steps_per_second": 0.226
1833
+ }
1834
+ ],
1835
+ "logging_steps": 10,
1836
+ "max_steps": 2597,
1837
+ "num_input_tokens_seen": 418855352,
1838
+ "num_train_epochs": 1,
1839
+ "save_steps": 500,
1840
+ "stateful_callbacks": {
1841
+ "TrainerControl": {
1842
+ "args": {
1843
+ "should_epoch_stop": false,
1844
+ "should_evaluate": false,
1845
+ "should_log": false,
1846
+ "should_save": true,
1847
+ "should_training_stop": true
1848
+ },
1849
+ "attributes": {}
1850
+ }
1851
+ },
1852
+ "total_flos": 3.315169149931487e+18,
1853
+ "train_batch_size": 8,
1854
+ "trial_name": null,
1855
+ "trial_params": null
1856
+ }
continuous/qwen25/pe_core_t16_384/finetune/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f1b555222e18ed2327a10a5a9c6978adbb81964a25d6cfcc6f0aef8fae900f2
3
+ size 6264
continuous/qwen25/pe_core_t16_384/finetune/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
continuous/qwen25/pe_core_t16_384/pretrain/config.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_faster_video": false,
3
+ "add_time_instruction": false,
4
+ "architectures": [
5
+ "Qwen2ForCausalLM"
6
+ ],
7
+ "attention_dropout": 0.0,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151645,
10
+ "faster_token_stride": 10,
11
+ "force_sample": false,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 1536,
14
+ "image_aspect_ratio": "square",
15
+ "image_crop_resolution": null,
16
+ "image_grid_pinpoints": null,
17
+ "image_split_resolution": null,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 8960,
20
+ "layer_types": [
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
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention"
49
+ ],
50
+ "max_position_embeddings": 32768,
51
+ "max_window_layers": 21,
52
+ "mm_hidden_size": 192,
53
+ "mm_newline_position": "grid",
54
+ "mm_patch_merge_type": "flat",
55
+ "mm_projector_lr": null,
56
+ "mm_projector_type": "mlp2x_gelu",
57
+ "mm_resampler_type": null,
58
+ "mm_spatial_pool_mode": "bilinear",
59
+ "mm_spatial_pool_stride": null,
60
+ "mm_tunable_parts": "mm_mlp_adapter",
61
+ "mm_use_im_patch_token": false,
62
+ "mm_use_im_start_end": false,
63
+ "mm_vision_select_feature": "patch",
64
+ "mm_vision_select_layer": -1,
65
+ "mm_vision_tower": "vtb_ssl:pe",
66
+ "mm_vision_tower_lr": null,
67
+ "model_type": "qwen2",
68
+ "num_attention_heads": 12,
69
+ "num_hidden_layers": 28,
70
+ "num_key_value_heads": 2,
71
+ "pad_token_id": 151643,
72
+ "pos_skipping_range": 4096,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_scaling": null,
75
+ "rope_theta": 1000000.0,
76
+ "sliding_window": null,
77
+ "tie_word_embeddings": true,
78
+ "tokenizer_model_max_length": 8192,
79
+ "tokenizer_padding_side": "right",
80
+ "transformers_version": "4.57.6",
81
+ "use_cache": true,
82
+ "use_mm_proj": true,
83
+ "use_pos_skipping": false,
84
+ "use_sliding_window": false,
85
+ "vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/pe/PE-Core-T16-384/PE-Core-T16-384.pt",
86
+ "vocab_size": 151936
87
+ }
continuous/qwen25/pe_core_t16_384/pretrain/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e3394aec551d8f3d2960e7238099fe45fd250bf20a9823bb9bec208d0467cbe
3
+ size 10631224
continuous/qwen25/pe_core_t16_384/pretrain/trainer_state.json ADDED
@@ -0,0 +1,421 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 545,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.01834862385321101,
14
+ "grad_norm": 0.3195981979370117,
15
+ "learning_rate": 0.00032142857142857147,
16
+ "loss": 4.658,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.03669724770642202,
21
+ "grad_norm": 0.18599539995193481,
22
+ "learning_rate": 0.0006785714285714287,
23
+ "loss": 3.773,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.05504587155963303,
28
+ "grad_norm": 0.06548872590065002,
29
+ "learning_rate": 0.0009999907688176173,
30
+ "loss": 3.422,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.07339449541284404,
35
+ "grad_norm": 0.04160045459866524,
36
+ "learning_rate": 0.0009988834393115766,
37
+ "loss": 3.3227,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.09174311926605505,
42
+ "grad_norm": 0.04526331275701523,
43
+ "learning_rate": 0.0009959345572940772,
44
+ "loss": 3.2847,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.11009174311926606,
49
+ "grad_norm": 0.049120448529720306,
50
+ "learning_rate": 0.0009911550081155983,
51
+ "loss": 3.2501,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.12844036697247707,
56
+ "grad_norm": 0.06738036125898361,
57
+ "learning_rate": 0.0009845624347566433,
58
+ "loss": 3.2254,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.14678899082568808,
63
+ "grad_norm": 0.0484769269824028,
64
+ "learning_rate": 0.0009761811727013547,
65
+ "loss": 3.2027,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.1651376146788991,
70
+ "grad_norm": 0.03982609137892723,
71
+ "learning_rate": 0.0009660421601067666,
72
+ "loss": 3.1796,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.1834862385321101,
77
+ "grad_norm": 0.06149926409125328,
78
+ "learning_rate": 0.0009541828235992926,
79
+ "loss": 3.1682,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.2018348623853211,
84
+ "grad_norm": 0.06128969416022301,
85
+ "learning_rate": 0.0009406469401200151,
86
+ "loss": 3.1436,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.22018348623853212,
91
+ "grad_norm": 0.06992045044898987,
92
+ "learning_rate": 0.0009254844753287492,
93
+ "loss": 3.1345,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.23853211009174313,
98
+ "grad_norm": 0.055654868483543396,
99
+ "learning_rate": 0.0009087513991633924,
100
+ "loss": 3.1017,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.25688073394495414,
105
+ "grad_norm": 0.07554488629102707,
106
+ "learning_rate": 0.0008905094792353917,
107
+ "loss": 3.0896,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.27522935779816515,
112
+ "grad_norm": 0.09819451719522476,
113
+ "learning_rate": 0.0008708260528239789,
114
+ "loss": 3.083,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.29357798165137616,
119
+ "grad_norm": 0.04817397519946098,
120
+ "learning_rate": 0.0008497737783108237,
121
+ "loss": 3.065,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.3119266055045872,
126
+ "grad_norm": 0.07595115154981613,
127
+ "learning_rate": 0.0008274303669726426,
128
+ "loss": 3.0573,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.3302752293577982,
133
+ "grad_norm": 0.06587675213813782,
134
+ "learning_rate": 0.0008038782961218137,
135
+ "loss": 3.0373,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.3486238532110092,
140
+ "grad_norm": 0.06800957769155502,
141
+ "learning_rate": 0.000779204504653894,
142
+ "loss": 3.0333,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.3669724770642202,
147
+ "grad_norm": 0.06334719061851501,
148
+ "learning_rate": 0.0007535000721258791,
149
+ "loss": 3.0074,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.3853211009174312,
154
+ "grad_norm": 0.07801689952611923,
155
+ "learning_rate": 0.00072685988254984,
156
+ "loss": 3.0,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.4036697247706422,
161
+ "grad_norm": 0.0681065246462822,
162
+ "learning_rate": 0.0006993822741429906,
163
+ "loss": 3.0033,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.42201834862385323,
168
+ "grad_norm": 0.05781600996851921,
169
+ "learning_rate": 0.0006711686763270818,
170
+ "loss": 3.0007,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.44036697247706424,
175
+ "grad_norm": 0.06390953063964844,
176
+ "learning_rate": 0.0006423232353170831,
177
+ "loss": 2.9833,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.45871559633027525,
182
+ "grad_norm": 0.06395537406206131,
183
+ "learning_rate": 0.0006129524296812335,
184
+ "loss": 2.9672,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.47706422018348627,
189
+ "grad_norm": 0.06994635611772537,
190
+ "learning_rate": 0.0005831646772915651,
191
+ "loss": 2.9756,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.4954128440366973,
196
+ "grad_norm": 0.0683479830622673,
197
+ "learning_rate": 0.0005530699351157809,
198
+ "loss": 2.9532,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.5137614678899083,
203
+ "grad_norm": 0.05616157874464989,
204
+ "learning_rate": 0.0005227792933277943,
205
+ "loss": 2.9563,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.5321100917431193,
210
+ "grad_norm": 0.06159662455320358,
211
+ "learning_rate": 0.0004924045652352092,
212
+ "loss": 2.9375,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.5504587155963303,
217
+ "grad_norm": 0.059963032603263855,
218
+ "learning_rate": 0.0004620578745374633,
219
+ "loss": 2.9358,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.5688073394495413,
224
+ "grad_norm": 0.07279516756534576,
225
+ "learning_rate": 0.0004318512414382059,
226
+ "loss": 2.9412,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.5871559633027523,
231
+ "grad_norm": 0.06351424008607864,
232
+ "learning_rate": 0.0004018961691397148,
233
+ "loss": 2.9369,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.6055045871559633,
238
+ "grad_norm": 0.08951804786920547,
239
+ "learning_rate": 0.0003723032322457458,
240
+ "loss": 2.919,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.6238532110091743,
245
+ "grad_norm": 0.0642656683921814,
246
+ "learning_rate": 0.0003431816685921625,
247
+ "loss": 2.9241,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.6422018348623854,
252
+ "grad_norm": 0.06514798104763031,
253
+ "learning_rate": 0.0003146389760120416,
254
+ "loss": 2.9277,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.6605504587155964,
259
+ "grad_norm": 0.07042760401964188,
260
+ "learning_rate": 0.00028678051552373485,
261
+ "loss": 2.9268,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.6788990825688074,
266
+ "grad_norm": 0.06049153208732605,
267
+ "learning_rate": 0.0002597091224066581,
268
+ "loss": 2.9163,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.6972477064220184,
273
+ "grad_norm": 0.05817561596632004,
274
+ "learning_rate": 0.0002335247266004629,
275
+ "loss": 2.9256,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.7155963302752294,
280
+ "grad_norm": 0.05961354076862335,
281
+ "learning_rate": 0.000208323983828828,
282
+ "loss": 2.912,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.7339449541284404,
287
+ "grad_norm": 0.06080279126763344,
288
+ "learning_rate": 0.0001841999188095224,
289
+ "loss": 2.9065,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.7522935779816514,
294
+ "grad_norm": 0.05501021444797516,
295
+ "learning_rate": 0.00016124158186777676,
296
+ "loss": 2.9085,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.7706422018348624,
301
+ "grad_norm": 0.06560225039720535,
302
+ "learning_rate": 0.0001395337202205211,
303
+ "loss": 2.9026,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.7889908256880734,
308
+ "grad_norm": 0.055996984243392944,
309
+ "learning_rate": 0.00011915646514489292,
310
+ "loss": 2.9052,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.8073394495412844,
315
+ "grad_norm": 0.05786696448922157,
316
+ "learning_rate": 0.00010018503618578818,
317
+ "loss": 2.9107,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.8256880733944955,
322
+ "grad_norm": 0.05415608361363411,
323
+ "learning_rate": 8.268946349432583e-05,
324
+ "loss": 2.9078,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.8440366972477065,
329
+ "grad_norm": 0.052878640592098236,
330
+ "learning_rate": 6.673432932217117e-05,
331
+ "loss": 2.8949,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.8623853211009175,
336
+ "grad_norm": 0.05728932097554207,
337
+ "learning_rate": 5.2378529625954684e-05,
338
+ "loss": 2.8866,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.8807339449541285,
343
+ "grad_norm": 0.051333554089069366,
344
+ "learning_rate": 3.9675056661785556e-05,
345
+ "loss": 2.8944,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.8990825688073395,
350
+ "grad_norm": 0.05312556028366089,
351
+ "learning_rate": 2.8670803372379538e-05,
352
+ "loss": 2.9006,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.9174311926605505,
357
+ "grad_norm": 0.048136208206415176,
358
+ "learning_rate": 1.9406390288876586e-05,
359
+ "loss": 2.9087,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.9357798165137615,
364
+ "grad_norm": 0.05489068850874901,
365
+ "learning_rate": 1.191601558631461e-05,
366
+ "loss": 2.8945,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.9541284403669725,
371
+ "grad_norm": 0.04619835689663887,
372
+ "learning_rate": 6.2273288462559305e-06,
373
+ "loss": 2.8918,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.9724770642201835,
378
+ "grad_norm": 0.053532302379608154,
379
+ "learning_rate": 2.361328992552314e-06,
380
+ "loss": 2.8947,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.9908256880733946,
385
+ "grad_norm": 0.05121740326285362,
386
+ "learning_rate": 3.322867770044202e-07,
387
+ "loss": 2.8943,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 1.0,
392
+ "step": 545,
393
+ "total_flos": 2.5245574375460045e+17,
394
+ "train_loss": 3.0501117566309937,
395
+ "train_runtime": 5315.2725,
396
+ "train_samples_per_second": 105.005,
397
+ "train_steps_per_second": 0.103
398
+ }
399
+ ],
400
+ "logging_steps": 10,
401
+ "max_steps": 545,
402
+ "num_input_tokens_seen": 31896544,
403
+ "num_train_epochs": 1,
404
+ "save_steps": 500,
405
+ "stateful_callbacks": {
406
+ "TrainerControl": {
407
+ "args": {
408
+ "should_epoch_stop": false,
409
+ "should_evaluate": false,
410
+ "should_log": false,
411
+ "should_save": false,
412
+ "should_training_stop": false
413
+ },
414
+ "attributes": {}
415
+ }
416
+ },
417
+ "total_flos": 2.5245574375460045e+17,
418
+ "train_batch_size": 32,
419
+ "trial_name": null,
420
+ "trial_params": null
421
+ }
continuous/qwen25/pe_lang_g14_448/finetune/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
+ }
continuous/qwen25/pe_lang_g14_448/finetune/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
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
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\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|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 %}
continuous/qwen25/pe_lang_g14_448/finetune/config.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_faster_video": false,
3
+ "add_time_instruction": false,
4
+ "architectures": [
5
+ "LlavaQwenForCausalLM"
6
+ ],
7
+ "attention_dropout": 0.0,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151645,
10
+ "faster_token_stride": 10,
11
+ "force_sample": false,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 1536,
14
+ "image_aspect_ratio": "square",
15
+ "image_crop_resolution": null,
16
+ "image_grid_pinpoints": null,
17
+ "image_split_resolution": null,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 8960,
20
+ "layer_types": [
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
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention"
49
+ ],
50
+ "max_position_embeddings": 32768,
51
+ "max_window_layers": 21,
52
+ "mm_hidden_size": 1536,
53
+ "mm_newline_position": "grid",
54
+ "mm_patch_merge_type": "flat",
55
+ "mm_projector_lr": 0.0001,
56
+ "mm_projector_type": "mlp2x_gelu",
57
+ "mm_resampler_type": null,
58
+ "mm_spatial_pool_mode": "bilinear",
59
+ "mm_spatial_pool_stride": null,
60
+ "mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
61
+ "mm_use_im_patch_token": false,
62
+ "mm_use_im_start_end": false,
63
+ "mm_vision_select_feature": "patch",
64
+ "mm_vision_select_layer": -1,
65
+ "mm_vision_tower": "vtb_ssl:pe",
66
+ "mm_vision_tower_lr": null,
67
+ "model_type": "qwen2",
68
+ "num_attention_heads": 12,
69
+ "num_hidden_layers": 28,
70
+ "num_key_value_heads": 2,
71
+ "pad_token_id": 151643,
72
+ "pos_skipping_range": 4096,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_scaling": null,
75
+ "rope_theta": 1000000.0,
76
+ "sliding_window": null,
77
+ "tie_word_embeddings": true,
78
+ "tokenizer_model_max_length": 8192,
79
+ "tokenizer_padding_side": "right",
80
+ "transformers_version": "4.57.6",
81
+ "use_cache": true,
82
+ "use_mm_proj": true,
83
+ "use_pos_skipping": false,
84
+ "use_sliding_window": false,
85
+ "vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/pe/PE-Lang-G14-448/PE-Lang-G14-448.pt",
86
+ "vocab_size": 151936
87
+ }
continuous/qwen25/pe_lang_g14_448/finetune/generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_sample": true,
3
+ "eos_token_id": [
4
+ 151645,
5
+ 151643
6
+ ],
7
+ "pad_token_id": 151643,
8
+ "repetition_penalty": 1.1,
9
+ "temperature": 0.7,
10
+ "top_k": 20,
11
+ "top_p": 0.8,
12
+ "transformers_version": "4.57.6"
13
+ }
continuous/qwen25/pe_lang_g14_448/finetune/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
continuous/qwen25/pe_lang_g14_448/finetune/model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6540e222eb2c15752d2612b8e0708c55fea6543527c211c278ffeb67851b787d
3
+ size 4988334784
continuous/qwen25/pe_lang_g14_448/finetune/model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6bf7b7c56ca3acb290a7943fe5c5b29cffdca8924a5540ed6640cf121f825908
3
+ size 2066708576
continuous/qwen25/pe_lang_g14_448/finetune/model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
continuous/qwen25/pe_lang_g14_448/finetune/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
+ }
continuous/qwen25/pe_lang_g14_448/finetune/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
3
+ size 11421896
continuous/qwen25/pe_lang_g14_448/finetune/tokenizer_config.json ADDED
@@ -0,0 +1,208 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ }
181
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "clean_up_tokenization_spaces": false,
199
+ "eos_token": "<|im_end|>",
200
+ "errors": "replace",
201
+ "extra_special_tokens": {},
202
+ "model_max_length": 8192,
203
+ "pad_token": "<|endoftext|>",
204
+ "padding_side": "right",
205
+ "split_special_tokens": false,
206
+ "tokenizer_class": "Qwen2Tokenizer",
207
+ "unk_token": null
208
+ }
continuous/qwen25/pe_lang_g14_448/finetune/trainer_state.json ADDED
@@ -0,0 +1,1856 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.9999037443449803,
6
+ "eval_steps": 500,
7
+ "global_step": 2597,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.003850226200789296,
14
+ "grad_norm": 12.324213981628418,
15
+ "learning_rate": 1.153846153846154e-06,
16
+ "loss": 2.2695,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.007700452401578592,
21
+ "grad_norm": 7.025111198425293,
22
+ "learning_rate": 2.435897435897436e-06,
23
+ "loss": 1.6699,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.01155067860236789,
28
+ "grad_norm": 1.996394157409668,
29
+ "learning_rate": 3.7179487179487184e-06,
30
+ "loss": 1.0835,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.015400904803157185,
35
+ "grad_norm": 0.39286160469055176,
36
+ "learning_rate": 5e-06,
37
+ "loss": 0.8885,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.019251131003946483,
42
+ "grad_norm": 0.23678062856197357,
43
+ "learning_rate": 6.282051282051282e-06,
44
+ "loss": 0.8571,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.02310135720473578,
49
+ "grad_norm": 0.17502111196517944,
50
+ "learning_rate": 7.564102564102564e-06,
51
+ "loss": 0.8092,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.026951583405525073,
56
+ "grad_norm": 0.13215219974517822,
57
+ "learning_rate": 8.846153846153847e-06,
58
+ "loss": 0.783,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.03080180960631437,
63
+ "grad_norm": 0.13992276787757874,
64
+ "learning_rate": 9.999996111488723e-06,
65
+ "loss": 0.7924,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.034652035807103666,
70
+ "grad_norm": 0.1533016860485077,
71
+ "learning_rate": 9.999529497453782e-06,
72
+ "loss": 0.7847,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.038502262007892966,
77
+ "grad_norm": 0.12892073392868042,
78
+ "learning_rate": 9.998285264324193e-06,
79
+ "loss": 0.7497,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.04235248820868226,
84
+ "grad_norm": 0.12179314345121384,
85
+ "learning_rate": 9.996263605626052e-06,
86
+ "loss": 0.7455,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.04620271440947156,
91
+ "grad_norm": 0.12279170006513596,
92
+ "learning_rate": 9.99346483580503e-06,
93
+ "loss": 0.7575,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.05005294061026085,
98
+ "grad_norm": 0.11094465106725693,
99
+ "learning_rate": 9.989889390177467e-06,
100
+ "loss": 0.7398,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.053903166811050146,
105
+ "grad_norm": 0.1266247183084488,
106
+ "learning_rate": 9.985537824862647e-06,
107
+ "loss": 0.7479,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.057753393011839446,
112
+ "grad_norm": 0.13799382746219635,
113
+ "learning_rate": 9.980410816696317e-06,
114
+ "loss": 0.7355,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.06160361921262874,
119
+ "grad_norm": 0.12321662902832031,
120
+ "learning_rate": 9.97450916312541e-06,
121
+ "loss": 0.7379,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.06545384541341803,
126
+ "grad_norm": 0.10648263990879059,
127
+ "learning_rate": 9.967833782084004e-06,
128
+ "loss": 0.7389,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.06930407161420733,
133
+ "grad_norm": 0.132377028465271,
134
+ "learning_rate": 9.960385711850554e-06,
135
+ "loss": 0.758,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.07315429781499663,
140
+ "grad_norm": 0.1284930258989334,
141
+ "learning_rate": 9.952166110886394e-06,
142
+ "loss": 0.7594,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.07700452401578593,
147
+ "grad_norm": 0.1353977471590042,
148
+ "learning_rate": 9.943176257655567e-06,
149
+ "loss": 0.7403,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.08085475021657522,
154
+ "grad_norm": 0.11774808168411255,
155
+ "learning_rate": 9.933417550425952e-06,
156
+ "loss": 0.7493,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.08470497641736452,
161
+ "grad_norm": 0.12515190243721008,
162
+ "learning_rate": 9.9228915070518e-06,
163
+ "loss": 0.7236,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.08855520261815382,
168
+ "grad_norm": 0.13562101125717163,
169
+ "learning_rate": 9.911599764737632e-06,
170
+ "loss": 0.7298,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.09240542881894312,
175
+ "grad_norm": 0.13133682310581207,
176
+ "learning_rate": 9.899544079783606e-06,
177
+ "loss": 0.7159,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.0962556550197324,
182
+ "grad_norm": 0.1176205649971962,
183
+ "learning_rate": 9.886726327312338e-06,
184
+ "loss": 0.7344,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.1001058812205217,
189
+ "grad_norm": 0.12539319694042206,
190
+ "learning_rate": 9.87314850097724e-06,
191
+ "loss": 0.7379,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.103956107421311,
196
+ "grad_norm": 0.11378435045480728,
197
+ "learning_rate": 9.858812712652452e-06,
198
+ "loss": 0.7227,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.10780633362210029,
203
+ "grad_norm": 0.11367890238761902,
204
+ "learning_rate": 9.843721192104341e-06,
205
+ "loss": 0.7241,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.11165655982288959,
210
+ "grad_norm": 0.1168658584356308,
211
+ "learning_rate": 9.827876286644698e-06,
212
+ "loss": 0.7423,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.11550678602367889,
217
+ "grad_norm": 0.11890362203121185,
218
+ "learning_rate": 9.811280460765643e-06,
219
+ "loss": 0.7062,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.11935701222446819,
224
+ "grad_norm": 0.11882569640874863,
225
+ "learning_rate": 9.793936295756292e-06,
226
+ "loss": 0.7153,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.12320723842525748,
231
+ "grad_norm": 0.12301485240459442,
232
+ "learning_rate": 9.775846489301279e-06,
233
+ "loss": 0.722,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.12705746462604678,
238
+ "grad_norm": 0.12254803627729416,
239
+ "learning_rate": 9.757013855061157e-06,
240
+ "loss": 0.7199,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.13090769082683607,
245
+ "grad_norm": 0.12338763475418091,
246
+ "learning_rate": 9.737441322234758e-06,
247
+ "loss": 0.7284,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.13475791702762538,
252
+ "grad_norm": 0.12786762416362762,
253
+ "learning_rate": 9.717131935103605e-06,
254
+ "loss": 0.7347,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.13860814322841467,
259
+ "grad_norm": 0.11916555464267731,
260
+ "learning_rate": 9.696088852558398e-06,
261
+ "loss": 0.7196,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.14245836942920398,
266
+ "grad_norm": 0.11739562451839447,
267
+ "learning_rate": 9.674315347607686e-06,
268
+ "loss": 0.7438,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.14630859562999327,
273
+ "grad_norm": 0.12059490382671356,
274
+ "learning_rate": 9.651814806868796e-06,
275
+ "loss": 0.7181,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.15015882183078255,
280
+ "grad_norm": 0.11872441321611404,
281
+ "learning_rate": 9.62859073004107e-06,
282
+ "loss": 0.7155,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.15400904803157187,
287
+ "grad_norm": 0.10904064774513245,
288
+ "learning_rate": 9.604646729361547e-06,
289
+ "loss": 0.7181,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.15785927423236115,
294
+ "grad_norm": 0.11471547186374664,
295
+ "learning_rate": 9.579986529043099e-06,
296
+ "loss": 0.6998,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.16170950043315044,
301
+ "grad_norm": 0.12376510351896286,
302
+ "learning_rate": 9.554613964695189e-06,
303
+ "loss": 0.7194,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.16555972663393975,
308
+ "grad_norm": 0.11777686327695847,
309
+ "learning_rate": 9.528532982727281e-06,
310
+ "loss": 0.7152,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.16940995283472904,
315
+ "grad_norm": 0.12025685608386993,
316
+ "learning_rate": 9.501747639735018e-06,
317
+ "loss": 0.7106,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.17326017903551832,
322
+ "grad_norm": 0.11889459937810898,
323
+ "learning_rate": 9.474262101869272e-06,
324
+ "loss": 0.698,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.17711040523630764,
329
+ "grad_norm": 0.12180490046739578,
330
+ "learning_rate": 9.446080644188141e-06,
331
+ "loss": 0.7139,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.18096063143709692,
336
+ "grad_norm": 0.12096218019723892,
337
+ "learning_rate": 9.417207649992017e-06,
338
+ "loss": 0.7142,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.18481085763788624,
343
+ "grad_norm": 0.1234583854675293,
344
+ "learning_rate": 9.387647610141816e-06,
345
+ "loss": 0.706,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.18866108383867553,
350
+ "grad_norm": 0.11031262576580048,
351
+ "learning_rate": 9.35740512236047e-06,
352
+ "loss": 0.701,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.1925113100394648,
357
+ "grad_norm": 0.11529544740915298,
358
+ "learning_rate": 9.326484890517809e-06,
359
+ "loss": 0.71,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.19636153624025413,
364
+ "grad_norm": 0.12345736473798752,
365
+ "learning_rate": 9.294891723898926e-06,
366
+ "loss": 0.6896,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.2002117624410434,
371
+ "grad_norm": 0.12189178913831711,
372
+ "learning_rate": 9.262630536456153e-06,
373
+ "loss": 0.7281,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.2040619886418327,
378
+ "grad_norm": 0.12327122688293457,
379
+ "learning_rate": 9.229706346044749e-06,
380
+ "loss": 0.7011,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.207912214842622,
385
+ "grad_norm": 0.12792851030826569,
386
+ "learning_rate": 9.196124273642422e-06,
387
+ "loss": 0.7165,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.2117624410434113,
392
+ "grad_norm": 0.11543191969394684,
393
+ "learning_rate": 9.161889542552842e-06,
394
+ "loss": 0.7267,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.21561266724420058,
399
+ "grad_norm": 0.1163366436958313,
400
+ "learning_rate": 9.127007477593193e-06,
401
+ "loss": 0.7195,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.2194628934449899,
406
+ "grad_norm": 0.12813590466976166,
407
+ "learning_rate": 9.091483504265974e-06,
408
+ "loss": 0.7282,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.22331311964577918,
413
+ "grad_norm": 0.12032407522201538,
414
+ "learning_rate": 9.055323147915116e-06,
415
+ "loss": 0.7044,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.2271633458465685,
420
+ "grad_norm": 0.12227418273687363,
421
+ "learning_rate": 9.018532032866585e-06,
422
+ "loss": 0.7195,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.23101357204735778,
427
+ "grad_norm": 0.13292452692985535,
428
+ "learning_rate": 8.981115881553579e-06,
429
+ "loss": 0.7169,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.23486379824814707,
434
+ "grad_norm": 0.12193329632282257,
435
+ "learning_rate": 8.943080513626476e-06,
436
+ "loss": 0.7288,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.23871402444893638,
441
+ "grad_norm": 0.12439188361167908,
442
+ "learning_rate": 8.904431845047641e-06,
443
+ "loss": 0.72,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.24256425064972567,
448
+ "grad_norm": 0.12147743999958038,
449
+ "learning_rate": 8.86517588717128e-06,
450
+ "loss": 0.725,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.24641447685051496,
455
+ "grad_norm": 0.1106458529829979,
456
+ "learning_rate": 8.82531874580844e-06,
457
+ "loss": 0.7276,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.25026470305130427,
462
+ "grad_norm": 0.11053787171840668,
463
+ "learning_rate": 8.784866620277314e-06,
464
+ "loss": 0.6836,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.25411492925209356,
469
+ "grad_norm": 0.1312021166086197,
470
+ "learning_rate": 8.743825802439018e-06,
471
+ "loss": 0.6982,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.25796515545288284,
476
+ "grad_norm": 0.12452798336744308,
477
+ "learning_rate": 8.702202675718956e-06,
478
+ "loss": 0.7239,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.26181538165367213,
483
+ "grad_norm": 0.11540570855140686,
484
+ "learning_rate": 8.660003714113959e-06,
485
+ "loss": 0.6958,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.26566560785446147,
490
+ "grad_norm": 0.11419165134429932,
491
+ "learning_rate": 8.617235481185323e-06,
492
+ "loss": 0.7065,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.26951583405525076,
497
+ "grad_norm": 0.11562362313270569,
498
+ "learning_rate": 8.573904629037931e-06,
499
+ "loss": 0.7138,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.27336606025604004,
504
+ "grad_norm": 0.11472730338573456,
505
+ "learning_rate": 8.530017897285585e-06,
506
+ "loss": 0.6991,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.27721628645682933,
511
+ "grad_norm": 0.11573010683059692,
512
+ "learning_rate": 8.485582112002746e-06,
513
+ "loss": 0.7103,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.2810665126576186,
518
+ "grad_norm": 0.11485695093870163,
519
+ "learning_rate": 8.440604184662811e-06,
520
+ "loss": 0.7102,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.28491673885840796,
525
+ "grad_norm": 0.12254331260919571,
526
+ "learning_rate": 8.395091111063118e-06,
527
+ "loss": 0.6974,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.28876696505919724,
532
+ "grad_norm": 0.10609365999698639,
533
+ "learning_rate": 8.349049970236822e-06,
534
+ "loss": 0.699,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.29261719125998653,
539
+ "grad_norm": 0.11689411103725433,
540
+ "learning_rate": 8.302487923351842e-06,
541
+ "loss": 0.687,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.2964674174607758,
546
+ "grad_norm": 0.11658143997192383,
547
+ "learning_rate": 8.255412212597021e-06,
548
+ "loss": 0.7077,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.3003176436615651,
553
+ "grad_norm": 0.12256301939487457,
554
+ "learning_rate": 8.207830160055688e-06,
555
+ "loss": 0.7033,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.3041678698623544,
560
+ "grad_norm": 0.11867094784975052,
561
+ "learning_rate": 8.159749166566786e-06,
562
+ "loss": 0.6906,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.30801809606314373,
567
+ "grad_norm": 0.12085304409265518,
568
+ "learning_rate": 8.11117671057377e-06,
569
+ "loss": 0.7144,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.311868322263933,
574
+ "grad_norm": 0.11776888370513916,
575
+ "learning_rate": 8.062120346961404e-06,
576
+ "loss": 0.7001,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.3157185484647223,
581
+ "grad_norm": 0.10413401573896408,
582
+ "learning_rate": 8.012587705880705e-06,
583
+ "loss": 0.7074,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.3195687746655116,
588
+ "grad_norm": 0.1255320906639099,
589
+ "learning_rate": 7.962586491562146e-06,
590
+ "loss": 0.694,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.3234190008663009,
595
+ "grad_norm": 0.12615175545215607,
596
+ "learning_rate": 7.912124481117357e-06,
597
+ "loss": 0.696,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.3272692270670902,
602
+ "grad_norm": 0.11574489623308182,
603
+ "learning_rate": 7.861209523329494e-06,
604
+ "loss": 0.7043,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.3311194532678795,
609
+ "grad_norm": 0.11616148054599762,
610
+ "learning_rate": 7.809849537432432e-06,
611
+ "loss": 0.6933,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.3349696794686688,
616
+ "grad_norm": 0.10648791491985321,
617
+ "learning_rate": 7.758052511879035e-06,
618
+ "loss": 0.6911,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.3388199056694581,
623
+ "grad_norm": 0.11016496270895004,
624
+ "learning_rate": 7.705826503098642e-06,
625
+ "loss": 0.6981,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.34267013187024736,
630
+ "grad_norm": 0.11787716299295425,
631
+ "learning_rate": 7.653179634243979e-06,
632
+ "loss": 0.6924,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.34652035807103665,
637
+ "grad_norm": 0.11555282026529312,
638
+ "learning_rate": 7.600120093927696e-06,
639
+ "loss": 0.6859,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.350370584271826,
644
+ "grad_norm": 0.11861473321914673,
645
+ "learning_rate": 7.546656134948729e-06,
646
+ "loss": 0.6836,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.3542208104726153,
651
+ "grad_norm": 0.11019252985715866,
652
+ "learning_rate": 7.492796073008661e-06,
653
+ "loss": 0.6813,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.35807103667340456,
658
+ "grad_norm": 0.12635816633701324,
659
+ "learning_rate": 7.438548285418326e-06,
660
+ "loss": 0.677,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.36192126287419385,
665
+ "grad_norm": 0.12196777760982513,
666
+ "learning_rate": 7.3839212097948e-06,
667
+ "loss": 0.7132,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.36577148907498314,
672
+ "grad_norm": 0.11073444783687592,
673
+ "learning_rate": 7.328923342749039e-06,
674
+ "loss": 0.7257,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.3696217152757725,
679
+ "grad_norm": 0.12053274363279343,
680
+ "learning_rate": 7.273563238564319e-06,
681
+ "loss": 0.704,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.37347194147656176,
686
+ "grad_norm": 0.11276710033416748,
687
+ "learning_rate": 7.217849507865724e-06,
688
+ "loss": 0.6798,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.37732216767735105,
693
+ "grad_norm": 0.11120860278606415,
694
+ "learning_rate": 7.161790816280858e-06,
695
+ "loss": 0.6979,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.38117239387814034,
700
+ "grad_norm": 0.12413940578699112,
701
+ "learning_rate": 7.1053958830920125e-06,
702
+ "loss": 0.6694,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.3850226200789296,
707
+ "grad_norm": 0.11231735348701477,
708
+ "learning_rate": 7.0486734798799715e-06,
709
+ "loss": 0.7002,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.3888728462797189,
714
+ "grad_norm": 0.11600877344608307,
715
+ "learning_rate": 6.991632429159708e-06,
716
+ "loss": 0.722,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.39272307248050825,
721
+ "grad_norm": 0.12726068496704102,
722
+ "learning_rate": 6.934281603008132e-06,
723
+ "loss": 0.6975,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.39657329868129754,
728
+ "grad_norm": 0.11292813718318939,
729
+ "learning_rate": 6.876629921684143e-06,
730
+ "loss": 0.7092,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.4004235248820868,
735
+ "grad_norm": 0.11842852830886841,
736
+ "learning_rate": 6.818686352241199e-06,
737
+ "loss": 0.7051,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.4042737510828761,
742
+ "grad_norm": 0.10833907127380371,
743
+ "learning_rate": 6.760459907132573e-06,
744
+ "loss": 0.6937,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.4081239772836654,
749
+ "grad_norm": 0.11265300959348679,
750
+ "learning_rate": 6.701959642809589e-06,
751
+ "loss": 0.6977,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.41197420348445474,
756
+ "grad_norm": 0.12088123708963394,
757
+ "learning_rate": 6.643194658312987e-06,
758
+ "loss": 0.6793,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.415824429685244,
763
+ "grad_norm": 0.1096087247133255,
764
+ "learning_rate": 6.584174093857676e-06,
765
+ "loss": 0.7181,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.4196746558860333,
770
+ "grad_norm": 0.12818241119384766,
771
+ "learning_rate": 6.5249071294110735e-06,
772
+ "loss": 0.6909,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.4235248820868226,
777
+ "grad_norm": 0.11633343249559402,
778
+ "learning_rate": 6.465402983265274e-06,
779
+ "loss": 0.7103,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.4273751082876119,
784
+ "grad_norm": 0.11809894442558289,
785
+ "learning_rate": 6.405670910603243e-06,
786
+ "loss": 0.7068,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.43122533448840117,
791
+ "grad_norm": 0.12198018282651901,
792
+ "learning_rate": 6.345720202059291e-06,
793
+ "loss": 0.6972,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.4350755606891905,
798
+ "grad_norm": 0.11708134412765503,
799
+ "learning_rate": 6.285560182274008e-06,
800
+ "loss": 0.6862,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.4389257868899798,
805
+ "grad_norm": 0.10934826731681824,
806
+ "learning_rate": 6.225200208443936e-06,
807
+ "loss": 0.6933,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.4427760130907691,
812
+ "grad_norm": 0.10771723091602325,
813
+ "learning_rate": 6.16464966886616e-06,
814
+ "loss": 0.6877,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.44662623929155837,
819
+ "grad_norm": 0.13413622975349426,
820
+ "learning_rate": 6.103917981478065e-06,
821
+ "loss": 0.7172,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.45047646549234766,
826
+ "grad_norm": 0.11088903248310089,
827
+ "learning_rate": 6.043014592392489e-06,
828
+ "loss": 0.7237,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.454326691693137,
833
+ "grad_norm": 0.10944590717554092,
834
+ "learning_rate": 5.981948974428485e-06,
835
+ "loss": 0.696,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.4581769178939263,
840
+ "grad_norm": 0.11830946058034897,
841
+ "learning_rate": 5.920730625637934e-06,
842
+ "loss": 0.676,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.46202714409471557,
847
+ "grad_norm": 0.12409171462059021,
848
+ "learning_rate": 5.8593690678282355e-06,
849
+ "loss": 0.6894,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.46587737029550486,
854
+ "grad_norm": 0.11819177865982056,
855
+ "learning_rate": 5.797873845081301e-06,
856
+ "loss": 0.69,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.46972759649629414,
861
+ "grad_norm": 0.11742081493139267,
862
+ "learning_rate": 5.736254522269075e-06,
863
+ "loss": 0.7028,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.47357782269708343,
868
+ "grad_norm": 0.11251910030841827,
869
+ "learning_rate": 5.6745206835658386e-06,
870
+ "loss": 0.7119,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.47742804889787277,
875
+ "grad_norm": 0.10965047031641006,
876
+ "learning_rate": 5.612681930957495e-06,
877
+ "loss": 0.6871,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.48127827509866206,
882
+ "grad_norm": 0.1076359748840332,
883
+ "learning_rate": 5.550747882748095e-06,
884
+ "loss": 0.6814,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.48512850129945134,
889
+ "grad_norm": 0.11864548921585083,
890
+ "learning_rate": 5.4887281720638195e-06,
891
+ "loss": 0.69,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.48897872750024063,
896
+ "grad_norm": 0.1313992589712143,
897
+ "learning_rate": 5.42663244535466e-06,
898
+ "loss": 0.6736,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.4928289537010299,
903
+ "grad_norm": 0.11974918842315674,
904
+ "learning_rate": 5.3644703608940205e-06,
905
+ "loss": 0.6819,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.49667917990181926,
910
+ "grad_norm": 0.1185317188501358,
911
+ "learning_rate": 5.302251587276491e-06,
912
+ "loss": 0.701,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.5005294061026085,
917
+ "grad_norm": 0.11500269174575806,
918
+ "learning_rate": 5.2399858019140005e-06,
919
+ "loss": 0.6956,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.5043796323033978,
924
+ "grad_norm": 0.115505650639534,
925
+ "learning_rate": 5.177682689530615e-06,
926
+ "loss": 0.7124,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.5082298585041871,
931
+ "grad_norm": 0.12790171802043915,
932
+ "learning_rate": 5.11535194065619e-06,
933
+ "loss": 0.6836,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.5120800847049765,
938
+ "grad_norm": 0.12581998109817505,
939
+ "learning_rate": 5.053003250119113e-06,
940
+ "loss": 0.6834,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.5159303109057657,
945
+ "grad_norm": 0.11929438263177872,
946
+ "learning_rate": 4.9906463155383925e-06,
947
+ "loss": 0.6947,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.519780537106555,
952
+ "grad_norm": 0.11798333376646042,
953
+ "learning_rate": 4.9282908358153046e-06,
954
+ "loss": 0.6908,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.5236307633073443,
959
+ "grad_norm": 0.13198168575763702,
960
+ "learning_rate": 4.8659465096248345e-06,
961
+ "loss": 0.6955,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.5274809895081336,
966
+ "grad_norm": 0.11525028944015503,
967
+ "learning_rate": 4.803623033907169e-06,
968
+ "loss": 0.6974,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.5313312157089229,
973
+ "grad_norm": 0.11475051194429398,
974
+ "learning_rate": 4.741330102359441e-06,
975
+ "loss": 0.7089,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.5351814419097122,
980
+ "grad_norm": 0.1284182369709015,
981
+ "learning_rate": 4.679077403927996e-06,
982
+ "loss": 0.6812,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.5390316681105015,
987
+ "grad_norm": 0.12111629545688629,
988
+ "learning_rate": 4.616874621301373e-06,
989
+ "loss": 0.6908,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.5428818943112907,
994
+ "grad_norm": 0.11966368556022644,
995
+ "learning_rate": 4.554731429404293e-06,
996
+ "loss": 0.7065,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.5467321205120801,
1001
+ "grad_norm": 0.12273744493722916,
1002
+ "learning_rate": 4.492657493892822e-06,
1003
+ "loss": 0.6851,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.5505823467128694,
1008
+ "grad_norm": 0.11892206221818924,
1009
+ "learning_rate": 4.430662469650999e-06,
1010
+ "loss": 0.6994,
1011
+ "step": 1430
1012
+ },
1013
+ {
1014
+ "epoch": 0.5544325729136587,
1015
+ "grad_norm": 0.11654770374298096,
1016
+ "learning_rate": 4.368755999289118e-06,
1017
+ "loss": 0.7053,
1018
+ "step": 1440
1019
+ },
1020
+ {
1021
+ "epoch": 0.558282799114448,
1022
+ "grad_norm": 0.12428013235330582,
1023
+ "learning_rate": 4.306947711643949e-06,
1024
+ "loss": 0.6819,
1025
+ "step": 1450
1026
+ },
1027
+ {
1028
+ "epoch": 0.5621330253152372,
1029
+ "grad_norm": 0.11648034304380417,
1030
+ "learning_rate": 4.245247220281068e-06,
1031
+ "loss": 0.6923,
1032
+ "step": 1460
1033
+ },
1034
+ {
1035
+ "epoch": 0.5659832515160266,
1036
+ "grad_norm": 0.11269496381282806,
1037
+ "learning_rate": 4.183664121999587e-06,
1038
+ "loss": 0.6847,
1039
+ "step": 1470
1040
+ },
1041
+ {
1042
+ "epoch": 0.5698334777168159,
1043
+ "grad_norm": 0.12039127945899963,
1044
+ "learning_rate": 4.122207995339474e-06,
1045
+ "loss": 0.6825,
1046
+ "step": 1480
1047
+ },
1048
+ {
1049
+ "epoch": 0.5736837039176051,
1050
+ "grad_norm": 0.114377461373806,
1051
+ "learning_rate": 4.060888399091734e-06,
1052
+ "loss": 0.6912,
1053
+ "step": 1490
1054
+ },
1055
+ {
1056
+ "epoch": 0.5775339301183945,
1057
+ "grad_norm": 0.11231505870819092,
1058
+ "learning_rate": 3.999714870811639e-06,
1059
+ "loss": 0.7071,
1060
+ "step": 1500
1061
+ },
1062
+ {
1063
+ "epoch": 0.5813841563191837,
1064
+ "grad_norm": 0.11400473862886429,
1065
+ "learning_rate": 3.938696925335275e-06,
1066
+ "loss": 0.6952,
1067
+ "step": 1510
1068
+ },
1069
+ {
1070
+ "epoch": 0.5852343825199731,
1071
+ "grad_norm": 0.1194034293293953,
1072
+ "learning_rate": 3.87784405329962e-06,
1073
+ "loss": 0.6992,
1074
+ "step": 1520
1075
+ },
1076
+ {
1077
+ "epoch": 0.5890846087207623,
1078
+ "grad_norm": 0.11712820082902908,
1079
+ "learning_rate": 3.817165719666382e-06,
1080
+ "loss": 0.6955,
1081
+ "step": 1530
1082
+ },
1083
+ {
1084
+ "epoch": 0.5929348349215516,
1085
+ "grad_norm": 0.11396987736225128,
1086
+ "learning_rate": 3.756671362249841e-06,
1087
+ "loss": 0.7107,
1088
+ "step": 1540
1089
+ },
1090
+ {
1091
+ "epoch": 0.596785061122341,
1092
+ "grad_norm": 0.11514045298099518,
1093
+ "learning_rate": 3.6963703902488907e-06,
1094
+ "loss": 0.6988,
1095
+ "step": 1550
1096
+ },
1097
+ {
1098
+ "epoch": 0.6006352873231302,
1099
+ "grad_norm": 0.11802313476800919,
1100
+ "learning_rate": 3.6362721827835655e-06,
1101
+ "loss": 0.7046,
1102
+ "step": 1560
1103
+ },
1104
+ {
1105
+ "epoch": 0.6044855135239195,
1106
+ "grad_norm": 0.12421932071447372,
1107
+ "learning_rate": 3.576386087436212e-06,
1108
+ "loss": 0.6987,
1109
+ "step": 1570
1110
+ },
1111
+ {
1112
+ "epoch": 0.6083357397247088,
1113
+ "grad_norm": 0.11891119927167892,
1114
+ "learning_rate": 3.5167214187975913e-06,
1115
+ "loss": 0.7025,
1116
+ "step": 1580
1117
+ },
1118
+ {
1119
+ "epoch": 0.6121859659254981,
1120
+ "grad_norm": 0.10714492946863174,
1121
+ "learning_rate": 3.457287457018086e-06,
1122
+ "loss": 0.698,
1123
+ "step": 1590
1124
+ },
1125
+ {
1126
+ "epoch": 0.6160361921262875,
1127
+ "grad_norm": 0.1079103872179985,
1128
+ "learning_rate": 3.3980934463643003e-06,
1129
+ "loss": 0.6854,
1130
+ "step": 1600
1131
+ },
1132
+ {
1133
+ "epoch": 0.6198864183270767,
1134
+ "grad_norm": 0.10634385049343109,
1135
+ "learning_rate": 3.3391485937812053e-06,
1136
+ "loss": 0.6635,
1137
+ "step": 1610
1138
+ },
1139
+ {
1140
+ "epoch": 0.623736644527866,
1141
+ "grad_norm": 0.11625577509403229,
1142
+ "learning_rate": 3.280462067460115e-06,
1143
+ "loss": 0.668,
1144
+ "step": 1620
1145
+ },
1146
+ {
1147
+ "epoch": 0.6275868707286553,
1148
+ "grad_norm": 0.11527043581008911,
1149
+ "learning_rate": 3.222042995412669e-06,
1150
+ "loss": 0.7066,
1151
+ "step": 1630
1152
+ },
1153
+ {
1154
+ "epoch": 0.6314370969294446,
1155
+ "grad_norm": 0.1031271293759346,
1156
+ "learning_rate": 3.163900464051089e-06,
1157
+ "loss": 0.6849,
1158
+ "step": 1640
1159
+ },
1160
+ {
1161
+ "epoch": 0.635287323130234,
1162
+ "grad_norm": 0.11496777832508087,
1163
+ "learning_rate": 3.1060435167748826e-06,
1164
+ "loss": 0.698,
1165
+ "step": 1650
1166
+ },
1167
+ {
1168
+ "epoch": 0.6391375493310232,
1169
+ "grad_norm": 0.11435095965862274,
1170
+ "learning_rate": 3.048481152564253e-06,
1171
+ "loss": 0.6855,
1172
+ "step": 1660
1173
+ },
1174
+ {
1175
+ "epoch": 0.6429877755318125,
1176
+ "grad_norm": 0.1263606995344162,
1177
+ "learning_rate": 2.9912223245804017e-06,
1178
+ "loss": 0.7079,
1179
+ "step": 1670
1180
+ },
1181
+ {
1182
+ "epoch": 0.6468380017326018,
1183
+ "grad_norm": 0.12091144919395447,
1184
+ "learning_rate": 2.934275938772981e-06,
1185
+ "loss": 0.6864,
1186
+ "step": 1680
1187
+ },
1188
+ {
1189
+ "epoch": 0.6506882279333911,
1190
+ "grad_norm": 0.11726419627666473,
1191
+ "learning_rate": 2.8776508524948676e-06,
1192
+ "loss": 0.676,
1193
+ "step": 1690
1194
+ },
1195
+ {
1196
+ "epoch": 0.6545384541341804,
1197
+ "grad_norm": 0.10472208261489868,
1198
+ "learning_rate": 2.821355873124503e-06,
1199
+ "loss": 0.6765,
1200
+ "step": 1700
1201
+ },
1202
+ {
1203
+ "epoch": 0.6583886803349697,
1204
+ "grad_norm": 0.12270388007164001,
1205
+ "learning_rate": 2.7653997566960144e-06,
1206
+ "loss": 0.6966,
1207
+ "step": 1710
1208
+ },
1209
+ {
1210
+ "epoch": 0.662238906535759,
1211
+ "grad_norm": 0.1153169497847557,
1212
+ "learning_rate": 2.709791206537309e-06,
1213
+ "loss": 0.6854,
1214
+ "step": 1720
1215
+ },
1216
+ {
1217
+ "epoch": 0.6660891327365482,
1218
+ "grad_norm": 0.11839717626571655,
1219
+ "learning_rate": 2.6545388719163747e-06,
1220
+ "loss": 0.6804,
1221
+ "step": 1730
1222
+ },
1223
+ {
1224
+ "epoch": 0.6699393589373376,
1225
+ "grad_norm": 0.1149812787771225,
1226
+ "learning_rate": 2.599651346695979e-06,
1227
+ "loss": 0.6953,
1228
+ "step": 1740
1229
+ },
1230
+ {
1231
+ "epoch": 0.6737895851381268,
1232
+ "grad_norm": 0.11533249914646149,
1233
+ "learning_rate": 2.545137167997007e-06,
1234
+ "loss": 0.6917,
1235
+ "step": 1750
1236
+ },
1237
+ {
1238
+ "epoch": 0.6776398113389162,
1239
+ "grad_norm": 0.11311566829681396,
1240
+ "learning_rate": 2.4910048148705972e-06,
1241
+ "loss": 0.6756,
1242
+ "step": 1760
1243
+ },
1244
+ {
1245
+ "epoch": 0.6814900375397055,
1246
+ "grad_norm": 0.12470650672912598,
1247
+ "learning_rate": 2.4372627069793327e-06,
1248
+ "loss": 0.7185,
1249
+ "step": 1770
1250
+ },
1251
+ {
1252
+ "epoch": 0.6853402637404947,
1253
+ "grad_norm": 0.12654200196266174,
1254
+ "learning_rate": 2.3839192032876486e-06,
1255
+ "loss": 0.7018,
1256
+ "step": 1780
1257
+ },
1258
+ {
1259
+ "epoch": 0.6891904899412841,
1260
+ "grad_norm": 0.11440509557723999,
1261
+ "learning_rate": 2.3309826007617047e-06,
1262
+ "loss": 0.7126,
1263
+ "step": 1790
1264
+ },
1265
+ {
1266
+ "epoch": 0.6930407161420733,
1267
+ "grad_norm": 0.11220594495534897,
1268
+ "learning_rate": 2.278461133078881e-06,
1269
+ "loss": 0.6764,
1270
+ "step": 1800
1271
+ },
1272
+ {
1273
+ "epoch": 0.6968909423428626,
1274
+ "grad_norm": 0.11749005317687988,
1275
+ "learning_rate": 2.2263629693471258e-06,
1276
+ "loss": 0.6724,
1277
+ "step": 1810
1278
+ },
1279
+ {
1280
+ "epoch": 0.700741168543652,
1281
+ "grad_norm": 0.11141106486320496,
1282
+ "learning_rate": 2.1746962128343414e-06,
1283
+ "loss": 0.6927,
1284
+ "step": 1820
1285
+ },
1286
+ {
1287
+ "epoch": 0.7045913947444412,
1288
+ "grad_norm": 0.10968391597270966,
1289
+ "learning_rate": 2.1234688997080183e-06,
1290
+ "loss": 0.6929,
1291
+ "step": 1830
1292
+ },
1293
+ {
1294
+ "epoch": 0.7084416209452306,
1295
+ "grad_norm": 0.13009047508239746,
1296
+ "learning_rate": 2.0726889977853053e-06,
1297
+ "loss": 0.6904,
1298
+ "step": 1840
1299
+ },
1300
+ {
1301
+ "epoch": 0.7122918471460198,
1302
+ "grad_norm": 0.12730343639850616,
1303
+ "learning_rate": 2.022364405293703e-06,
1304
+ "loss": 0.7027,
1305
+ "step": 1850
1306
+ },
1307
+ {
1308
+ "epoch": 0.7161420733468091,
1309
+ "grad_norm": 0.11544299125671387,
1310
+ "learning_rate": 1.9725029496425814e-06,
1311
+ "loss": 0.6978,
1312
+ "step": 1860
1313
+ },
1314
+ {
1315
+ "epoch": 0.7199922995475985,
1316
+ "grad_norm": 0.10448940098285675,
1317
+ "learning_rate": 1.923112386205727e-06,
1318
+ "loss": 0.6935,
1319
+ "step": 1870
1320
+ },
1321
+ {
1322
+ "epoch": 0.7238425257483877,
1323
+ "grad_norm": 0.11842341721057892,
1324
+ "learning_rate": 1.874200397115079e-06,
1325
+ "loss": 0.7003,
1326
+ "step": 1880
1327
+ },
1328
+ {
1329
+ "epoch": 0.727692751949177,
1330
+ "grad_norm": 0.11826151609420776,
1331
+ "learning_rate": 1.825774590065857e-06,
1332
+ "loss": 0.685,
1333
+ "step": 1890
1334
+ },
1335
+ {
1336
+ "epoch": 0.7315429781499663,
1337
+ "grad_norm": 0.12879306077957153,
1338
+ "learning_rate": 1.7778424971332813e-06,
1339
+ "loss": 0.6946,
1340
+ "step": 1900
1341
+ },
1342
+ {
1343
+ "epoch": 0.7353932043507556,
1344
+ "grad_norm": 0.1125655323266983,
1345
+ "learning_rate": 1.7304115736010407e-06,
1346
+ "loss": 0.6953,
1347
+ "step": 1910
1348
+ },
1349
+ {
1350
+ "epoch": 0.739243430551545,
1351
+ "grad_norm": 0.11263454705476761,
1352
+ "learning_rate": 1.6834891968017108e-06,
1353
+ "loss": 0.6894,
1354
+ "step": 1920
1355
+ },
1356
+ {
1357
+ "epoch": 0.7430936567523342,
1358
+ "grad_norm": 0.11885105818510056,
1359
+ "learning_rate": 1.6370826649692884e-06,
1360
+ "loss": 0.6956,
1361
+ "step": 1930
1362
+ },
1363
+ {
1364
+ "epoch": 0.7469438829531235,
1365
+ "grad_norm": 0.125907763838768,
1366
+ "learning_rate": 1.5911991961040463e-06,
1367
+ "loss": 0.7093,
1368
+ "step": 1940
1369
+ },
1370
+ {
1371
+ "epoch": 0.7507941091539128,
1372
+ "grad_norm": 0.11634443700313568,
1373
+ "learning_rate": 1.545845926849851e-06,
1374
+ "loss": 0.6963,
1375
+ "step": 1950
1376
+ },
1377
+ {
1378
+ "epoch": 0.7546443353547021,
1379
+ "grad_norm": 0.12008029222488403,
1380
+ "learning_rate": 1.5010299113841397e-06,
1381
+ "loss": 0.7119,
1382
+ "step": 1960
1383
+ },
1384
+ {
1385
+ "epoch": 0.7584945615554913,
1386
+ "grad_norm": 0.1207604631781578,
1387
+ "learning_rate": 1.4567581203207238e-06,
1388
+ "loss": 0.7089,
1389
+ "step": 1970
1390
+ },
1391
+ {
1392
+ "epoch": 0.7623447877562807,
1393
+ "grad_norm": 0.11659514158964157,
1394
+ "learning_rate": 1.413037439625593e-06,
1395
+ "loss": 0.6922,
1396
+ "step": 1980
1397
+ },
1398
+ {
1399
+ "epoch": 0.76619501395707,
1400
+ "grad_norm": 0.12583735585212708,
1401
+ "learning_rate": 1.3698746695458847e-06,
1402
+ "loss": 0.6932,
1403
+ "step": 1990
1404
+ },
1405
+ {
1406
+ "epoch": 0.7700452401578592,
1407
+ "grad_norm": 0.11470282822847366,
1408
+ "learning_rate": 1.3272765235521745e-06,
1409
+ "loss": 0.6957,
1410
+ "step": 2000
1411
+ },
1412
+ {
1413
+ "epoch": 0.7738954663586486,
1414
+ "grad_norm": 0.11217229068279266,
1415
+ "learning_rate": 1.2852496272942877e-06,
1416
+ "loss": 0.6962,
1417
+ "step": 2010
1418
+ },
1419
+ {
1420
+ "epoch": 0.7777456925594378,
1421
+ "grad_norm": 0.10893014073371887,
1422
+ "learning_rate": 1.2438005175707396e-06,
1423
+ "loss": 0.6781,
1424
+ "step": 2020
1425
+ },
1426
+ {
1427
+ "epoch": 0.7815959187602272,
1428
+ "grad_norm": 0.1269926279783249,
1429
+ "learning_rate": 1.2029356413120297e-06,
1430
+ "loss": 0.6657,
1431
+ "step": 2030
1432
+ },
1433
+ {
1434
+ "epoch": 0.7854461449610165,
1435
+ "grad_norm": 0.11553125083446503,
1436
+ "learning_rate": 1.1626613545778786e-06,
1437
+ "loss": 0.6831,
1438
+ "step": 2040
1439
+ },
1440
+ {
1441
+ "epoch": 0.7892963711618057,
1442
+ "grad_norm": 0.1132592111825943,
1443
+ "learning_rate": 1.1229839215686294e-06,
1444
+ "loss": 0.6908,
1445
+ "step": 2050
1446
+ },
1447
+ {
1448
+ "epoch": 0.7931465973625951,
1449
+ "grad_norm": 0.11090149730443954,
1450
+ "learning_rate": 1.0839095136509188e-06,
1451
+ "loss": 0.6694,
1452
+ "step": 2060
1453
+ },
1454
+ {
1455
+ "epoch": 0.7969968235633843,
1456
+ "grad_norm": 0.11828117072582245,
1457
+ "learning_rate": 1.045444208397791e-06,
1458
+ "loss": 0.6809,
1459
+ "step": 2070
1460
+ },
1461
+ {
1462
+ "epoch": 0.8008470497641736,
1463
+ "grad_norm": 0.1127399355173111,
1464
+ "learning_rate": 1.007593988643402e-06,
1465
+ "loss": 0.6731,
1466
+ "step": 2080
1467
+ },
1468
+ {
1469
+ "epoch": 0.804697275964963,
1470
+ "grad_norm": 0.10861848294734955,
1471
+ "learning_rate": 9.703647415524613e-07,
1472
+ "loss": 0.6961,
1473
+ "step": 2090
1474
+ },
1475
+ {
1476
+ "epoch": 0.8085475021657522,
1477
+ "grad_norm": 0.11485666781663895,
1478
+ "learning_rate": 9.337622577045485e-07,
1479
+ "loss": 0.6896,
1480
+ "step": 2100
1481
+ },
1482
+ {
1483
+ "epoch": 0.8123977283665416,
1484
+ "grad_norm": 0.11830513924360275,
1485
+ "learning_rate": 8.977922301934588e-07,
1486
+ "loss": 0.7045,
1487
+ "step": 2110
1488
+ },
1489
+ {
1490
+ "epoch": 0.8162479545673308,
1491
+ "grad_norm": 0.11683208495378494,
1492
+ "learning_rate": 8.624602537417009e-07,
1493
+ "loss": 0.7018,
1494
+ "step": 2120
1495
+ },
1496
+ {
1497
+ "epoch": 0.8200981807681201,
1498
+ "grad_norm": 0.11990165710449219,
1499
+ "learning_rate": 8.277718238303123e-07,
1500
+ "loss": 0.6809,
1501
+ "step": 2130
1502
+ },
1503
+ {
1504
+ "epoch": 0.8239484069689095,
1505
+ "grad_norm": 0.11294158548116684,
1506
+ "learning_rate": 7.937323358440935e-07,
1507
+ "loss": 0.6947,
1508
+ "step": 2140
1509
+ },
1510
+ {
1511
+ "epoch": 0.8277986331696987,
1512
+ "grad_norm": 0.13749928772449493,
1513
+ "learning_rate": 7.603470842324223e-07,
1514
+ "loss": 0.6759,
1515
+ "step": 2150
1516
+ },
1517
+ {
1518
+ "epoch": 0.831648859370488,
1519
+ "grad_norm": 0.11176960915327072,
1520
+ "learning_rate": 7.276212616857575e-07,
1521
+ "loss": 0.6971,
1522
+ "step": 2160
1523
+ },
1524
+ {
1525
+ "epoch": 0.8354990855712773,
1526
+ "grad_norm": 0.11233169585466385,
1527
+ "learning_rate": 6.955599583279837e-07,
1528
+ "loss": 0.6979,
1529
+ "step": 2170
1530
+ },
1531
+ {
1532
+ "epoch": 0.8393493117720666,
1533
+ "grad_norm": 0.1153334453701973,
1534
+ "learning_rate": 6.641681609246981e-07,
1535
+ "loss": 0.6897,
1536
+ "step": 2180
1537
+ },
1538
+ {
1539
+ "epoch": 0.8431995379728559,
1540
+ "grad_norm": 0.11645404249429703,
1541
+ "learning_rate": 6.334507521075711e-07,
1542
+ "loss": 0.697,
1543
+ "step": 2190
1544
+ },
1545
+ {
1546
+ "epoch": 0.8470497641736452,
1547
+ "grad_norm": 0.12157875299453735,
1548
+ "learning_rate": 6.034125096149179e-07,
1549
+ "loss": 0.6861,
1550
+ "step": 2200
1551
+ },
1552
+ {
1553
+ "epoch": 0.8508999903744345,
1554
+ "grad_norm": 0.1343868225812912,
1555
+ "learning_rate": 5.740581055485722e-07,
1556
+ "loss": 0.6694,
1557
+ "step": 2210
1558
+ },
1559
+ {
1560
+ "epoch": 0.8547502165752238,
1561
+ "grad_norm": 0.10732828825712204,
1562
+ "learning_rate": 5.453921056471961e-07,
1563
+ "loss": 0.6911,
1564
+ "step": 2220
1565
+ },
1566
+ {
1567
+ "epoch": 0.8586004427760131,
1568
+ "grad_norm": 0.1239297017455101,
1569
+ "learning_rate": 5.174189685761288e-07,
1570
+ "loss": 0.6869,
1571
+ "step": 2230
1572
+ },
1573
+ {
1574
+ "epoch": 0.8624506689768023,
1575
+ "grad_norm": 0.11052046716213226,
1576
+ "learning_rate": 4.901430452338979e-07,
1577
+ "loss": 0.7054,
1578
+ "step": 2240
1579
+ },
1580
+ {
1581
+ "epoch": 0.8663008951775917,
1582
+ "grad_norm": 0.1172553300857544,
1583
+ "learning_rate": 4.6356857807548286e-07,
1584
+ "loss": 0.6861,
1585
+ "step": 2250
1586
+ },
1587
+ {
1588
+ "epoch": 0.870151121378381,
1589
+ "grad_norm": 0.11583167314529419,
1590
+ "learning_rate": 4.376997004524541e-07,
1591
+ "loss": 0.6748,
1592
+ "step": 2260
1593
+ },
1594
+ {
1595
+ "epoch": 0.8740013475791703,
1596
+ "grad_norm": 0.11184808611869812,
1597
+ "learning_rate": 4.12540435970073e-07,
1598
+ "loss": 0.6737,
1599
+ "step": 2270
1600
+ },
1601
+ {
1602
+ "epoch": 0.8778515737799596,
1603
+ "grad_norm": 0.11348651349544525,
1604
+ "learning_rate": 3.880946978614703e-07,
1605
+ "loss": 0.6987,
1606
+ "step": 2280
1607
+ },
1608
+ {
1609
+ "epoch": 0.8817017999807488,
1610
+ "grad_norm": 0.11190453171730042,
1611
+ "learning_rate": 3.643662883789878e-07,
1612
+ "loss": 0.6925,
1613
+ "step": 2290
1614
+ },
1615
+ {
1616
+ "epoch": 0.8855520261815382,
1617
+ "grad_norm": 0.10971527546644211,
1618
+ "learning_rate": 3.41358898202781e-07,
1619
+ "loss": 0.6795,
1620
+ "step": 2300
1621
+ },
1622
+ {
1623
+ "epoch": 0.8894022523823275,
1624
+ "grad_norm": 0.10762371867895126,
1625
+ "learning_rate": 3.190761058667741e-07,
1626
+ "loss": 0.6827,
1627
+ "step": 2310
1628
+ },
1629
+ {
1630
+ "epoch": 0.8932524785831167,
1631
+ "grad_norm": 0.10925888270139694,
1632
+ "learning_rate": 2.9752137720206463e-07,
1633
+ "loss": 0.6887,
1634
+ "step": 2320
1635
+ },
1636
+ {
1637
+ "epoch": 0.8971027047839061,
1638
+ "grad_norm": 0.11614745110273361,
1639
+ "learning_rate": 2.7669806479785067e-07,
1640
+ "loss": 0.7,
1641
+ "step": 2330
1642
+ },
1643
+ {
1644
+ "epoch": 0.9009529309846953,
1645
+ "grad_norm": 0.1134771853685379,
1646
+ "learning_rate": 2.566094074799763e-07,
1647
+ "loss": 0.688,
1648
+ "step": 2340
1649
+ },
1650
+ {
1651
+ "epoch": 0.9048031571854847,
1652
+ "grad_norm": 0.11493874341249466,
1653
+ "learning_rate": 2.3725852980716557e-07,
1654
+ "loss": 0.6789,
1655
+ "step": 2350
1656
+ },
1657
+ {
1658
+ "epoch": 0.908653383386274,
1659
+ "grad_norm": 0.11317159235477448,
1660
+ "learning_rate": 2.1864844158503983e-07,
1661
+ "loss": 0.6898,
1662
+ "step": 2360
1663
+ },
1664
+ {
1665
+ "epoch": 0.9125036095870632,
1666
+ "grad_norm": 0.12302514165639877,
1667
+ "learning_rate": 2.007820373979702e-07,
1668
+ "loss": 0.6782,
1669
+ "step": 2370
1670
+ },
1671
+ {
1672
+ "epoch": 0.9163538357878526,
1673
+ "grad_norm": 0.11632615327835083,
1674
+ "learning_rate": 1.8366209615886532e-07,
1675
+ "loss": 0.685,
1676
+ "step": 2380
1677
+ },
1678
+ {
1679
+ "epoch": 0.9202040619886418,
1680
+ "grad_norm": 0.11606108397245407,
1681
+ "learning_rate": 1.6729128067693613e-07,
1682
+ "loss": 0.7065,
1683
+ "step": 2390
1684
+ },
1685
+ {
1686
+ "epoch": 0.9240542881894311,
1687
+ "grad_norm": 0.11533889919519424,
1688
+ "learning_rate": 1.5167213724353426e-07,
1689
+ "loss": 0.6954,
1690
+ "step": 2400
1691
+ },
1692
+ {
1693
+ "epoch": 0.9279045143902205,
1694
+ "grad_norm": 0.11229761689901352,
1695
+ "learning_rate": 1.3680709523610058e-07,
1696
+ "loss": 0.6782,
1697
+ "step": 2410
1698
+ },
1699
+ {
1700
+ "epoch": 0.9317547405910097,
1701
+ "grad_norm": 0.1148187443614006,
1702
+ "learning_rate": 1.2269846674030418e-07,
1703
+ "loss": 0.7066,
1704
+ "step": 2420
1705
+ },
1706
+ {
1707
+ "epoch": 0.935604966791799,
1708
+ "grad_norm": 0.1273847371339798,
1709
+ "learning_rate": 1.0934844619042595e-07,
1710
+ "loss": 0.6994,
1711
+ "step": 2430
1712
+ },
1713
+ {
1714
+ "epoch": 0.9394551929925883,
1715
+ "grad_norm": 0.12011973559856415,
1716
+ "learning_rate": 9.675911002803961e-08,
1717
+ "loss": 0.687,
1718
+ "step": 2440
1719
+ },
1720
+ {
1721
+ "epoch": 0.9433054191933776,
1722
+ "grad_norm": 0.12248039990663528,
1723
+ "learning_rate": 8.493241637904381e-08,
1724
+ "loss": 0.6833,
1725
+ "step": 2450
1726
+ },
1727
+ {
1728
+ "epoch": 0.9471556453941669,
1729
+ "grad_norm": 0.11954186856746674,
1730
+ "learning_rate": 7.387020474909689e-08,
1731
+ "loss": 0.7063,
1732
+ "step": 2460
1733
+ },
1734
+ {
1735
+ "epoch": 0.9510058715949562,
1736
+ "grad_norm": 0.12359010428190231,
1737
+ "learning_rate": 6.357419573750856e-08,
1738
+ "loss": 0.6935,
1739
+ "step": 2470
1740
+ },
1741
+ {
1742
+ "epoch": 0.9548560977957455,
1743
+ "grad_norm": 0.11653745919466019,
1744
+ "learning_rate": 5.4045990769613365e-08,
1745
+ "loss": 0.6843,
1746
+ "step": 2480
1747
+ },
1748
+ {
1749
+ "epoch": 0.9587063239965348,
1750
+ "grad_norm": 0.11980461329221725,
1751
+ "learning_rate": 4.5287071847694434e-08,
1752
+ "loss": 0.6841,
1753
+ "step": 2490
1754
+ },
1755
+ {
1756
+ "epoch": 0.9625565501973241,
1757
+ "grad_norm": 0.11589724570512772,
1758
+ "learning_rate": 3.7298801320467817e-08,
1759
+ "loss": 0.6868,
1760
+ "step": 2500
1761
+ },
1762
+ {
1763
+ "epoch": 0.9664067763981133,
1764
+ "grad_norm": 0.1163303479552269,
1765
+ "learning_rate": 3.0082421671192576e-08,
1766
+ "loss": 0.6793,
1767
+ "step": 2510
1768
+ },
1769
+ {
1770
+ "epoch": 0.9702570025989027,
1771
+ "grad_norm": 0.11008146405220032,
1772
+ "learning_rate": 2.3639055324410355e-08,
1773
+ "loss": 0.6966,
1774
+ "step": 2520
1775
+ },
1776
+ {
1777
+ "epoch": 0.974107228799692,
1778
+ "grad_norm": 0.11519956588745117,
1779
+ "learning_rate": 1.7969704471369477e-08,
1780
+ "loss": 0.6848,
1781
+ "step": 2530
1782
+ },
1783
+ {
1784
+ "epoch": 0.9779574550004813,
1785
+ "grad_norm": 0.10692780464887619,
1786
+ "learning_rate": 1.3075250914142413e-08,
1787
+ "loss": 0.6882,
1788
+ "step": 2540
1789
+ },
1790
+ {
1791
+ "epoch": 0.9818076812012706,
1792
+ "grad_norm": 0.10820350795984268,
1793
+ "learning_rate": 8.956455928476049e-09,
1794
+ "loss": 0.6991,
1795
+ "step": 2550
1796
+ },
1797
+ {
1798
+ "epoch": 0.9856579074020598,
1799
+ "grad_norm": 0.10323736071586609,
1800
+ "learning_rate": 5.6139601453786275e-09,
1801
+ "loss": 0.6777,
1802
+ "step": 2560
1803
+ },
1804
+ {
1805
+ "epoch": 0.9895081336028492,
1806
+ "grad_norm": 0.12698744237422943,
1807
+ "learning_rate": 3.0482834514805602e-09,
1808
+ "loss": 0.6832,
1809
+ "step": 2570
1810
+ },
1811
+ {
1812
+ "epoch": 0.9933583598036385,
1813
+ "grad_norm": 0.11778843402862549,
1814
+ "learning_rate": 1.2598249081713366e-09,
1815
+ "loss": 0.6638,
1816
+ "step": 2580
1817
+ },
1818
+ {
1819
+ "epoch": 0.9972085860044277,
1820
+ "grad_norm": 0.11506417393684387,
1821
+ "learning_rate": 2.488626895302826e-10,
1822
+ "loss": 0.702,
1823
+ "step": 2590
1824
+ },
1825
+ {
1826
+ "epoch": 0.9999037443449803,
1827
+ "step": 2597,
1828
+ "total_flos": 6.090715375756902e+18,
1829
+ "train_loss": 0.7143064417378188,
1830
+ "train_runtime": 17726.8053,
1831
+ "train_samples_per_second": 37.508,
1832
+ "train_steps_per_second": 0.147
1833
+ }
1834
+ ],
1835
+ "logging_steps": 10,
1836
+ "max_steps": 2597,
1837
+ "num_input_tokens_seen": 332174164,
1838
+ "num_train_epochs": 1,
1839
+ "save_steps": 500,
1840
+ "stateful_callbacks": {
1841
+ "TrainerControl": {
1842
+ "args": {
1843
+ "should_epoch_stop": false,
1844
+ "should_evaluate": false,
1845
+ "should_log": false,
1846
+ "should_save": true,
1847
+ "should_training_stop": true
1848
+ },
1849
+ "attributes": {}
1850
+ }
1851
+ },
1852
+ "total_flos": 6.090715375756902e+18,
1853
+ "train_batch_size": 4,
1854
+ "trial_name": null,
1855
+ "trial_params": null
1856
+ }
continuous/qwen25/pe_lang_g14_448/finetune/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07634782d4b074e8086f2e246bd1ec0e93dc8234e2dc66a46b58a0b5657019d0
3
+ size 6264
continuous/qwen25/pe_lang_g14_448/finetune/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
continuous/qwen25/pe_lang_g14_448/pretrain/config.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_faster_video": false,
3
+ "add_time_instruction": false,
4
+ "architectures": [
5
+ "Qwen2ForCausalLM"
6
+ ],
7
+ "attention_dropout": 0.0,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151645,
10
+ "faster_token_stride": 10,
11
+ "force_sample": false,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 1536,
14
+ "image_aspect_ratio": "square",
15
+ "image_crop_resolution": null,
16
+ "image_grid_pinpoints": null,
17
+ "image_split_resolution": null,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 8960,
20
+ "layer_types": [
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
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention"
49
+ ],
50
+ "max_position_embeddings": 32768,
51
+ "max_window_layers": 21,
52
+ "mm_hidden_size": 1536,
53
+ "mm_newline_position": "grid",
54
+ "mm_patch_merge_type": "flat",
55
+ "mm_projector_lr": null,
56
+ "mm_projector_type": "mlp2x_gelu",
57
+ "mm_resampler_type": null,
58
+ "mm_spatial_pool_mode": "bilinear",
59
+ "mm_spatial_pool_stride": null,
60
+ "mm_tunable_parts": "mm_mlp_adapter",
61
+ "mm_use_im_patch_token": false,
62
+ "mm_use_im_start_end": false,
63
+ "mm_vision_select_feature": "patch",
64
+ "mm_vision_select_layer": -1,
65
+ "mm_vision_tower": "vtb_ssl:pe",
66
+ "mm_vision_tower_lr": null,
67
+ "model_type": "qwen2",
68
+ "num_attention_heads": 12,
69
+ "num_hidden_layers": 28,
70
+ "num_key_value_heads": 2,
71
+ "pad_token_id": 151643,
72
+ "pos_skipping_range": 4096,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_scaling": null,
75
+ "rope_theta": 1000000.0,
76
+ "sliding_window": null,
77
+ "tie_word_embeddings": true,
78
+ "tokenizer_model_max_length": 8192,
79
+ "tokenizer_padding_side": "right",
80
+ "transformers_version": "4.57.6",
81
+ "use_cache": true,
82
+ "use_mm_proj": true,
83
+ "use_pos_skipping": false,
84
+ "use_sliding_window": false,
85
+ "vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/pe/PE-Lang-G14-448/PE-Lang-G14-448.pt",
86
+ "vocab_size": 151936
87
+ }
continuous/qwen25/pe_lang_g14_448/pretrain/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:10296e9c524e52b3024cd74a9957f260dd0c135206e49000fcf5b39753be4459
3
+ size 18888760
continuous/qwen25/pe_lang_g14_448/pretrain/trainer_state.json ADDED
@@ -0,0 +1,421 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 545,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.01834862385321101,
14
+ "grad_norm": 1.3058937788009644,
15
+ "learning_rate": 0.00032142857142857147,
16
+ "loss": 4.8238,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.03669724770642202,
21
+ "grad_norm": 0.3120851218700409,
22
+ "learning_rate": 0.0006785714285714287,
23
+ "loss": 3.6452,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.05504587155963303,
28
+ "grad_norm": 0.08367841690778732,
29
+ "learning_rate": 0.0009999907688176173,
30
+ "loss": 3.2861,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.07339449541284404,
35
+ "grad_norm": 0.09412956982851028,
36
+ "learning_rate": 0.0009988834393115766,
37
+ "loss": 3.2093,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.09174311926605505,
42
+ "grad_norm": 0.10817833244800568,
43
+ "learning_rate": 0.0009959345572940772,
44
+ "loss": 3.1695,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.11009174311926606,
49
+ "grad_norm": 0.075778529047966,
50
+ "learning_rate": 0.0009911550081155983,
51
+ "loss": 3.1156,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.12844036697247707,
56
+ "grad_norm": 0.07564650475978851,
57
+ "learning_rate": 0.0009845624347566433,
58
+ "loss": 3.056,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.14678899082568808,
63
+ "grad_norm": 0.11779944598674774,
64
+ "learning_rate": 0.0009761811727013547,
65
+ "loss": 3.0007,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.1651376146788991,
70
+ "grad_norm": 0.087154321372509,
71
+ "learning_rate": 0.0009660421601067666,
72
+ "loss": 2.935,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.1834862385321101,
77
+ "grad_norm": 0.10271885991096497,
78
+ "learning_rate": 0.0009541828235992926,
79
+ "loss": 2.896,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.2018348623853211,
84
+ "grad_norm": 0.08745159953832626,
85
+ "learning_rate": 0.0009406469401200151,
86
+ "loss": 2.8463,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.22018348623853212,
91
+ "grad_norm": 0.11118987947702408,
92
+ "learning_rate": 0.0009254844753287492,
93
+ "loss": 2.8208,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.23853211009174313,
98
+ "grad_norm": 0.09521395713090897,
99
+ "learning_rate": 0.0009087513991633924,
100
+ "loss": 2.7777,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.25688073394495414,
105
+ "grad_norm": 0.08775249123573303,
106
+ "learning_rate": 0.0008905094792353917,
107
+ "loss": 2.7538,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.27522935779816515,
112
+ "grad_norm": 0.09323590248823166,
113
+ "learning_rate": 0.0008708260528239789,
114
+ "loss": 2.7319,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.29357798165137616,
119
+ "grad_norm": 0.11063627898693085,
120
+ "learning_rate": 0.0008497737783108237,
121
+ "loss": 2.6844,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.3119266055045872,
126
+ "grad_norm": 0.10772769898176193,
127
+ "learning_rate": 0.0008274303669726426,
128
+ "loss": 2.6392,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.3302752293577982,
133
+ "grad_norm": 0.13286030292510986,
134
+ "learning_rate": 0.0008038782961218137,
135
+ "loss": 2.5959,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.3486238532110092,
140
+ "grad_norm": 0.10534809529781342,
141
+ "learning_rate": 0.000779204504653894,
142
+ "loss": 2.5671,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.3669724770642202,
147
+ "grad_norm": 0.09921610355377197,
148
+ "learning_rate": 0.0007535000721258791,
149
+ "loss": 2.5308,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.3853211009174312,
154
+ "grad_norm": 0.09231391549110413,
155
+ "learning_rate": 0.00072685988254984,
156
+ "loss": 2.5133,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.4036697247706422,
161
+ "grad_norm": 0.07924599945545197,
162
+ "learning_rate": 0.0006993822741429906,
163
+ "loss": 2.5182,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.42201834862385323,
168
+ "grad_norm": 0.07964698225259781,
169
+ "learning_rate": 0.0006711686763270818,
170
+ "loss": 2.5048,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.44036697247706424,
175
+ "grad_norm": 0.08114204555749893,
176
+ "learning_rate": 0.0006423232353170831,
177
+ "loss": 2.4908,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.45871559633027525,
182
+ "grad_norm": 0.08259791135787964,
183
+ "learning_rate": 0.0006129524296812335,
184
+ "loss": 2.4914,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.47706422018348627,
189
+ "grad_norm": 0.0790342316031456,
190
+ "learning_rate": 0.0005831646772915651,
191
+ "loss": 2.4943,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.4954128440366973,
196
+ "grad_norm": 0.08207990974187851,
197
+ "learning_rate": 0.0005530699351157809,
198
+ "loss": 2.4714,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.5137614678899083,
203
+ "grad_norm": 0.0746670663356781,
204
+ "learning_rate": 0.0005227792933277943,
205
+ "loss": 2.4704,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.5321100917431193,
210
+ "grad_norm": 0.07783375680446625,
211
+ "learning_rate": 0.0004924045652352092,
212
+ "loss": 2.4629,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.5504587155963303,
217
+ "grad_norm": 0.06455393135547638,
218
+ "learning_rate": 0.0004620578745374633,
219
+ "loss": 2.4598,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.5688073394495413,
224
+ "grad_norm": 0.08729340881109238,
225
+ "learning_rate": 0.0004318512414382059,
226
+ "loss": 2.4674,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.5871559633027523,
231
+ "grad_norm": 0.08352471143007278,
232
+ "learning_rate": 0.0004018961691397148,
233
+ "loss": 2.4645,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.6055045871559633,
238
+ "grad_norm": 0.08595695346593857,
239
+ "learning_rate": 0.0003723032322457458,
240
+ "loss": 2.449,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.6238532110091743,
245
+ "grad_norm": 0.10282009094953537,
246
+ "learning_rate": 0.0003431816685921625,
247
+ "loss": 2.4517,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.6422018348623854,
252
+ "grad_norm": 0.07496550679206848,
253
+ "learning_rate": 0.0003146389760120416,
254
+ "loss": 2.4527,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.6605504587155964,
259
+ "grad_norm": 0.0878211036324501,
260
+ "learning_rate": 0.00028678051552373485,
261
+ "loss": 2.4529,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.6788990825688074,
266
+ "grad_norm": 0.07584265619516373,
267
+ "learning_rate": 0.0002597091224066581,
268
+ "loss": 2.4444,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.6972477064220184,
273
+ "grad_norm": 0.07265079021453857,
274
+ "learning_rate": 0.0002335247266004629,
275
+ "loss": 2.45,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.7155963302752294,
280
+ "grad_norm": 0.08033710718154907,
281
+ "learning_rate": 0.000208323983828828,
282
+ "loss": 2.4416,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.7339449541284404,
287
+ "grad_norm": 0.0716070756316185,
288
+ "learning_rate": 0.0001841999188095224,
289
+ "loss": 2.4347,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.7522935779816514,
294
+ "grad_norm": 0.06784600019454956,
295
+ "learning_rate": 0.00016124158186777676,
296
+ "loss": 2.4398,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.7706422018348624,
301
+ "grad_norm": 0.0642361268401146,
302
+ "learning_rate": 0.0001395337202205211,
303
+ "loss": 2.4325,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.7889908256880734,
308
+ "grad_norm": 0.06370564550161362,
309
+ "learning_rate": 0.00011915646514489292,
310
+ "loss": 2.4305,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.8073394495412844,
315
+ "grad_norm": 0.06430336087942123,
316
+ "learning_rate": 0.00010018503618578818,
317
+ "loss": 2.4389,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.8256880733944955,
322
+ "grad_norm": 0.06707969307899475,
323
+ "learning_rate": 8.268946349432583e-05,
324
+ "loss": 2.433,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.8440366972477065,
329
+ "grad_norm": 0.06378036737442017,
330
+ "learning_rate": 6.673432932217117e-05,
331
+ "loss": 2.4296,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.8623853211009175,
336
+ "grad_norm": 0.058744750916957855,
337
+ "learning_rate": 5.2378529625954684e-05,
338
+ "loss": 2.4198,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.8807339449541285,
343
+ "grad_norm": 0.06687592715024948,
344
+ "learning_rate": 3.9675056661785556e-05,
345
+ "loss": 2.4249,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.8990825688073395,
350
+ "grad_norm": 0.06803320348262787,
351
+ "learning_rate": 2.8670803372379538e-05,
352
+ "loss": 2.4318,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.9174311926605505,
357
+ "grad_norm": 0.05692101642489433,
358
+ "learning_rate": 1.9406390288876586e-05,
359
+ "loss": 2.4381,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.9357798165137615,
364
+ "grad_norm": 0.05888695269823074,
365
+ "learning_rate": 1.191601558631461e-05,
366
+ "loss": 2.4272,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.9541284403669725,
371
+ "grad_norm": 0.0672849714756012,
372
+ "learning_rate": 6.2273288462559305e-06,
373
+ "loss": 2.4223,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.9724770642201835,
378
+ "grad_norm": 0.061098866164684296,
379
+ "learning_rate": 2.361328992552314e-06,
380
+ "loss": 2.4268,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.9908256880733946,
385
+ "grad_norm": 0.058041829615831375,
386
+ "learning_rate": 3.322867770044202e-07,
387
+ "loss": 2.4184,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 1.0,
392
+ "step": 545,
393
+ "total_flos": 5.848521369997804e+17,
394
+ "train_loss": 2.655074600779682,
395
+ "train_runtime": 7750.551,
396
+ "train_samples_per_second": 72.011,
397
+ "train_steps_per_second": 0.07
398
+ }
399
+ ],
400
+ "logging_steps": 10,
401
+ "max_steps": 545,
402
+ "num_input_tokens_seen": 31896544,
403
+ "num_train_epochs": 1,
404
+ "save_steps": 500,
405
+ "stateful_callbacks": {
406
+ "TrainerControl": {
407
+ "args": {
408
+ "should_epoch_stop": false,
409
+ "should_evaluate": false,
410
+ "should_log": false,
411
+ "should_save": false,
412
+ "should_training_stop": false
413
+ },
414
+ "attributes": {}
415
+ }
416
+ },
417
+ "total_flos": 5.848521369997804e+17,
418
+ "train_batch_size": 32,
419
+ "trial_name": null,
420
+ "trial_params": null
421
+ }
continuous/qwen25/pe_lang_g14_448_tiling/finetune/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
+ }
continuous/qwen25/pe_lang_g14_448_tiling/finetune/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
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
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\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|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 %}
continuous/qwen25/pe_lang_g14_448_tiling/finetune/config.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_faster_video": false,
3
+ "add_time_instruction": false,
4
+ "architectures": [
5
+ "LlavaQwenForCausalLM"
6
+ ],
7
+ "attention_dropout": 0.0,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151645,
10
+ "faster_token_stride": 10,
11
+ "force_sample": false,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 1536,
14
+ "image_aspect_ratio": "square",
15
+ "image_crop_resolution": null,
16
+ "image_grid_pinpoints": null,
17
+ "image_split_resolution": null,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 8960,
20
+ "layer_types": [
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
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention"
49
+ ],
50
+ "max_position_embeddings": 32768,
51
+ "max_window_layers": 21,
52
+ "mm_hidden_size": 1536,
53
+ "mm_newline_position": "grid",
54
+ "mm_patch_merge_type": "flat",
55
+ "mm_projector_lr": 0.0001,
56
+ "mm_projector_type": "mlp2x_gelu",
57
+ "mm_resampler_type": null,
58
+ "mm_spatial_pool_mode": "bilinear",
59
+ "mm_spatial_pool_stride": null,
60
+ "mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
61
+ "mm_use_im_patch_token": false,
62
+ "mm_use_im_start_end": false,
63
+ "mm_vision_select_feature": "patch",
64
+ "mm_vision_select_layer": -1,
65
+ "mm_vision_tower": "vtb_ssl:pe",
66
+ "mm_vision_tower_lr": null,
67
+ "model_type": "qwen2",
68
+ "num_attention_heads": 12,
69
+ "num_hidden_layers": 28,
70
+ "num_key_value_heads": 2,
71
+ "pad_token_id": 151643,
72
+ "pos_skipping_range": 4096,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_scaling": null,
75
+ "rope_theta": 1000000.0,
76
+ "sliding_window": null,
77
+ "tie_word_embeddings": true,
78
+ "tokenizer_model_max_length": 8192,
79
+ "tokenizer_padding_side": "right",
80
+ "transformers_version": "4.57.6",
81
+ "use_cache": true,
82
+ "use_mm_proj": true,
83
+ "use_pos_skipping": false,
84
+ "use_sliding_window": false,
85
+ "vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/pe/PE-Lang-G14-448-Tiling/PE-Lang-G14-448-Tiling.pt",
86
+ "vocab_size": 151936
87
+ }
continuous/qwen25/pe_lang_g14_448_tiling/finetune/generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_sample": true,
3
+ "eos_token_id": [
4
+ 151645,
5
+ 151643
6
+ ],
7
+ "pad_token_id": 151643,
8
+ "repetition_penalty": 1.1,
9
+ "temperature": 0.7,
10
+ "top_k": 20,
11
+ "top_p": 0.8,
12
+ "transformers_version": "4.57.6"
13
+ }
continuous/qwen25/pe_lang_g14_448_tiling/finetune/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
continuous/qwen25/pe_lang_g14_448_tiling/finetune/model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4cb5bb8d60607e1430a947609cecfbdeea607d59552f661cc9b703069ec68a9b
3
+ size 4988334784