Hugging Face's logo

Leonardo6
/
sft-clip-instruct-base-patch16

Safetensors
clip
Model card Files
xet
Community
Leonardo6 commited on
Commit
c3c65c2
·
verified ·
1 Parent(s): 769079e

Add files using upload-large-folder tool

Browse files
Files changed (50) hide show
  1. checkpoint-100/config.json +51 -0
  2. checkpoint-100/merges.txt +0 -0
  3. checkpoint-100/preprocessor_config.json +28 -0
  4. checkpoint-100/special_tokens_map.json +30 -0
  5. checkpoint-100/tokenizer.json +0 -0
  6. checkpoint-100/tokenizer_config.json +32 -0
  7. checkpoint-100/trainer_state.json +734 -0
  8. checkpoint-100/vocab.json +0 -0
  9. checkpoint-200/tokenizer.json +0 -0
  10. checkpoint-300/config.json +51 -0
  11. checkpoint-300/preprocessor_config.json +28 -0
  12. checkpoint-300/special_tokens_map.json +30 -0
  13. checkpoint-300/tokenizer.json +0 -0
  14. checkpoint-300/tokenizer_config.json +32 -0
  15. checkpoint-300/trainer_state.json +2134 -0
  16. checkpoint-300/vocab.json +0 -0
  17. checkpoint-400/config.json +51 -0
  18. checkpoint-400/merges.txt +0 -0
  19. checkpoint-400/preprocessor_config.json +28 -0
  20. checkpoint-400/special_tokens_map.json +30 -0
  21. checkpoint-400/tokenizer.json +0 -0
  22. checkpoint-400/tokenizer_config.json +32 -0
  23. checkpoint-400/trainer_state.json +2834 -0
  24. checkpoint-400/vocab.json +0 -0
  25. checkpoint-500/config.json +51 -0
  26. checkpoint-500/merges.txt +0 -0
  27. checkpoint-500/preprocessor_config.json +28 -0
  28. checkpoint-500/special_tokens_map.json +30 -0
  29. checkpoint-500/tokenizer.json +0 -0
  30. checkpoint-500/tokenizer_config.json +32 -0
  31. checkpoint-500/vocab.json +0 -0
  32. checkpoint-600/config.json +51 -0
  33. checkpoint-600/merges.txt +0 -0
  34. checkpoint-600/special_tokens_map.json +30 -0
  35. checkpoint-600/trainer_state.json +0 -0
  36. checkpoint-700/config.json +51 -0
  37. checkpoint-700/merges.txt +0 -0
  38. checkpoint-700/preprocessor_config.json +28 -0
  39. checkpoint-700/special_tokens_map.json +30 -0
  40. checkpoint-700/tokenizer.json +0 -0
  41. checkpoint-700/tokenizer_config.json +32 -0
  42. checkpoint-700/trainer_state.json +0 -0
  43. checkpoint-700/vocab.json +0 -0
  44. checkpoint-800/tokenizer.json +0 -0
  45. checkpoint-900/config.json +51 -0
  46. checkpoint-900/tokenizer.json +0 -0
  47. checkpoint-900/trainer_state.json +0 -0
  48. checkpoint-900/vocab.json +0 -0
  49. config.json +51 -0
  50. tokenizer.json +0 -0
checkpoint-100/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CLIPModel"
4
+ ],
5
+ "bos_token_id": 49406,
6
+ "eos_token_id": 49407,
7
+ "initializer_factor": 1.0,
8
+ "logit_scale_init_value": 2.6592,
9
+ "model_type": "clip",
10
+ "pad_token_id": 49407,
11
+ "projection_dim": 512,
12
+ "text_config": {
13
+ "attention_dropout": 0.0,
14
+ "bos_token_id": 0,
15
+ "dropout": 0.0,
16
+ "eos_token_id": 2,
17
+ "hidden_act": "quick_gelu",
18
+ "hidden_size": 512,
19
+ "initializer_factor": 1.0,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 2048,
22
+ "layer_norm_eps": 1e-05,
23
+ "max_position_embeddings": 77,
24
+ "model_type": "clip_text_model",
25
+ "num_attention_heads": 8,
26
+ "num_hidden_layers": 12,
27
+ "projection_dim": 512,
28
+ "torch_dtype": "float32",
29
+ "vocab_size": 49408
30
+ },
31
+ "torch_dtype": "bfloat16",
32
+ "transformers_version": "4.56.0.dev0",
33
+ "vision_config": {
34
+ "attention_dropout": 0.0,
35
+ "dropout": 0.0,
36
+ "hidden_act": "quick_gelu",
37
+ "hidden_size": 768,
38
+ "image_size": 224,
39
+ "initializer_factor": 1.0,
40
+ "initializer_range": 0.02,
41
+ "intermediate_size": 3072,
42
+ "layer_norm_eps": 1e-05,
43
+ "model_type": "clip_vision_model",
44
+ "num_attention_heads": 12,
45
+ "num_channels": 3,
46
+ "num_hidden_layers": 12,
47
+ "patch_size": 16,
48
+ "projection_dim": 512,
49
+ "torch_dtype": "float32"
50
+ }
51
+ }
checkpoint-100/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-100/preprocessor_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "processor_class": "CLIPProcessor",
23
+ "resample": 3,
24
+ "rescale_factor": 0.00392156862745098,
25
+ "size": {
26
+ "shortest_edge": 224
27
+ }
28
+ }
checkpoint-100/special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|startoftext|>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<|endoftext|>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<|endoftext|>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
checkpoint-100/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-100/tokenizer_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "49406": {
5
+ "content": "<|startoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "49407": {
13
+ "content": "<|endoftext|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ }
20
+ },
21
+ "bos_token": "<|startoftext|>",
22
+ "clean_up_tokenization_spaces": false,
23
+ "do_lower_case": true,
24
+ "eos_token": "<|endoftext|>",
25
+ "errors": "replace",
26
+ "extra_special_tokens": {},
27
+ "model_max_length": 77,
28
+ "pad_token": "<|endoftext|>",
29
+ "processor_class": "CLIPProcessor",
30
+ "tokenizer_class": "CLIPTokenizer",
31
+ "unk_token": "<|endoftext|>"
32
+ }
checkpoint-100/trainer_state.json ADDED
@@ -0,0 +1,734 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.32786885245901637,
6
+ "eval_steps": 500,
7
+ "global_step": 100,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.003278688524590164,
14
+ "grad_norm": 58.999969482421875,
15
+ "learning_rate": 0.0,
16
+ "loss": 3.8125,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.006557377049180328,
21
+ "grad_norm": 56.5999870300293,
22
+ "learning_rate": 3.5714285714285716e-07,
23
+ "loss": 3.8984,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.009836065573770493,
28
+ "grad_norm": 58.442317962646484,
29
+ "learning_rate": 7.142857142857143e-07,
30
+ "loss": 3.8359,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.013114754098360656,
35
+ "grad_norm": 59.13441467285156,
36
+ "learning_rate": 1.0714285714285714e-06,
37
+ "loss": 3.7422,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.01639344262295082,
42
+ "grad_norm": 59.55280685424805,
43
+ "learning_rate": 1.4285714285714286e-06,
44
+ "loss": 3.7383,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.019672131147540985,
49
+ "grad_norm": 57.113956451416016,
50
+ "learning_rate": 1.7857142857142859e-06,
51
+ "loss": 3.8281,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.022950819672131147,
56
+ "grad_norm": 44.69753646850586,
57
+ "learning_rate": 2.1428571428571427e-06,
58
+ "loss": 3.6562,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.02622950819672131,
63
+ "grad_norm": 42.74599075317383,
64
+ "learning_rate": 2.5e-06,
65
+ "loss": 3.668,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.029508196721311476,
70
+ "grad_norm": 27.557815551757812,
71
+ "learning_rate": 2.8571428571428573e-06,
72
+ "loss": 3.3438,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.03278688524590164,
77
+ "grad_norm": 25.87570571899414,
78
+ "learning_rate": 3.2142857142857147e-06,
79
+ "loss": 3.3633,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.036065573770491806,
84
+ "grad_norm": 25.204158782958984,
85
+ "learning_rate": 3.5714285714285718e-06,
86
+ "loss": 3.2852,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.03934426229508197,
91
+ "grad_norm": 25.082277297973633,
92
+ "learning_rate": 3.928571428571429e-06,
93
+ "loss": 3.2188,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.04262295081967213,
98
+ "grad_norm": 21.52399253845215,
99
+ "learning_rate": 4.2857142857142855e-06,
100
+ "loss": 3.1016,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.04590163934426229,
105
+ "grad_norm": 21.302213668823242,
106
+ "learning_rate": 4.642857142857144e-06,
107
+ "loss": 3.0938,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.04918032786885246,
112
+ "grad_norm": 17.48920249938965,
113
+ "learning_rate": 5e-06,
114
+ "loss": 3.0703,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.05245901639344262,
119
+ "grad_norm": 14.723689079284668,
120
+ "learning_rate": 5.357142857142857e-06,
121
+ "loss": 2.7812,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.05573770491803279,
126
+ "grad_norm": 13.425204277038574,
127
+ "learning_rate": 5.7142857142857145e-06,
128
+ "loss": 2.8398,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.05901639344262295,
133
+ "grad_norm": 12.121846199035645,
134
+ "learning_rate": 6.071428571428571e-06,
135
+ "loss": 2.7539,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.06229508196721312,
140
+ "grad_norm": 12.97847843170166,
141
+ "learning_rate": 6.4285714285714295e-06,
142
+ "loss": 2.6719,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.06557377049180328,
147
+ "grad_norm": 12.588919639587402,
148
+ "learning_rate": 6.785714285714287e-06,
149
+ "loss": 2.6758,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.06885245901639345,
154
+ "grad_norm": 10.492659568786621,
155
+ "learning_rate": 7.1428571428571436e-06,
156
+ "loss": 2.6562,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.07213114754098361,
161
+ "grad_norm": 9.676924705505371,
162
+ "learning_rate": 7.500000000000001e-06,
163
+ "loss": 2.6094,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.07540983606557378,
168
+ "grad_norm": 10.222829818725586,
169
+ "learning_rate": 7.857142857142858e-06,
170
+ "loss": 2.6211,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.07868852459016394,
175
+ "grad_norm": 10.020594596862793,
176
+ "learning_rate": 8.214285714285714e-06,
177
+ "loss": 2.5273,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.08196721311475409,
182
+ "grad_norm": 9.404995918273926,
183
+ "learning_rate": 8.571428571428571e-06,
184
+ "loss": 2.5352,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.08524590163934426,
189
+ "grad_norm": 10.194731712341309,
190
+ "learning_rate": 8.92857142857143e-06,
191
+ "loss": 2.543,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.08852459016393442,
196
+ "grad_norm": 9.045639991760254,
197
+ "learning_rate": 9.285714285714288e-06,
198
+ "loss": 2.4961,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.09180327868852459,
203
+ "grad_norm": 9.108716011047363,
204
+ "learning_rate": 9.642857142857144e-06,
205
+ "loss": 2.4336,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.09508196721311475,
210
+ "grad_norm": 9.336297988891602,
211
+ "learning_rate": 1e-05,
212
+ "loss": 2.457,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.09836065573770492,
217
+ "grad_norm": 8.170738220214844,
218
+ "learning_rate": 9.999968638844173e-06,
219
+ "loss": 2.4688,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.10163934426229508,
224
+ "grad_norm": 8.507989883422852,
225
+ "learning_rate": 9.999874555770099e-06,
226
+ "loss": 2.4062,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.10491803278688525,
231
+ "grad_norm": 8.38835334777832,
232
+ "learning_rate": 9.999717751958002e-06,
233
+ "loss": 2.3711,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.10819672131147541,
238
+ "grad_norm": 8.36460018157959,
239
+ "learning_rate": 9.9994982293749e-06,
240
+ "loss": 2.4805,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.11147540983606558,
245
+ "grad_norm": 8.547989845275879,
246
+ "learning_rate": 9.999215990774584e-06,
247
+ "loss": 2.3906,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.11475409836065574,
252
+ "grad_norm": 8.938812255859375,
253
+ "learning_rate": 9.998871039697589e-06,
254
+ "loss": 2.3906,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.1180327868852459,
259
+ "grad_norm": 8.230469703674316,
260
+ "learning_rate": 9.998463380471138e-06,
261
+ "loss": 2.3164,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.12131147540983607,
266
+ "grad_norm": 7.883755683898926,
267
+ "learning_rate": 9.997993018209098e-06,
268
+ "loss": 2.3633,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.12459016393442623,
273
+ "grad_norm": 8.220748901367188,
274
+ "learning_rate": 9.997459958811911e-06,
275
+ "loss": 2.3516,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.12786885245901639,
280
+ "grad_norm": 8.738327980041504,
281
+ "learning_rate": 9.996864208966522e-06,
282
+ "loss": 2.3398,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.13114754098360656,
287
+ "grad_norm": 7.944454669952393,
288
+ "learning_rate": 9.996205776146288e-06,
289
+ "loss": 2.3867,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.13442622950819672,
294
+ "grad_norm": 7.631208419799805,
295
+ "learning_rate": 9.995484668610897e-06,
296
+ "loss": 2.3242,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.1377049180327869,
301
+ "grad_norm": 9.301526069641113,
302
+ "learning_rate": 9.994700895406258e-06,
303
+ "loss": 2.3203,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.14098360655737704,
308
+ "grad_norm": 7.625834941864014,
309
+ "learning_rate": 9.99385446636438e-06,
310
+ "loss": 2.3398,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.14426229508196722,
315
+ "grad_norm": 7.908705234527588,
316
+ "learning_rate": 9.992945392103264e-06,
317
+ "loss": 2.3477,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.14754098360655737,
322
+ "grad_norm": 8.002193450927734,
323
+ "learning_rate": 9.991973684026755e-06,
324
+ "loss": 2.2852,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.15081967213114755,
329
+ "grad_norm": 8.058524131774902,
330
+ "learning_rate": 9.99093935432441e-06,
331
+ "loss": 2.3711,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.1540983606557377,
336
+ "grad_norm": 8.803773880004883,
337
+ "learning_rate": 9.989842415971338e-06,
338
+ "loss": 2.2344,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.15737704918032788,
343
+ "grad_norm": 7.851852893829346,
344
+ "learning_rate": 9.988682882728043e-06,
345
+ "loss": 2.3047,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.16065573770491803,
350
+ "grad_norm": 8.29660415649414,
351
+ "learning_rate": 9.987460769140242e-06,
352
+ "loss": 2.3047,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.16393442622950818,
357
+ "grad_norm": 7.779457092285156,
358
+ "learning_rate": 9.986176090538697e-06,
359
+ "loss": 2.1992,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.16721311475409836,
364
+ "grad_norm": 8.602660179138184,
365
+ "learning_rate": 9.984828863039008e-06,
366
+ "loss": 2.25,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.17049180327868851,
371
+ "grad_norm": 8.079516410827637,
372
+ "learning_rate": 9.983419103541421e-06,
373
+ "loss": 2.3203,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.1737704918032787,
378
+ "grad_norm": 8.443635940551758,
379
+ "learning_rate": 9.981946829730611e-06,
380
+ "loss": 2.3242,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.17704918032786884,
385
+ "grad_norm": 7.873276710510254,
386
+ "learning_rate": 9.980412060075459e-06,
387
+ "loss": 2.2656,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.18032786885245902,
392
+ "grad_norm": 8.633429527282715,
393
+ "learning_rate": 9.978814813828827e-06,
394
+ "loss": 2.2227,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.18360655737704917,
399
+ "grad_norm": 9.135279655456543,
400
+ "learning_rate": 9.97715511102731e-06,
401
+ "loss": 2.2148,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.18688524590163935,
406
+ "grad_norm": 7.584745407104492,
407
+ "learning_rate": 9.975432972490985e-06,
408
+ "loss": 2.2773,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.1901639344262295,
413
+ "grad_norm": 9.086216926574707,
414
+ "learning_rate": 9.973648419823161e-06,
415
+ "loss": 2.2656,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.19344262295081968,
420
+ "grad_norm": 8.223559379577637,
421
+ "learning_rate": 9.971801475410084e-06,
422
+ "loss": 2.2773,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.19672131147540983,
427
+ "grad_norm": 8.726387023925781,
428
+ "learning_rate": 9.969892162420682e-06,
429
+ "loss": 2.3125,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.2,
434
+ "grad_norm": 8.304938316345215,
435
+ "learning_rate": 9.96792050480626e-06,
436
+ "loss": 2.168,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.20327868852459016,
441
+ "grad_norm": 8.20266342163086,
442
+ "learning_rate": 9.965886527300201e-06,
443
+ "loss": 2.25,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.20655737704918034,
448
+ "grad_norm": 7.833676815032959,
449
+ "learning_rate": 9.963790255417663e-06,
450
+ "loss": 2.2422,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.2098360655737705,
455
+ "grad_norm": 7.891449928283691,
456
+ "learning_rate": 9.961631715455245e-06,
457
+ "loss": 2.1758,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.21311475409836064,
462
+ "grad_norm": 8.999512672424316,
463
+ "learning_rate": 9.959410934490673e-06,
464
+ "loss": 2.1484,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.21639344262295082,
469
+ "grad_norm": 7.718410015106201,
470
+ "learning_rate": 9.95712794038245e-06,
471
+ "loss": 2.1172,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.21967213114754097,
476
+ "grad_norm": 8.272541999816895,
477
+ "learning_rate": 9.954782761769509e-06,
478
+ "loss": 2.25,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.22295081967213115,
483
+ "grad_norm": 9.607975006103516,
484
+ "learning_rate": 9.952375428070853e-06,
485
+ "loss": 2.2812,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.2262295081967213,
490
+ "grad_norm": 8.297865867614746,
491
+ "learning_rate": 9.949905969485192e-06,
492
+ "loss": 2.1406,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.22950819672131148,
497
+ "grad_norm": 8.192915916442871,
498
+ "learning_rate": 9.947374416990554e-06,
499
+ "loss": 2.2109,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.23278688524590163,
504
+ "grad_norm": 8.489333152770996,
505
+ "learning_rate": 9.944780802343906e-06,
506
+ "loss": 2.1914,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.2360655737704918,
511
+ "grad_norm": 8.244175910949707,
512
+ "learning_rate": 9.942125158080747e-06,
513
+ "loss": 2.1797,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.23934426229508196,
518
+ "grad_norm": 8.009477615356445,
519
+ "learning_rate": 9.939407517514709e-06,
520
+ "loss": 2.2578,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.24262295081967214,
525
+ "grad_norm": 8.284704208374023,
526
+ "learning_rate": 9.936627914737129e-06,
527
+ "loss": 2.2305,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.2459016393442623,
532
+ "grad_norm": 9.479103088378906,
533
+ "learning_rate": 9.933786384616631e-06,
534
+ "loss": 2.1406,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.24918032786885247,
539
+ "grad_norm": 8.558723449707031,
540
+ "learning_rate": 9.930882962798683e-06,
541
+ "loss": 2.2812,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.25245901639344265,
546
+ "grad_norm": 8.608718872070312,
547
+ "learning_rate": 9.927917685705148e-06,
548
+ "loss": 2.2461,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.25573770491803277,
553
+ "grad_norm": 7.863804817199707,
554
+ "learning_rate": 9.924890590533837e-06,
555
+ "loss": 2.1914,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.25901639344262295,
560
+ "grad_norm": 8.842366218566895,
561
+ "learning_rate": 9.921801715258027e-06,
562
+ "loss": 2.0996,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.26229508196721313,
567
+ "grad_norm": 8.273311614990234,
568
+ "learning_rate": 9.918651098626e-06,
569
+ "loss": 2.1641,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.26557377049180325,
574
+ "grad_norm": 8.011392593383789,
575
+ "learning_rate": 9.915438780160547e-06,
576
+ "loss": 2.168,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.26885245901639343,
581
+ "grad_norm": 7.946706771850586,
582
+ "learning_rate": 9.912164800158474e-06,
583
+ "loss": 2.2227,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.2721311475409836,
588
+ "grad_norm": 9.032291412353516,
589
+ "learning_rate": 9.908829199690106e-06,
590
+ "loss": 2.2891,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.2754098360655738,
595
+ "grad_norm": 8.33333969116211,
596
+ "learning_rate": 9.905432020598751e-06,
597
+ "loss": 2.2344,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.2786885245901639,
602
+ "grad_norm": 7.967901229858398,
603
+ "learning_rate": 9.901973305500197e-06,
604
+ "loss": 2.1172,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.2819672131147541,
609
+ "grad_norm": 8.24918270111084,
610
+ "learning_rate": 9.898453097782164e-06,
611
+ "loss": 2.1602,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.28524590163934427,
616
+ "grad_norm": 8.877111434936523,
617
+ "learning_rate": 9.894871441603766e-06,
618
+ "loss": 2.2383,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.28852459016393445,
623
+ "grad_norm": 8.32304573059082,
624
+ "learning_rate": 9.891228381894954e-06,
625
+ "loss": 2.1719,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.29180327868852457,
630
+ "grad_norm": 9.091806411743164,
631
+ "learning_rate": 9.887523964355953e-06,
632
+ "loss": 2.168,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.29508196721311475,
637
+ "grad_norm": 8.54736614227295,
638
+ "learning_rate": 9.88375823545669e-06,
639
+ "loss": 2.1211,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.2983606557377049,
644
+ "grad_norm": 9.3049955368042,
645
+ "learning_rate": 9.879931242436208e-06,
646
+ "loss": 2.1719,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.3016393442622951,
651
+ "grad_norm": 8.523479461669922,
652
+ "learning_rate": 9.876043033302079e-06,
653
+ "loss": 2.1719,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.30491803278688523,
658
+ "grad_norm": 7.741846561431885,
659
+ "learning_rate": 9.872093656829792e-06,
660
+ "loss": 2.0391,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.3081967213114754,
665
+ "grad_norm": 8.36119270324707,
666
+ "learning_rate": 9.868083162562155e-06,
667
+ "loss": 2.0938,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.3114754098360656,
672
+ "grad_norm": 8.163023948669434,
673
+ "learning_rate": 9.864011600808663e-06,
674
+ "loss": 2.0938,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.31475409836065577,
679
+ "grad_norm": 8.827863693237305,
680
+ "learning_rate": 9.859879022644865e-06,
681
+ "loss": 2.1406,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.3180327868852459,
686
+ "grad_norm": 8.434041023254395,
687
+ "learning_rate": 9.855685479911736e-06,
688
+ "loss": 2.168,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.32131147540983607,
693
+ "grad_norm": 8.144036293029785,
694
+ "learning_rate": 9.851431025215012e-06,
695
+ "loss": 2.0742,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.32459016393442625,
700
+ "grad_norm": 8.843683242797852,
701
+ "learning_rate": 9.847115711924542e-06,
702
+ "loss": 2.1016,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.32786885245901637,
707
+ "grad_norm": 8.300407409667969,
708
+ "learning_rate": 9.842739594173607e-06,
709
+ "loss": 2.1953,
710
+ "step": 100
711
+ }
712
+ ],
713
+ "logging_steps": 1,
714
+ "max_steps": 915,
715
+ "num_input_tokens_seen": 0,
716
+ "num_train_epochs": 3,
717
+ "save_steps": 100,
718
+ "stateful_callbacks": {
719
+ "TrainerControl": {
720
+ "args": {
721
+ "should_epoch_stop": false,
722
+ "should_evaluate": false,
723
+ "should_log": false,
724
+ "should_save": true,
725
+ "should_training_stop": false
726
+ },
727
+ "attributes": {}
728
+ }
729
+ },
730
+ "total_flos": 1.1732503551279104e+16,
731
+ "train_batch_size": 512,
732
+ "trial_name": null,
733
+ "trial_params": null
734
+ }
checkpoint-100/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-200/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-300/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CLIPModel"
4
+ ],
5
+ "bos_token_id": 49406,
6
+ "eos_token_id": 49407,
7
+ "initializer_factor": 1.0,
8
+ "logit_scale_init_value": 2.6592,
9
+ "model_type": "clip",
10
+ "pad_token_id": 49407,
11
+ "projection_dim": 512,
12
+ "text_config": {
13
+ "attention_dropout": 0.0,
14
+ "bos_token_id": 0,
15
+ "dropout": 0.0,
16
+ "eos_token_id": 2,
17
+ "hidden_act": "quick_gelu",
18
+ "hidden_size": 512,
19
+ "initializer_factor": 1.0,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 2048,
22
+ "layer_norm_eps": 1e-05,
23
+ "max_position_embeddings": 77,
24
+ "model_type": "clip_text_model",
25
+ "num_attention_heads": 8,
26
+ "num_hidden_layers": 12,
27
+ "projection_dim": 512,
28
+ "torch_dtype": "float32",
29
+ "vocab_size": 49408
30
+ },
31
+ "torch_dtype": "bfloat16",
32
+ "transformers_version": "4.56.0.dev0",
33
+ "vision_config": {
34
+ "attention_dropout": 0.0,
35
+ "dropout": 0.0,
36
+ "hidden_act": "quick_gelu",
37
+ "hidden_size": 768,
38
+ "image_size": 224,
39
+ "initializer_factor": 1.0,
40
+ "initializer_range": 0.02,
41
+ "intermediate_size": 3072,
42
+ "layer_norm_eps": 1e-05,
43
+ "model_type": "clip_vision_model",
44
+ "num_attention_heads": 12,
45
+ "num_channels": 3,
46
+ "num_hidden_layers": 12,
47
+ "patch_size": 16,
48
+ "projection_dim": 512,
49
+ "torch_dtype": "float32"
50
+ }
51
+ }
checkpoint-300/preprocessor_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "processor_class": "CLIPProcessor",
23
+ "resample": 3,
24
+ "rescale_factor": 0.00392156862745098,
25
+ "size": {
26
+ "shortest_edge": 224
27
+ }
28
+ }
checkpoint-300/special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|startoftext|>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<|endoftext|>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<|endoftext|>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
checkpoint-300/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-300/tokenizer_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "49406": {
5
+ "content": "<|startoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "49407": {
13
+ "content": "<|endoftext|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ }
20
+ },
21
+ "bos_token": "<|startoftext|>",
22
+ "clean_up_tokenization_spaces": false,
23
+ "do_lower_case": true,
24
+ "eos_token": "<|endoftext|>",
25
+ "errors": "replace",
26
+ "extra_special_tokens": {},
27
+ "model_max_length": 77,
28
+ "pad_token": "<|endoftext|>",
29
+ "processor_class": "CLIPProcessor",
30
+ "tokenizer_class": "CLIPTokenizer",
31
+ "unk_token": "<|endoftext|>"
32
+ }
checkpoint-300/trainer_state.json ADDED
@@ -0,0 +1,2134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.9836065573770492,
6
+ "eval_steps": 500,
7
+ "global_step": 300,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.003278688524590164,
14
+ "grad_norm": 58.999969482421875,
15
+ "learning_rate": 0.0,
16
+ "loss": 3.8125,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.006557377049180328,
21
+ "grad_norm": 56.5999870300293,
22
+ "learning_rate": 3.5714285714285716e-07,
23
+ "loss": 3.8984,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.009836065573770493,
28
+ "grad_norm": 58.442317962646484,
29
+ "learning_rate": 7.142857142857143e-07,
30
+ "loss": 3.8359,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.013114754098360656,
35
+ "grad_norm": 59.13441467285156,
36
+ "learning_rate": 1.0714285714285714e-06,
37
+ "loss": 3.7422,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.01639344262295082,
42
+ "grad_norm": 59.55280685424805,
43
+ "learning_rate": 1.4285714285714286e-06,
44
+ "loss": 3.7383,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.019672131147540985,
49
+ "grad_norm": 57.113956451416016,
50
+ "learning_rate": 1.7857142857142859e-06,
51
+ "loss": 3.8281,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.022950819672131147,
56
+ "grad_norm": 44.69753646850586,
57
+ "learning_rate": 2.1428571428571427e-06,
58
+ "loss": 3.6562,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.02622950819672131,
63
+ "grad_norm": 42.74599075317383,
64
+ "learning_rate": 2.5e-06,
65
+ "loss": 3.668,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.029508196721311476,
70
+ "grad_norm": 27.557815551757812,
71
+ "learning_rate": 2.8571428571428573e-06,
72
+ "loss": 3.3438,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.03278688524590164,
77
+ "grad_norm": 25.87570571899414,
78
+ "learning_rate": 3.2142857142857147e-06,
79
+ "loss": 3.3633,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.036065573770491806,
84
+ "grad_norm": 25.204158782958984,
85
+ "learning_rate": 3.5714285714285718e-06,
86
+ "loss": 3.2852,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.03934426229508197,
91
+ "grad_norm": 25.082277297973633,
92
+ "learning_rate": 3.928571428571429e-06,
93
+ "loss": 3.2188,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.04262295081967213,
98
+ "grad_norm": 21.52399253845215,
99
+ "learning_rate": 4.2857142857142855e-06,
100
+ "loss": 3.1016,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.04590163934426229,
105
+ "grad_norm": 21.302213668823242,
106
+ "learning_rate": 4.642857142857144e-06,
107
+ "loss": 3.0938,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.04918032786885246,
112
+ "grad_norm": 17.48920249938965,
113
+ "learning_rate": 5e-06,
114
+ "loss": 3.0703,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.05245901639344262,
119
+ "grad_norm": 14.723689079284668,
120
+ "learning_rate": 5.357142857142857e-06,
121
+ "loss": 2.7812,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.05573770491803279,
126
+ "grad_norm": 13.425204277038574,
127
+ "learning_rate": 5.7142857142857145e-06,
128
+ "loss": 2.8398,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.05901639344262295,
133
+ "grad_norm": 12.121846199035645,
134
+ "learning_rate": 6.071428571428571e-06,
135
+ "loss": 2.7539,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.06229508196721312,
140
+ "grad_norm": 12.97847843170166,
141
+ "learning_rate": 6.4285714285714295e-06,
142
+ "loss": 2.6719,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.06557377049180328,
147
+ "grad_norm": 12.588919639587402,
148
+ "learning_rate": 6.785714285714287e-06,
149
+ "loss": 2.6758,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.06885245901639345,
154
+ "grad_norm": 10.492659568786621,
155
+ "learning_rate": 7.1428571428571436e-06,
156
+ "loss": 2.6562,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.07213114754098361,
161
+ "grad_norm": 9.676924705505371,
162
+ "learning_rate": 7.500000000000001e-06,
163
+ "loss": 2.6094,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.07540983606557378,
168
+ "grad_norm": 10.222829818725586,
169
+ "learning_rate": 7.857142857142858e-06,
170
+ "loss": 2.6211,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.07868852459016394,
175
+ "grad_norm": 10.020594596862793,
176
+ "learning_rate": 8.214285714285714e-06,
177
+ "loss": 2.5273,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.08196721311475409,
182
+ "grad_norm": 9.404995918273926,
183
+ "learning_rate": 8.571428571428571e-06,
184
+ "loss": 2.5352,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.08524590163934426,
189
+ "grad_norm": 10.194731712341309,
190
+ "learning_rate": 8.92857142857143e-06,
191
+ "loss": 2.543,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.08852459016393442,
196
+ "grad_norm": 9.045639991760254,
197
+ "learning_rate": 9.285714285714288e-06,
198
+ "loss": 2.4961,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.09180327868852459,
203
+ "grad_norm": 9.108716011047363,
204
+ "learning_rate": 9.642857142857144e-06,
205
+ "loss": 2.4336,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.09508196721311475,
210
+ "grad_norm": 9.336297988891602,
211
+ "learning_rate": 1e-05,
212
+ "loss": 2.457,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.09836065573770492,
217
+ "grad_norm": 8.170738220214844,
218
+ "learning_rate": 9.999968638844173e-06,
219
+ "loss": 2.4688,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.10163934426229508,
224
+ "grad_norm": 8.507989883422852,
225
+ "learning_rate": 9.999874555770099e-06,
226
+ "loss": 2.4062,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.10491803278688525,
231
+ "grad_norm": 8.38835334777832,
232
+ "learning_rate": 9.999717751958002e-06,
233
+ "loss": 2.3711,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.10819672131147541,
238
+ "grad_norm": 8.36460018157959,
239
+ "learning_rate": 9.9994982293749e-06,
240
+ "loss": 2.4805,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.11147540983606558,
245
+ "grad_norm": 8.547989845275879,
246
+ "learning_rate": 9.999215990774584e-06,
247
+ "loss": 2.3906,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.11475409836065574,
252
+ "grad_norm": 8.938812255859375,
253
+ "learning_rate": 9.998871039697589e-06,
254
+ "loss": 2.3906,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.1180327868852459,
259
+ "grad_norm": 8.230469703674316,
260
+ "learning_rate": 9.998463380471138e-06,
261
+ "loss": 2.3164,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.12131147540983607,
266
+ "grad_norm": 7.883755683898926,
267
+ "learning_rate": 9.997993018209098e-06,
268
+ "loss": 2.3633,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.12459016393442623,
273
+ "grad_norm": 8.220748901367188,
274
+ "learning_rate": 9.997459958811911e-06,
275
+ "loss": 2.3516,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.12786885245901639,
280
+ "grad_norm": 8.738327980041504,
281
+ "learning_rate": 9.996864208966522e-06,
282
+ "loss": 2.3398,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.13114754098360656,
287
+ "grad_norm": 7.944454669952393,
288
+ "learning_rate": 9.996205776146288e-06,
289
+ "loss": 2.3867,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.13442622950819672,
294
+ "grad_norm": 7.631208419799805,
295
+ "learning_rate": 9.995484668610897e-06,
296
+ "loss": 2.3242,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.1377049180327869,
301
+ "grad_norm": 9.301526069641113,
302
+ "learning_rate": 9.994700895406258e-06,
303
+ "loss": 2.3203,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.14098360655737704,
308
+ "grad_norm": 7.625834941864014,
309
+ "learning_rate": 9.99385446636438e-06,
310
+ "loss": 2.3398,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.14426229508196722,
315
+ "grad_norm": 7.908705234527588,
316
+ "learning_rate": 9.992945392103264e-06,
317
+ "loss": 2.3477,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.14754098360655737,
322
+ "grad_norm": 8.002193450927734,
323
+ "learning_rate": 9.991973684026755e-06,
324
+ "loss": 2.2852,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.15081967213114755,
329
+ "grad_norm": 8.058524131774902,
330
+ "learning_rate": 9.99093935432441e-06,
331
+ "loss": 2.3711,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.1540983606557377,
336
+ "grad_norm": 8.803773880004883,
337
+ "learning_rate": 9.989842415971338e-06,
338
+ "loss": 2.2344,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.15737704918032788,
343
+ "grad_norm": 7.851852893829346,
344
+ "learning_rate": 9.988682882728043e-06,
345
+ "loss": 2.3047,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.16065573770491803,
350
+ "grad_norm": 8.29660415649414,
351
+ "learning_rate": 9.987460769140242e-06,
352
+ "loss": 2.3047,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.16393442622950818,
357
+ "grad_norm": 7.779457092285156,
358
+ "learning_rate": 9.986176090538697e-06,
359
+ "loss": 2.1992,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.16721311475409836,
364
+ "grad_norm": 8.602660179138184,
365
+ "learning_rate": 9.984828863039008e-06,
366
+ "loss": 2.25,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.17049180327868851,
371
+ "grad_norm": 8.079516410827637,
372
+ "learning_rate": 9.983419103541421e-06,
373
+ "loss": 2.3203,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.1737704918032787,
378
+ "grad_norm": 8.443635940551758,
379
+ "learning_rate": 9.981946829730611e-06,
380
+ "loss": 2.3242,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.17704918032786884,
385
+ "grad_norm": 7.873276710510254,
386
+ "learning_rate": 9.980412060075459e-06,
387
+ "loss": 2.2656,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.18032786885245902,
392
+ "grad_norm": 8.633429527282715,
393
+ "learning_rate": 9.978814813828827e-06,
394
+ "loss": 2.2227,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.18360655737704917,
399
+ "grad_norm": 9.135279655456543,
400
+ "learning_rate": 9.97715511102731e-06,
401
+ "loss": 2.2148,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.18688524590163935,
406
+ "grad_norm": 7.584745407104492,
407
+ "learning_rate": 9.975432972490985e-06,
408
+ "loss": 2.2773,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.1901639344262295,
413
+ "grad_norm": 9.086216926574707,
414
+ "learning_rate": 9.973648419823161e-06,
415
+ "loss": 2.2656,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.19344262295081968,
420
+ "grad_norm": 8.223559379577637,
421
+ "learning_rate": 9.971801475410084e-06,
422
+ "loss": 2.2773,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.19672131147540983,
427
+ "grad_norm": 8.726387023925781,
428
+ "learning_rate": 9.969892162420682e-06,
429
+ "loss": 2.3125,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.2,
434
+ "grad_norm": 8.304938316345215,
435
+ "learning_rate": 9.96792050480626e-06,
436
+ "loss": 2.168,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.20327868852459016,
441
+ "grad_norm": 8.20266342163086,
442
+ "learning_rate": 9.965886527300201e-06,
443
+ "loss": 2.25,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.20655737704918034,
448
+ "grad_norm": 7.833676815032959,
449
+ "learning_rate": 9.963790255417663e-06,
450
+ "loss": 2.2422,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.2098360655737705,
455
+ "grad_norm": 7.891449928283691,
456
+ "learning_rate": 9.961631715455245e-06,
457
+ "loss": 2.1758,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.21311475409836064,
462
+ "grad_norm": 8.999512672424316,
463
+ "learning_rate": 9.959410934490673e-06,
464
+ "loss": 2.1484,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.21639344262295082,
469
+ "grad_norm": 7.718410015106201,
470
+ "learning_rate": 9.95712794038245e-06,
471
+ "loss": 2.1172,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.21967213114754097,
476
+ "grad_norm": 8.272541999816895,
477
+ "learning_rate": 9.954782761769509e-06,
478
+ "loss": 2.25,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.22295081967213115,
483
+ "grad_norm": 9.607975006103516,
484
+ "learning_rate": 9.952375428070853e-06,
485
+ "loss": 2.2812,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.2262295081967213,
490
+ "grad_norm": 8.297865867614746,
491
+ "learning_rate": 9.949905969485192e-06,
492
+ "loss": 2.1406,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.22950819672131148,
497
+ "grad_norm": 8.192915916442871,
498
+ "learning_rate": 9.947374416990554e-06,
499
+ "loss": 2.2109,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.23278688524590163,
504
+ "grad_norm": 8.489333152770996,
505
+ "learning_rate": 9.944780802343906e-06,
506
+ "loss": 2.1914,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.2360655737704918,
511
+ "grad_norm": 8.244175910949707,
512
+ "learning_rate": 9.942125158080747e-06,
513
+ "loss": 2.1797,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.23934426229508196,
518
+ "grad_norm": 8.009477615356445,
519
+ "learning_rate": 9.939407517514709e-06,
520
+ "loss": 2.2578,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.24262295081967214,
525
+ "grad_norm": 8.284704208374023,
526
+ "learning_rate": 9.936627914737129e-06,
527
+ "loss": 2.2305,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.2459016393442623,
532
+ "grad_norm": 9.479103088378906,
533
+ "learning_rate": 9.933786384616631e-06,
534
+ "loss": 2.1406,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.24918032786885247,
539
+ "grad_norm": 8.558723449707031,
540
+ "learning_rate": 9.930882962798683e-06,
541
+ "loss": 2.2812,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.25245901639344265,
546
+ "grad_norm": 8.608718872070312,
547
+ "learning_rate": 9.927917685705148e-06,
548
+ "loss": 2.2461,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.25573770491803277,
553
+ "grad_norm": 7.863804817199707,
554
+ "learning_rate": 9.924890590533837e-06,
555
+ "loss": 2.1914,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.25901639344262295,
560
+ "grad_norm": 8.842366218566895,
561
+ "learning_rate": 9.921801715258027e-06,
562
+ "loss": 2.0996,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.26229508196721313,
567
+ "grad_norm": 8.273311614990234,
568
+ "learning_rate": 9.918651098626e-06,
569
+ "loss": 2.1641,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.26557377049180325,
574
+ "grad_norm": 8.011392593383789,
575
+ "learning_rate": 9.915438780160547e-06,
576
+ "loss": 2.168,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.26885245901639343,
581
+ "grad_norm": 7.946706771850586,
582
+ "learning_rate": 9.912164800158474e-06,
583
+ "loss": 2.2227,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.2721311475409836,
588
+ "grad_norm": 9.032291412353516,
589
+ "learning_rate": 9.908829199690106e-06,
590
+ "loss": 2.2891,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.2754098360655738,
595
+ "grad_norm": 8.33333969116211,
596
+ "learning_rate": 9.905432020598751e-06,
597
+ "loss": 2.2344,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.2786885245901639,
602
+ "grad_norm": 7.967901229858398,
603
+ "learning_rate": 9.901973305500197e-06,
604
+ "loss": 2.1172,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.2819672131147541,
609
+ "grad_norm": 8.24918270111084,
610
+ "learning_rate": 9.898453097782164e-06,
611
+ "loss": 2.1602,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.28524590163934427,
616
+ "grad_norm": 8.877111434936523,
617
+ "learning_rate": 9.894871441603766e-06,
618
+ "loss": 2.2383,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.28852459016393445,
623
+ "grad_norm": 8.32304573059082,
624
+ "learning_rate": 9.891228381894954e-06,
625
+ "loss": 2.1719,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.29180327868852457,
630
+ "grad_norm": 9.091806411743164,
631
+ "learning_rate": 9.887523964355953e-06,
632
+ "loss": 2.168,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.29508196721311475,
637
+ "grad_norm": 8.54736614227295,
638
+ "learning_rate": 9.88375823545669e-06,
639
+ "loss": 2.1211,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.2983606557377049,
644
+ "grad_norm": 9.3049955368042,
645
+ "learning_rate": 9.879931242436208e-06,
646
+ "loss": 2.1719,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.3016393442622951,
651
+ "grad_norm": 8.523479461669922,
652
+ "learning_rate": 9.876043033302079e-06,
653
+ "loss": 2.1719,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.30491803278688523,
658
+ "grad_norm": 7.741846561431885,
659
+ "learning_rate": 9.872093656829792e-06,
660
+ "loss": 2.0391,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.3081967213114754,
665
+ "grad_norm": 8.36119270324707,
666
+ "learning_rate": 9.868083162562155e-06,
667
+ "loss": 2.0938,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.3114754098360656,
672
+ "grad_norm": 8.163023948669434,
673
+ "learning_rate": 9.864011600808663e-06,
674
+ "loss": 2.0938,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.31475409836065577,
679
+ "grad_norm": 8.827863693237305,
680
+ "learning_rate": 9.859879022644865e-06,
681
+ "loss": 2.1406,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.3180327868852459,
686
+ "grad_norm": 8.434041023254395,
687
+ "learning_rate": 9.855685479911736e-06,
688
+ "loss": 2.168,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.32131147540983607,
693
+ "grad_norm": 8.144036293029785,
694
+ "learning_rate": 9.851431025215012e-06,
695
+ "loss": 2.0742,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.32459016393442625,
700
+ "grad_norm": 8.843683242797852,
701
+ "learning_rate": 9.847115711924542e-06,
702
+ "loss": 2.1016,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.32786885245901637,
707
+ "grad_norm": 8.300407409667969,
708
+ "learning_rate": 9.842739594173607e-06,
709
+ "loss": 2.1953,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 0.33114754098360655,
714
+ "grad_norm": 8.825713157653809,
715
+ "learning_rate": 9.838302726858258e-06,
716
+ "loss": 2.1406,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 0.3344262295081967,
721
+ "grad_norm": 8.763128280639648,
722
+ "learning_rate": 9.833805165636603e-06,
723
+ "loss": 2.1523,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 0.3377049180327869,
728
+ "grad_norm": 8.53940486907959,
729
+ "learning_rate": 9.829246966928135e-06,
730
+ "loss": 2.0703,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 0.34098360655737703,
735
+ "grad_norm": 8.830872535705566,
736
+ "learning_rate": 9.824628187913001e-06,
737
+ "loss": 2.1602,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 0.3442622950819672,
742
+ "grad_norm": 8.062509536743164,
743
+ "learning_rate": 9.819948886531305e-06,
744
+ "loss": 2.1211,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 0.3475409836065574,
749
+ "grad_norm": 8.8229398727417,
750
+ "learning_rate": 9.815209121482363e-06,
751
+ "loss": 2.1172,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 0.35081967213114756,
756
+ "grad_norm": 9.78320598602295,
757
+ "learning_rate": 9.810408952223982e-06,
758
+ "loss": 2.0586,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 0.3540983606557377,
763
+ "grad_norm": 8.267208099365234,
764
+ "learning_rate": 9.805548438971702e-06,
765
+ "loss": 2.1406,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 0.35737704918032787,
770
+ "grad_norm": 10.9806489944458,
771
+ "learning_rate": 9.80062764269805e-06,
772
+ "loss": 2.1035,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 0.36065573770491804,
777
+ "grad_norm": 9.699831008911133,
778
+ "learning_rate": 9.795646625131771e-06,
779
+ "loss": 2.1797,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 0.3639344262295082,
784
+ "grad_norm": 10.373594284057617,
785
+ "learning_rate": 9.790605448757049e-06,
786
+ "loss": 2.1445,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 0.36721311475409835,
791
+ "grad_norm": 8.51201343536377,
792
+ "learning_rate": 9.785504176812733e-06,
793
+ "loss": 2.1875,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 0.3704918032786885,
798
+ "grad_norm": 10.090012550354004,
799
+ "learning_rate": 9.780342873291535e-06,
800
+ "loss": 2.2109,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 0.3737704918032787,
805
+ "grad_norm": 8.939981460571289,
806
+ "learning_rate": 9.775121602939234e-06,
807
+ "loss": 2.0273,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 0.3770491803278688,
812
+ "grad_norm": 8.556815147399902,
813
+ "learning_rate": 9.76984043125386e-06,
814
+ "loss": 2.002,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 0.380327868852459,
819
+ "grad_norm": 9.271822929382324,
820
+ "learning_rate": 9.764499424484873e-06,
821
+ "loss": 2.0742,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 0.3836065573770492,
826
+ "grad_norm": 9.027301788330078,
827
+ "learning_rate": 9.759098649632326e-06,
828
+ "loss": 2.1133,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 0.38688524590163936,
833
+ "grad_norm": 9.81689453125,
834
+ "learning_rate": 9.753638174446042e-06,
835
+ "loss": 2.0645,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 0.3901639344262295,
840
+ "grad_norm": 9.42857551574707,
841
+ "learning_rate": 9.748118067424742e-06,
842
+ "loss": 2.1172,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 0.39344262295081966,
847
+ "grad_norm": 8.636916160583496,
848
+ "learning_rate": 9.742538397815203e-06,
849
+ "loss": 2.0859,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 0.39672131147540984,
854
+ "grad_norm": 9.01211166381836,
855
+ "learning_rate": 9.73689923561138e-06,
856
+ "loss": 2.0605,
857
+ "step": 121
858
+ },
859
+ {
860
+ "epoch": 0.4,
861
+ "grad_norm": 8.320965766906738,
862
+ "learning_rate": 9.73120065155353e-06,
863
+ "loss": 2.0352,
864
+ "step": 122
865
+ },
866
+ {
867
+ "epoch": 0.40327868852459015,
868
+ "grad_norm": 8.969964027404785,
869
+ "learning_rate": 9.725442717127324e-06,
870
+ "loss": 2.1055,
871
+ "step": 123
872
+ },
873
+ {
874
+ "epoch": 0.4065573770491803,
875
+ "grad_norm": 7.825037002563477,
876
+ "learning_rate": 9.719625504562959e-06,
877
+ "loss": 2.1133,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 0.4098360655737705,
882
+ "grad_norm": 9.594252586364746,
883
+ "learning_rate": 9.713749086834234e-06,
884
+ "loss": 2.1367,
885
+ "step": 125
886
+ },
887
+ {
888
+ "epoch": 0.4131147540983607,
889
+ "grad_norm": 9.286437034606934,
890
+ "learning_rate": 9.707813537657652e-06,
891
+ "loss": 2.0293,
892
+ "step": 126
893
+ },
894
+ {
895
+ "epoch": 0.4163934426229508,
896
+ "grad_norm": 11.219290733337402,
897
+ "learning_rate": 9.701818931491485e-06,
898
+ "loss": 2.1992,
899
+ "step": 127
900
+ },
901
+ {
902
+ "epoch": 0.419672131147541,
903
+ "grad_norm": 8.125706672668457,
904
+ "learning_rate": 9.695765343534846e-06,
905
+ "loss": 2.0273,
906
+ "step": 128
907
+ },
908
+ {
909
+ "epoch": 0.42295081967213116,
910
+ "grad_norm": 9.294842720031738,
911
+ "learning_rate": 9.689652849726739e-06,
912
+ "loss": 2.082,
913
+ "step": 129
914
+ },
915
+ {
916
+ "epoch": 0.4262295081967213,
917
+ "grad_norm": 8.297966957092285,
918
+ "learning_rate": 9.683481526745114e-06,
919
+ "loss": 2.084,
920
+ "step": 130
921
+ },
922
+ {
923
+ "epoch": 0.42950819672131146,
924
+ "grad_norm": 9.68375015258789,
925
+ "learning_rate": 9.677251452005897e-06,
926
+ "loss": 2.0996,
927
+ "step": 131
928
+ },
929
+ {
930
+ "epoch": 0.43278688524590164,
931
+ "grad_norm": 8.27182388305664,
932
+ "learning_rate": 9.670962703662032e-06,
933
+ "loss": 2.0312,
934
+ "step": 132
935
+ },
936
+ {
937
+ "epoch": 0.4360655737704918,
938
+ "grad_norm": 9.681833267211914,
939
+ "learning_rate": 9.664615360602478e-06,
940
+ "loss": 2.0547,
941
+ "step": 133
942
+ },
943
+ {
944
+ "epoch": 0.43934426229508194,
945
+ "grad_norm": 8.213553428649902,
946
+ "learning_rate": 9.658209502451245e-06,
947
+ "loss": 1.9746,
948
+ "step": 134
949
+ },
950
+ {
951
+ "epoch": 0.4426229508196721,
952
+ "grad_norm": 9.513401985168457,
953
+ "learning_rate": 9.651745209566379e-06,
954
+ "loss": 2.0625,
955
+ "step": 135
956
+ },
957
+ {
958
+ "epoch": 0.4459016393442623,
959
+ "grad_norm": 9.364225387573242,
960
+ "learning_rate": 9.645222563038959e-06,
961
+ "loss": 2.0078,
962
+ "step": 136
963
+ },
964
+ {
965
+ "epoch": 0.4491803278688525,
966
+ "grad_norm": 8.659600257873535,
967
+ "learning_rate": 9.638641644692077e-06,
968
+ "loss": 2.0469,
969
+ "step": 137
970
+ },
971
+ {
972
+ "epoch": 0.4524590163934426,
973
+ "grad_norm": 9.638226509094238,
974
+ "learning_rate": 9.632002537079816e-06,
975
+ "loss": 2.0391,
976
+ "step": 138
977
+ },
978
+ {
979
+ "epoch": 0.4557377049180328,
980
+ "grad_norm": 9.299017906188965,
981
+ "learning_rate": 9.625305323486214e-06,
982
+ "loss": 2.0508,
983
+ "step": 139
984
+ },
985
+ {
986
+ "epoch": 0.45901639344262296,
987
+ "grad_norm": 9.940561294555664,
988
+ "learning_rate": 9.61855008792421e-06,
989
+ "loss": 2.082,
990
+ "step": 140
991
+ },
992
+ {
993
+ "epoch": 0.46229508196721314,
994
+ "grad_norm": 9.618109703063965,
995
+ "learning_rate": 9.611736915134605e-06,
996
+ "loss": 1.998,
997
+ "step": 141
998
+ },
999
+ {
1000
+ "epoch": 0.46557377049180326,
1001
+ "grad_norm": 8.827125549316406,
1002
+ "learning_rate": 9.604865890584987e-06,
1003
+ "loss": 2.1055,
1004
+ "step": 142
1005
+ },
1006
+ {
1007
+ "epoch": 0.46885245901639344,
1008
+ "grad_norm": 10.079780578613281,
1009
+ "learning_rate": 9.597937100468668e-06,
1010
+ "loss": 2.0859,
1011
+ "step": 143
1012
+ },
1013
+ {
1014
+ "epoch": 0.4721311475409836,
1015
+ "grad_norm": 9.007691383361816,
1016
+ "learning_rate": 9.590950631703589e-06,
1017
+ "loss": 2.125,
1018
+ "step": 144
1019
+ },
1020
+ {
1021
+ "epoch": 0.47540983606557374,
1022
+ "grad_norm": 9.498886108398438,
1023
+ "learning_rate": 9.583906571931248e-06,
1024
+ "loss": 2.1172,
1025
+ "step": 145
1026
+ },
1027
+ {
1028
+ "epoch": 0.4786885245901639,
1029
+ "grad_norm": 10.918264389038086,
1030
+ "learning_rate": 9.576805009515588e-06,
1031
+ "loss": 2.0508,
1032
+ "step": 146
1033
+ },
1034
+ {
1035
+ "epoch": 0.4819672131147541,
1036
+ "grad_norm": 8.792062759399414,
1037
+ "learning_rate": 9.569646033541888e-06,
1038
+ "loss": 2.1484,
1039
+ "step": 147
1040
+ },
1041
+ {
1042
+ "epoch": 0.4852459016393443,
1043
+ "grad_norm": 9.490158081054688,
1044
+ "learning_rate": 9.562429733815655e-06,
1045
+ "loss": 2.1133,
1046
+ "step": 148
1047
+ },
1048
+ {
1049
+ "epoch": 0.4885245901639344,
1050
+ "grad_norm": 9.344221115112305,
1051
+ "learning_rate": 9.55515620086149e-06,
1052
+ "loss": 2.0684,
1053
+ "step": 149
1054
+ },
1055
+ {
1056
+ "epoch": 0.4918032786885246,
1057
+ "grad_norm": 10.885297775268555,
1058
+ "learning_rate": 9.54782552592195e-06,
1059
+ "loss": 2.0957,
1060
+ "step": 150
1061
+ },
1062
+ {
1063
+ "epoch": 0.49508196721311476,
1064
+ "grad_norm": 8.593646049499512,
1065
+ "learning_rate": 9.540437800956412e-06,
1066
+ "loss": 2.0625,
1067
+ "step": 151
1068
+ },
1069
+ {
1070
+ "epoch": 0.49836065573770494,
1071
+ "grad_norm": 8.966227531433105,
1072
+ "learning_rate": 9.532993118639915e-06,
1073
+ "loss": 2.0,
1074
+ "step": 152
1075
+ },
1076
+ {
1077
+ "epoch": 0.5016393442622951,
1078
+ "grad_norm": 10.044122695922852,
1079
+ "learning_rate": 9.525491572361995e-06,
1080
+ "loss": 2.0176,
1081
+ "step": 153
1082
+ },
1083
+ {
1084
+ "epoch": 0.5049180327868853,
1085
+ "grad_norm": 8.53621768951416,
1086
+ "learning_rate": 9.517933256225516e-06,
1087
+ "loss": 2.1094,
1088
+ "step": 154
1089
+ },
1090
+ {
1091
+ "epoch": 0.5081967213114754,
1092
+ "grad_norm": 10.106084823608398,
1093
+ "learning_rate": 9.510318265045488e-06,
1094
+ "loss": 2.0195,
1095
+ "step": 155
1096
+ },
1097
+ {
1098
+ "epoch": 0.5114754098360655,
1099
+ "grad_norm": 9.872212409973145,
1100
+ "learning_rate": 9.502646694347886e-06,
1101
+ "loss": 2.1055,
1102
+ "step": 156
1103
+ },
1104
+ {
1105
+ "epoch": 0.5147540983606558,
1106
+ "grad_norm": 8.695504188537598,
1107
+ "learning_rate": 9.494918640368435e-06,
1108
+ "loss": 1.9961,
1109
+ "step": 157
1110
+ },
1111
+ {
1112
+ "epoch": 0.5180327868852459,
1113
+ "grad_norm": 9.583698272705078,
1114
+ "learning_rate": 9.487134200051422e-06,
1115
+ "loss": 2.0801,
1116
+ "step": 158
1117
+ },
1118
+ {
1119
+ "epoch": 0.521311475409836,
1120
+ "grad_norm": 9.81774616241455,
1121
+ "learning_rate": 9.479293471048461e-06,
1122
+ "loss": 2.043,
1123
+ "step": 159
1124
+ },
1125
+ {
1126
+ "epoch": 0.5245901639344263,
1127
+ "grad_norm": 8.542972564697266,
1128
+ "learning_rate": 9.471396551717283e-06,
1129
+ "loss": 1.9922,
1130
+ "step": 160
1131
+ },
1132
+ {
1133
+ "epoch": 0.5278688524590164,
1134
+ "grad_norm": 9.894759178161621,
1135
+ "learning_rate": 9.463443541120496e-06,
1136
+ "loss": 2.0469,
1137
+ "step": 161
1138
+ },
1139
+ {
1140
+ "epoch": 0.5311475409836065,
1141
+ "grad_norm": 9.396486282348633,
1142
+ "learning_rate": 9.45543453902434e-06,
1143
+ "loss": 1.9688,
1144
+ "step": 162
1145
+ },
1146
+ {
1147
+ "epoch": 0.5344262295081967,
1148
+ "grad_norm": 9.620316505432129,
1149
+ "learning_rate": 9.447369645897441e-06,
1150
+ "loss": 2.0391,
1151
+ "step": 163
1152
+ },
1153
+ {
1154
+ "epoch": 0.5377049180327869,
1155
+ "grad_norm": 10.236593246459961,
1156
+ "learning_rate": 9.439248962909547e-06,
1157
+ "loss": 1.957,
1158
+ "step": 164
1159
+ },
1160
+ {
1161
+ "epoch": 0.5409836065573771,
1162
+ "grad_norm": 11.980149269104004,
1163
+ "learning_rate": 9.431072591930261e-06,
1164
+ "loss": 2.0293,
1165
+ "step": 165
1166
+ },
1167
+ {
1168
+ "epoch": 0.5442622950819672,
1169
+ "grad_norm": 14.74476146697998,
1170
+ "learning_rate": 9.42284063552776e-06,
1171
+ "loss": 1.9844,
1172
+ "step": 166
1173
+ },
1174
+ {
1175
+ "epoch": 0.5475409836065573,
1176
+ "grad_norm": 9.34021282196045,
1177
+ "learning_rate": 9.414553196967508e-06,
1178
+ "loss": 2.0703,
1179
+ "step": 167
1180
+ },
1181
+ {
1182
+ "epoch": 0.5508196721311476,
1183
+ "grad_norm": 12.805119514465332,
1184
+ "learning_rate": 9.406210380210971e-06,
1185
+ "loss": 2.0117,
1186
+ "step": 168
1187
+ },
1188
+ {
1189
+ "epoch": 0.5540983606557377,
1190
+ "grad_norm": 9.839883804321289,
1191
+ "learning_rate": 9.397812289914297e-06,
1192
+ "loss": 1.9727,
1193
+ "step": 169
1194
+ },
1195
+ {
1196
+ "epoch": 0.5573770491803278,
1197
+ "grad_norm": 14.596024513244629,
1198
+ "learning_rate": 9.389359031427013e-06,
1199
+ "loss": 2.0742,
1200
+ "step": 170
1201
+ },
1202
+ {
1203
+ "epoch": 0.5606557377049181,
1204
+ "grad_norm": 13.544656753540039,
1205
+ "learning_rate": 9.380850710790701e-06,
1206
+ "loss": 2.0742,
1207
+ "step": 171
1208
+ },
1209
+ {
1210
+ "epoch": 0.5639344262295082,
1211
+ "grad_norm": 9.45289421081543,
1212
+ "learning_rate": 9.372287434737673e-06,
1213
+ "loss": 2.0586,
1214
+ "step": 172
1215
+ },
1216
+ {
1217
+ "epoch": 0.5672131147540984,
1218
+ "grad_norm": 10.174077987670898,
1219
+ "learning_rate": 9.363669310689617e-06,
1220
+ "loss": 2.1367,
1221
+ "step": 173
1222
+ },
1223
+ {
1224
+ "epoch": 0.5704918032786885,
1225
+ "grad_norm": 12.690667152404785,
1226
+ "learning_rate": 9.354996446756268e-06,
1227
+ "loss": 1.9395,
1228
+ "step": 174
1229
+ },
1230
+ {
1231
+ "epoch": 0.5737704918032787,
1232
+ "grad_norm": 10.625876426696777,
1233
+ "learning_rate": 9.346268951734042e-06,
1234
+ "loss": 2.0215,
1235
+ "step": 175
1236
+ },
1237
+ {
1238
+ "epoch": 0.5770491803278689,
1239
+ "grad_norm": 11.856694221496582,
1240
+ "learning_rate": 9.337486935104672e-06,
1241
+ "loss": 1.9766,
1242
+ "step": 176
1243
+ },
1244
+ {
1245
+ "epoch": 0.580327868852459,
1246
+ "grad_norm": 10.2452974319458,
1247
+ "learning_rate": 9.328650507033832e-06,
1248
+ "loss": 2.0586,
1249
+ "step": 177
1250
+ },
1251
+ {
1252
+ "epoch": 0.5836065573770491,
1253
+ "grad_norm": 12.020367622375488,
1254
+ "learning_rate": 9.319759778369764e-06,
1255
+ "loss": 2.0352,
1256
+ "step": 178
1257
+ },
1258
+ {
1259
+ "epoch": 0.5868852459016394,
1260
+ "grad_norm": 11.364824295043945,
1261
+ "learning_rate": 9.310814860641874e-06,
1262
+ "loss": 2.0352,
1263
+ "step": 179
1264
+ },
1265
+ {
1266
+ "epoch": 0.5901639344262295,
1267
+ "grad_norm": 12.439066886901855,
1268
+ "learning_rate": 9.301815866059353e-06,
1269
+ "loss": 2.0703,
1270
+ "step": 180
1271
+ },
1272
+ {
1273
+ "epoch": 0.5934426229508196,
1274
+ "grad_norm": 12.66345500946045,
1275
+ "learning_rate": 9.292762907509742e-06,
1276
+ "loss": 1.9922,
1277
+ "step": 181
1278
+ },
1279
+ {
1280
+ "epoch": 0.5967213114754099,
1281
+ "grad_norm": 11.868645668029785,
1282
+ "learning_rate": 9.283656098557545e-06,
1283
+ "loss": 1.9863,
1284
+ "step": 182
1285
+ },
1286
+ {
1287
+ "epoch": 0.6,
1288
+ "grad_norm": 11.235689163208008,
1289
+ "learning_rate": 9.27449555344278e-06,
1290
+ "loss": 2.0195,
1291
+ "step": 183
1292
+ },
1293
+ {
1294
+ "epoch": 0.6032786885245902,
1295
+ "grad_norm": 10.719315528869629,
1296
+ "learning_rate": 9.26528138707956e-06,
1297
+ "loss": 1.9883,
1298
+ "step": 184
1299
+ },
1300
+ {
1301
+ "epoch": 0.6065573770491803,
1302
+ "grad_norm": 11.960482597351074,
1303
+ "learning_rate": 9.256013715054648e-06,
1304
+ "loss": 1.9082,
1305
+ "step": 185
1306
+ },
1307
+ {
1308
+ "epoch": 0.6098360655737705,
1309
+ "grad_norm": 9.54014778137207,
1310
+ "learning_rate": 9.24669265362601e-06,
1311
+ "loss": 2.0156,
1312
+ "step": 186
1313
+ },
1314
+ {
1315
+ "epoch": 0.6131147540983607,
1316
+ "grad_norm": 10.753348350524902,
1317
+ "learning_rate": 9.237318319721346e-06,
1318
+ "loss": 1.9746,
1319
+ "step": 187
1320
+ },
1321
+ {
1322
+ "epoch": 0.6163934426229508,
1323
+ "grad_norm": 9.75783634185791,
1324
+ "learning_rate": 9.227890830936634e-06,
1325
+ "loss": 2.0195,
1326
+ "step": 188
1327
+ },
1328
+ {
1329
+ "epoch": 0.6196721311475409,
1330
+ "grad_norm": 15.201295852661133,
1331
+ "learning_rate": 9.218410305534657e-06,
1332
+ "loss": 2.0117,
1333
+ "step": 189
1334
+ },
1335
+ {
1336
+ "epoch": 0.6229508196721312,
1337
+ "grad_norm": 10.927349090576172,
1338
+ "learning_rate": 9.208876862443504e-06,
1339
+ "loss": 1.9316,
1340
+ "step": 190
1341
+ },
1342
+ {
1343
+ "epoch": 0.6262295081967213,
1344
+ "grad_norm": 13.822660446166992,
1345
+ "learning_rate": 9.199290621255095e-06,
1346
+ "loss": 2.0371,
1347
+ "step": 191
1348
+ },
1349
+ {
1350
+ "epoch": 0.6295081967213115,
1351
+ "grad_norm": 13.426775932312012,
1352
+ "learning_rate": 9.189651702223672e-06,
1353
+ "loss": 2.0273,
1354
+ "step": 192
1355
+ },
1356
+ {
1357
+ "epoch": 0.6327868852459017,
1358
+ "grad_norm": 9.90354061126709,
1359
+ "learning_rate": 9.179960226264288e-06,
1360
+ "loss": 1.957,
1361
+ "step": 193
1362
+ },
1363
+ {
1364
+ "epoch": 0.6360655737704918,
1365
+ "grad_norm": 20.340970993041992,
1366
+ "learning_rate": 9.170216314951301e-06,
1367
+ "loss": 2.0488,
1368
+ "step": 194
1369
+ },
1370
+ {
1371
+ "epoch": 0.639344262295082,
1372
+ "grad_norm": 9.366877555847168,
1373
+ "learning_rate": 9.16042009051684e-06,
1374
+ "loss": 1.9688,
1375
+ "step": 195
1376
+ },
1377
+ {
1378
+ "epoch": 0.6426229508196721,
1379
+ "grad_norm": 14.084966659545898,
1380
+ "learning_rate": 9.150571675849273e-06,
1381
+ "loss": 2.0938,
1382
+ "step": 196
1383
+ },
1384
+ {
1385
+ "epoch": 0.6459016393442623,
1386
+ "grad_norm": 16.520889282226562,
1387
+ "learning_rate": 9.140671194491666e-06,
1388
+ "loss": 2.0508,
1389
+ "step": 197
1390
+ },
1391
+ {
1392
+ "epoch": 0.6491803278688525,
1393
+ "grad_norm": 13.539576530456543,
1394
+ "learning_rate": 9.130718770640235e-06,
1395
+ "loss": 1.9922,
1396
+ "step": 198
1397
+ },
1398
+ {
1399
+ "epoch": 0.6524590163934426,
1400
+ "grad_norm": 14.922486305236816,
1401
+ "learning_rate": 9.120714529142784e-06,
1402
+ "loss": 2.0371,
1403
+ "step": 199
1404
+ },
1405
+ {
1406
+ "epoch": 0.6557377049180327,
1407
+ "grad_norm": 12.108205795288086,
1408
+ "learning_rate": 9.110658595497148e-06,
1409
+ "loss": 1.9902,
1410
+ "step": 200
1411
+ },
1412
+ {
1413
+ "epoch": 0.659016393442623,
1414
+ "grad_norm": 10.225913047790527,
1415
+ "learning_rate": 9.100551095849603e-06,
1416
+ "loss": 1.8574,
1417
+ "step": 201
1418
+ },
1419
+ {
1420
+ "epoch": 0.6622950819672131,
1421
+ "grad_norm": 10.973288536071777,
1422
+ "learning_rate": 9.090392156993301e-06,
1423
+ "loss": 1.9902,
1424
+ "step": 202
1425
+ },
1426
+ {
1427
+ "epoch": 0.6655737704918033,
1428
+ "grad_norm": 12.73608684539795,
1429
+ "learning_rate": 9.080181906366668e-06,
1430
+ "loss": 1.9551,
1431
+ "step": 203
1432
+ },
1433
+ {
1434
+ "epoch": 0.6688524590163935,
1435
+ "grad_norm": 11.886429786682129,
1436
+ "learning_rate": 9.069920472051804e-06,
1437
+ "loss": 1.957,
1438
+ "step": 204
1439
+ },
1440
+ {
1441
+ "epoch": 0.6721311475409836,
1442
+ "grad_norm": 13.453667640686035,
1443
+ "learning_rate": 9.05960798277289e-06,
1444
+ "loss": 1.9512,
1445
+ "step": 205
1446
+ },
1447
+ {
1448
+ "epoch": 0.6754098360655738,
1449
+ "grad_norm": 19.135602951049805,
1450
+ "learning_rate": 9.04924456789456e-06,
1451
+ "loss": 1.9688,
1452
+ "step": 206
1453
+ },
1454
+ {
1455
+ "epoch": 0.6786885245901639,
1456
+ "grad_norm": 16.149311065673828,
1457
+ "learning_rate": 9.038830357420276e-06,
1458
+ "loss": 1.9688,
1459
+ "step": 207
1460
+ },
1461
+ {
1462
+ "epoch": 0.6819672131147541,
1463
+ "grad_norm": 14.444845199584961,
1464
+ "learning_rate": 9.028365481990713e-06,
1465
+ "loss": 2.0625,
1466
+ "step": 208
1467
+ },
1468
+ {
1469
+ "epoch": 0.6852459016393443,
1470
+ "grad_norm": 12.08676528930664,
1471
+ "learning_rate": 9.017850072882107e-06,
1472
+ "loss": 1.9121,
1473
+ "step": 209
1474
+ },
1475
+ {
1476
+ "epoch": 0.6885245901639344,
1477
+ "grad_norm": 11.756418228149414,
1478
+ "learning_rate": 9.007284262004608e-06,
1479
+ "loss": 1.9102,
1480
+ "step": 210
1481
+ },
1482
+ {
1483
+ "epoch": 0.6918032786885245,
1484
+ "grad_norm": 12.56296443939209,
1485
+ "learning_rate": 8.996668181900636e-06,
1486
+ "loss": 2.0312,
1487
+ "step": 211
1488
+ },
1489
+ {
1490
+ "epoch": 0.6950819672131148,
1491
+ "grad_norm": 10.218338966369629,
1492
+ "learning_rate": 8.986001965743205e-06,
1493
+ "loss": 1.9629,
1494
+ "step": 212
1495
+ },
1496
+ {
1497
+ "epoch": 0.6983606557377049,
1498
+ "grad_norm": 9.9441556930542,
1499
+ "learning_rate": 8.975285747334267e-06,
1500
+ "loss": 1.9844,
1501
+ "step": 213
1502
+ },
1503
+ {
1504
+ "epoch": 0.7016393442622951,
1505
+ "grad_norm": 10.612693786621094,
1506
+ "learning_rate": 8.964519661103014e-06,
1507
+ "loss": 1.9746,
1508
+ "step": 214
1509
+ },
1510
+ {
1511
+ "epoch": 0.7049180327868853,
1512
+ "grad_norm": 10.07176399230957,
1513
+ "learning_rate": 8.953703842104212e-06,
1514
+ "loss": 1.9531,
1515
+ "step": 215
1516
+ },
1517
+ {
1518
+ "epoch": 0.7081967213114754,
1519
+ "grad_norm": 9.502864837646484,
1520
+ "learning_rate": 8.942838426016496e-06,
1521
+ "loss": 1.9492,
1522
+ "step": 216
1523
+ },
1524
+ {
1525
+ "epoch": 0.7114754098360656,
1526
+ "grad_norm": 9.91391372680664,
1527
+ "learning_rate": 8.931923549140668e-06,
1528
+ "loss": 1.9844,
1529
+ "step": 217
1530
+ },
1531
+ {
1532
+ "epoch": 0.7147540983606557,
1533
+ "grad_norm": 9.780139923095703,
1534
+ "learning_rate": 8.920959348397988e-06,
1535
+ "loss": 1.9473,
1536
+ "step": 218
1537
+ },
1538
+ {
1539
+ "epoch": 0.7180327868852459,
1540
+ "grad_norm": 15.030618667602539,
1541
+ "learning_rate": 8.909945961328462e-06,
1542
+ "loss": 1.9922,
1543
+ "step": 219
1544
+ },
1545
+ {
1546
+ "epoch": 0.7213114754098361,
1547
+ "grad_norm": 10.924345016479492,
1548
+ "learning_rate": 8.898883526089108e-06,
1549
+ "loss": 2.0195,
1550
+ "step": 220
1551
+ },
1552
+ {
1553
+ "epoch": 0.7245901639344262,
1554
+ "grad_norm": 10.839616775512695,
1555
+ "learning_rate": 8.887772181452228e-06,
1556
+ "loss": 2.0234,
1557
+ "step": 221
1558
+ },
1559
+ {
1560
+ "epoch": 0.7278688524590164,
1561
+ "grad_norm": 11.212786674499512,
1562
+ "learning_rate": 8.876612066803666e-06,
1563
+ "loss": 1.9727,
1564
+ "step": 222
1565
+ },
1566
+ {
1567
+ "epoch": 0.7311475409836066,
1568
+ "grad_norm": 9.742143630981445,
1569
+ "learning_rate": 8.86540332214106e-06,
1570
+ "loss": 1.9844,
1571
+ "step": 223
1572
+ },
1573
+ {
1574
+ "epoch": 0.7344262295081967,
1575
+ "grad_norm": 12.224595069885254,
1576
+ "learning_rate": 8.854146088072087e-06,
1577
+ "loss": 2.0566,
1578
+ "step": 224
1579
+ },
1580
+ {
1581
+ "epoch": 0.7377049180327869,
1582
+ "grad_norm": 9.907976150512695,
1583
+ "learning_rate": 8.842840505812691e-06,
1584
+ "loss": 2.0215,
1585
+ "step": 225
1586
+ },
1587
+ {
1588
+ "epoch": 0.740983606557377,
1589
+ "grad_norm": 10.605637550354004,
1590
+ "learning_rate": 8.83148671718533e-06,
1591
+ "loss": 1.959,
1592
+ "step": 226
1593
+ },
1594
+ {
1595
+ "epoch": 0.7442622950819672,
1596
+ "grad_norm": 10.371489524841309,
1597
+ "learning_rate": 8.82008486461717e-06,
1598
+ "loss": 1.9492,
1599
+ "step": 227
1600
+ },
1601
+ {
1602
+ "epoch": 0.7475409836065574,
1603
+ "grad_norm": 13.147061347961426,
1604
+ "learning_rate": 8.808635091138327e-06,
1605
+ "loss": 1.9375,
1606
+ "step": 228
1607
+ },
1608
+ {
1609
+ "epoch": 0.7508196721311475,
1610
+ "grad_norm": 13.442245483398438,
1611
+ "learning_rate": 8.79713754038005e-06,
1612
+ "loss": 1.9219,
1613
+ "step": 229
1614
+ },
1615
+ {
1616
+ "epoch": 0.7540983606557377,
1617
+ "grad_norm": 10.87454891204834,
1618
+ "learning_rate": 8.785592356572931e-06,
1619
+ "loss": 1.9609,
1620
+ "step": 230
1621
+ },
1622
+ {
1623
+ "epoch": 0.7573770491803279,
1624
+ "grad_norm": 12.752360343933105,
1625
+ "learning_rate": 8.773999684545096e-06,
1626
+ "loss": 1.9414,
1627
+ "step": 231
1628
+ },
1629
+ {
1630
+ "epoch": 0.760655737704918,
1631
+ "grad_norm": 10.660215377807617,
1632
+ "learning_rate": 8.76235966972038e-06,
1633
+ "loss": 1.9395,
1634
+ "step": 232
1635
+ },
1636
+ {
1637
+ "epoch": 0.7639344262295082,
1638
+ "grad_norm": 13.825894355773926,
1639
+ "learning_rate": 8.750672458116513e-06,
1640
+ "loss": 1.9102,
1641
+ "step": 233
1642
+ },
1643
+ {
1644
+ "epoch": 0.7672131147540984,
1645
+ "grad_norm": 11.399506568908691,
1646
+ "learning_rate": 8.738938196343278e-06,
1647
+ "loss": 2.002,
1648
+ "step": 234
1649
+ },
1650
+ {
1651
+ "epoch": 0.7704918032786885,
1652
+ "grad_norm": 10.732869148254395,
1653
+ "learning_rate": 8.72715703160068e-06,
1654
+ "loss": 1.8887,
1655
+ "step": 235
1656
+ },
1657
+ {
1658
+ "epoch": 0.7737704918032787,
1659
+ "grad_norm": 13.854918479919434,
1660
+ "learning_rate": 8.715329111677102e-06,
1661
+ "loss": 1.9922,
1662
+ "step": 236
1663
+ },
1664
+ {
1665
+ "epoch": 0.7770491803278688,
1666
+ "grad_norm": 12.330432891845703,
1667
+ "learning_rate": 8.703454584947433e-06,
1668
+ "loss": 1.9395,
1669
+ "step": 237
1670
+ },
1671
+ {
1672
+ "epoch": 0.780327868852459,
1673
+ "grad_norm": 9.50846004486084,
1674
+ "learning_rate": 8.691533600371227e-06,
1675
+ "loss": 1.8965,
1676
+ "step": 238
1677
+ },
1678
+ {
1679
+ "epoch": 0.7836065573770492,
1680
+ "grad_norm": 10.240988731384277,
1681
+ "learning_rate": 8.67956630749083e-06,
1682
+ "loss": 1.9121,
1683
+ "step": 239
1684
+ },
1685
+ {
1686
+ "epoch": 0.7868852459016393,
1687
+ "grad_norm": 10.392232894897461,
1688
+ "learning_rate": 8.667552856429494e-06,
1689
+ "loss": 1.8809,
1690
+ "step": 240
1691
+ },
1692
+ {
1693
+ "epoch": 0.7901639344262295,
1694
+ "grad_norm": 10.917051315307617,
1695
+ "learning_rate": 8.655493397889503e-06,
1696
+ "loss": 1.9805,
1697
+ "step": 241
1698
+ },
1699
+ {
1700
+ "epoch": 0.7934426229508197,
1701
+ "grad_norm": 10.163880348205566,
1702
+ "learning_rate": 8.64338808315028e-06,
1703
+ "loss": 1.8711,
1704
+ "step": 242
1705
+ },
1706
+ {
1707
+ "epoch": 0.7967213114754098,
1708
+ "grad_norm": 11.161616325378418,
1709
+ "learning_rate": 8.631237064066491e-06,
1710
+ "loss": 1.9102,
1711
+ "step": 243
1712
+ },
1713
+ {
1714
+ "epoch": 0.8,
1715
+ "grad_norm": 11.414627075195312,
1716
+ "learning_rate": 8.61904049306614e-06,
1717
+ "loss": 1.9727,
1718
+ "step": 244
1719
+ },
1720
+ {
1721
+ "epoch": 0.8032786885245902,
1722
+ "grad_norm": 10.282430648803711,
1723
+ "learning_rate": 8.606798523148648e-06,
1724
+ "loss": 1.9414,
1725
+ "step": 245
1726
+ },
1727
+ {
1728
+ "epoch": 0.8065573770491803,
1729
+ "grad_norm": 12.212276458740234,
1730
+ "learning_rate": 8.594511307882945e-06,
1731
+ "loss": 1.8496,
1732
+ "step": 246
1733
+ },
1734
+ {
1735
+ "epoch": 0.8098360655737705,
1736
+ "grad_norm": 14.458395957946777,
1737
+ "learning_rate": 8.582179001405544e-06,
1738
+ "loss": 1.9414,
1739
+ "step": 247
1740
+ },
1741
+ {
1742
+ "epoch": 0.8131147540983606,
1743
+ "grad_norm": 10.135568618774414,
1744
+ "learning_rate": 8.569801758418596e-06,
1745
+ "loss": 1.8691,
1746
+ "step": 248
1747
+ },
1748
+ {
1749
+ "epoch": 0.8163934426229508,
1750
+ "grad_norm": 14.193105697631836,
1751
+ "learning_rate": 8.55737973418796e-06,
1752
+ "loss": 1.834,
1753
+ "step": 249
1754
+ },
1755
+ {
1756
+ "epoch": 0.819672131147541,
1757
+ "grad_norm": 10.561907768249512,
1758
+ "learning_rate": 8.544913084541253e-06,
1759
+ "loss": 2.043,
1760
+ "step": 250
1761
+ },
1762
+ {
1763
+ "epoch": 0.8229508196721311,
1764
+ "grad_norm": 12.753805160522461,
1765
+ "learning_rate": 8.532401965865889e-06,
1766
+ "loss": 1.9141,
1767
+ "step": 251
1768
+ },
1769
+ {
1770
+ "epoch": 0.8262295081967214,
1771
+ "grad_norm": 11.008598327636719,
1772
+ "learning_rate": 8.519846535107129e-06,
1773
+ "loss": 1.9258,
1774
+ "step": 252
1775
+ },
1776
+ {
1777
+ "epoch": 0.8295081967213115,
1778
+ "grad_norm": 10.186516761779785,
1779
+ "learning_rate": 8.507246949766095e-06,
1780
+ "loss": 1.9062,
1781
+ "step": 253
1782
+ },
1783
+ {
1784
+ "epoch": 0.8327868852459016,
1785
+ "grad_norm": 12.843867301940918,
1786
+ "learning_rate": 8.494603367897817e-06,
1787
+ "loss": 1.8516,
1788
+ "step": 254
1789
+ },
1790
+ {
1791
+ "epoch": 0.8360655737704918,
1792
+ "grad_norm": 11.588765144348145,
1793
+ "learning_rate": 8.481915948109226e-06,
1794
+ "loss": 1.9121,
1795
+ "step": 255
1796
+ },
1797
+ {
1798
+ "epoch": 0.839344262295082,
1799
+ "grad_norm": 12.169210433959961,
1800
+ "learning_rate": 8.469184849557188e-06,
1801
+ "loss": 1.9238,
1802
+ "step": 256
1803
+ },
1804
+ {
1805
+ "epoch": 0.8426229508196721,
1806
+ "grad_norm": 17.143991470336914,
1807
+ "learning_rate": 8.456410231946484e-06,
1808
+ "loss": 1.8887,
1809
+ "step": 257
1810
+ },
1811
+ {
1812
+ "epoch": 0.8459016393442623,
1813
+ "grad_norm": 9.951960563659668,
1814
+ "learning_rate": 8.443592255527824e-06,
1815
+ "loss": 1.9355,
1816
+ "step": 258
1817
+ },
1818
+ {
1819
+ "epoch": 0.8491803278688524,
1820
+ "grad_norm": 11.594623565673828,
1821
+ "learning_rate": 8.430731081095832e-06,
1822
+ "loss": 1.9512,
1823
+ "step": 259
1824
+ },
1825
+ {
1826
+ "epoch": 0.8524590163934426,
1827
+ "grad_norm": 10.37341022491455,
1828
+ "learning_rate": 8.417826869987026e-06,
1829
+ "loss": 1.9883,
1830
+ "step": 260
1831
+ },
1832
+ {
1833
+ "epoch": 0.8557377049180328,
1834
+ "grad_norm": 13.815895080566406,
1835
+ "learning_rate": 8.404879784077793e-06,
1836
+ "loss": 1.9746,
1837
+ "step": 261
1838
+ },
1839
+ {
1840
+ "epoch": 0.8590163934426229,
1841
+ "grad_norm": 10.756362915039062,
1842
+ "learning_rate": 8.39188998578237e-06,
1843
+ "loss": 1.9023,
1844
+ "step": 262
1845
+ },
1846
+ {
1847
+ "epoch": 0.8622950819672132,
1848
+ "grad_norm": 11.125969886779785,
1849
+ "learning_rate": 8.37885763805079e-06,
1850
+ "loss": 1.873,
1851
+ "step": 263
1852
+ },
1853
+ {
1854
+ "epoch": 0.8655737704918033,
1855
+ "grad_norm": 10.486440658569336,
1856
+ "learning_rate": 8.365782904366845e-06,
1857
+ "loss": 1.8789,
1858
+ "step": 264
1859
+ },
1860
+ {
1861
+ "epoch": 0.8688524590163934,
1862
+ "grad_norm": 10.452739715576172,
1863
+ "learning_rate": 8.352665948746044e-06,
1864
+ "loss": 1.9453,
1865
+ "step": 265
1866
+ },
1867
+ {
1868
+ "epoch": 0.8721311475409836,
1869
+ "grad_norm": 10.094730377197266,
1870
+ "learning_rate": 8.33950693573354e-06,
1871
+ "loss": 1.8516,
1872
+ "step": 266
1873
+ },
1874
+ {
1875
+ "epoch": 0.8754098360655738,
1876
+ "grad_norm": 11.559701919555664,
1877
+ "learning_rate": 8.326306030402076e-06,
1878
+ "loss": 1.9844,
1879
+ "step": 267
1880
+ },
1881
+ {
1882
+ "epoch": 0.8786885245901639,
1883
+ "grad_norm": 11.006271362304688,
1884
+ "learning_rate": 8.313063398349914e-06,
1885
+ "loss": 1.8887,
1886
+ "step": 268
1887
+ },
1888
+ {
1889
+ "epoch": 0.8819672131147541,
1890
+ "grad_norm": 12.529583930969238,
1891
+ "learning_rate": 8.29977920569875e-06,
1892
+ "loss": 1.877,
1893
+ "step": 269
1894
+ },
1895
+ {
1896
+ "epoch": 0.8852459016393442,
1897
+ "grad_norm": 10.77204704284668,
1898
+ "learning_rate": 8.28645361909164e-06,
1899
+ "loss": 1.998,
1900
+ "step": 270
1901
+ },
1902
+ {
1903
+ "epoch": 0.8885245901639345,
1904
+ "grad_norm": 13.109503746032715,
1905
+ "learning_rate": 8.273086805690903e-06,
1906
+ "loss": 1.8711,
1907
+ "step": 271
1908
+ },
1909
+ {
1910
+ "epoch": 0.8918032786885246,
1911
+ "grad_norm": 15.495031356811523,
1912
+ "learning_rate": 8.259678933176025e-06,
1913
+ "loss": 1.916,
1914
+ "step": 272
1915
+ },
1916
+ {
1917
+ "epoch": 0.8950819672131147,
1918
+ "grad_norm": 11.081769943237305,
1919
+ "learning_rate": 8.24623016974156e-06,
1920
+ "loss": 1.8809,
1921
+ "step": 273
1922
+ },
1923
+ {
1924
+ "epoch": 0.898360655737705,
1925
+ "grad_norm": 10.445962905883789,
1926
+ "learning_rate": 8.232740684095012e-06,
1927
+ "loss": 1.8301,
1928
+ "step": 274
1929
+ },
1930
+ {
1931
+ "epoch": 0.9016393442622951,
1932
+ "grad_norm": 11.152101516723633,
1933
+ "learning_rate": 8.219210645454727e-06,
1934
+ "loss": 1.9258,
1935
+ "step": 275
1936
+ },
1937
+ {
1938
+ "epoch": 0.9049180327868852,
1939
+ "grad_norm": 10.69294261932373,
1940
+ "learning_rate": 8.205640223547766e-06,
1941
+ "loss": 1.8242,
1942
+ "step": 276
1943
+ },
1944
+ {
1945
+ "epoch": 0.9081967213114754,
1946
+ "grad_norm": 13.176959991455078,
1947
+ "learning_rate": 8.192029588607772e-06,
1948
+ "loss": 1.9453,
1949
+ "step": 277
1950
+ },
1951
+ {
1952
+ "epoch": 0.9114754098360656,
1953
+ "grad_norm": 12.394086837768555,
1954
+ "learning_rate": 8.178378911372846e-06,
1955
+ "loss": 1.9434,
1956
+ "step": 278
1957
+ },
1958
+ {
1959
+ "epoch": 0.9147540983606557,
1960
+ "grad_norm": 13.449362754821777,
1961
+ "learning_rate": 8.16468836308339e-06,
1962
+ "loss": 1.9961,
1963
+ "step": 279
1964
+ },
1965
+ {
1966
+ "epoch": 0.9180327868852459,
1967
+ "grad_norm": 10.922250747680664,
1968
+ "learning_rate": 8.150958115479976e-06,
1969
+ "loss": 1.9531,
1970
+ "step": 280
1971
+ },
1972
+ {
1973
+ "epoch": 0.921311475409836,
1974
+ "grad_norm": 14.1903657913208,
1975
+ "learning_rate": 8.137188340801177e-06,
1976
+ "loss": 1.9336,
1977
+ "step": 281
1978
+ },
1979
+ {
1980
+ "epoch": 0.9245901639344263,
1981
+ "grad_norm": 13.266737937927246,
1982
+ "learning_rate": 8.12337921178141e-06,
1983
+ "loss": 1.9688,
1984
+ "step": 282
1985
+ },
1986
+ {
1987
+ "epoch": 0.9278688524590164,
1988
+ "grad_norm": 12.7962064743042,
1989
+ "learning_rate": 8.109530901648776e-06,
1990
+ "loss": 1.875,
1991
+ "step": 283
1992
+ },
1993
+ {
1994
+ "epoch": 0.9311475409836065,
1995
+ "grad_norm": 12.458824157714844,
1996
+ "learning_rate": 8.095643584122879e-06,
1997
+ "loss": 1.9141,
1998
+ "step": 284
1999
+ },
2000
+ {
2001
+ "epoch": 0.9344262295081968,
2002
+ "grad_norm": 10.26793384552002,
2003
+ "learning_rate": 8.08171743341265e-06,
2004
+ "loss": 1.8633,
2005
+ "step": 285
2006
+ },
2007
+ {
2008
+ "epoch": 0.9377049180327869,
2009
+ "grad_norm": 12.272446632385254,
2010
+ "learning_rate": 8.067752624214164e-06,
2011
+ "loss": 1.8945,
2012
+ "step": 286
2013
+ },
2014
+ {
2015
+ "epoch": 0.940983606557377,
2016
+ "grad_norm": 11.974457740783691,
2017
+ "learning_rate": 8.053749331708444e-06,
2018
+ "loss": 1.877,
2019
+ "step": 287
2020
+ },
2021
+ {
2022
+ "epoch": 0.9442622950819672,
2023
+ "grad_norm": 10.955528259277344,
2024
+ "learning_rate": 8.039707731559262e-06,
2025
+ "loss": 1.8867,
2026
+ "step": 288
2027
+ },
2028
+ {
2029
+ "epoch": 0.9475409836065574,
2030
+ "grad_norm": 12.32547378540039,
2031
+ "learning_rate": 8.025627999910946e-06,
2032
+ "loss": 1.9629,
2033
+ "step": 289
2034
+ },
2035
+ {
2036
+ "epoch": 0.9508196721311475,
2037
+ "grad_norm": 12.567078590393066,
2038
+ "learning_rate": 8.011510313386156e-06,
2039
+ "loss": 1.877,
2040
+ "step": 290
2041
+ },
2042
+ {
2043
+ "epoch": 0.9540983606557377,
2044
+ "grad_norm": 10.757590293884277,
2045
+ "learning_rate": 7.997354849083682e-06,
2046
+ "loss": 1.8926,
2047
+ "step": 291
2048
+ },
2049
+ {
2050
+ "epoch": 0.9573770491803278,
2051
+ "grad_norm": 12.629182815551758,
2052
+ "learning_rate": 7.98316178457621e-06,
2053
+ "loss": 1.9199,
2054
+ "step": 292
2055
+ },
2056
+ {
2057
+ "epoch": 0.9606557377049181,
2058
+ "grad_norm": 11.572930335998535,
2059
+ "learning_rate": 7.968931297908105e-06,
2060
+ "loss": 1.9492,
2061
+ "step": 293
2062
+ },
2063
+ {
2064
+ "epoch": 0.9639344262295082,
2065
+ "grad_norm": 13.813660621643066,
2066
+ "learning_rate": 7.954663567593169e-06,
2067
+ "loss": 1.793,
2068
+ "step": 294
2069
+ },
2070
+ {
2071
+ "epoch": 0.9672131147540983,
2072
+ "grad_norm": 10.30826473236084,
2073
+ "learning_rate": 7.940358772612408e-06,
2074
+ "loss": 1.8809,
2075
+ "step": 295
2076
+ },
2077
+ {
2078
+ "epoch": 0.9704918032786886,
2079
+ "grad_norm": 11.040604591369629,
2080
+ "learning_rate": 7.926017092411786e-06,
2081
+ "loss": 1.8438,
2082
+ "step": 296
2083
+ },
2084
+ {
2085
+ "epoch": 0.9737704918032787,
2086
+ "grad_norm": 11.625853538513184,
2087
+ "learning_rate": 7.911638706899968e-06,
2088
+ "loss": 1.8652,
2089
+ "step": 297
2090
+ },
2091
+ {
2092
+ "epoch": 0.9770491803278688,
2093
+ "grad_norm": 12.791728973388672,
2094
+ "learning_rate": 7.89722379644607e-06,
2095
+ "loss": 1.8555,
2096
+ "step": 298
2097
+ },
2098
+ {
2099
+ "epoch": 0.980327868852459,
2100
+ "grad_norm": 15.870569229125977,
2101
+ "learning_rate": 7.882772541877392e-06,
2102
+ "loss": 1.9043,
2103
+ "step": 299
2104
+ },
2105
+ {
2106
+ "epoch": 0.9836065573770492,
2107
+ "grad_norm": 11.703765869140625,
2108
+ "learning_rate": 7.868285124477156e-06,
2109
+ "loss": 1.8613,
2110
+ "step": 300
2111
+ }
2112
+ ],
2113
+ "logging_steps": 1,
2114
+ "max_steps": 915,
2115
+ "num_input_tokens_seen": 0,
2116
+ "num_train_epochs": 3,
2117
+ "save_steps": 100,
2118
+ "stateful_callbacks": {
2119
+ "TrainerControl": {
2120
+ "args": {
2121
+ "should_epoch_stop": false,
2122
+ "should_evaluate": false,
2123
+ "should_log": false,
2124
+ "should_save": true,
2125
+ "should_training_stop": false
2126
+ },
2127
+ "attributes": {}
2128
+ }
2129
+ },
2130
+ "total_flos": 3.5197453267369984e+16,
2131
+ "train_batch_size": 512,
2132
+ "trial_name": null,
2133
+ "trial_params": null
2134
+ }
checkpoint-300/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-400/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CLIPModel"
4
+ ],
5
+ "bos_token_id": 49406,
6
+ "eos_token_id": 49407,
7
+ "initializer_factor": 1.0,
8
+ "logit_scale_init_value": 2.6592,
9
+ "model_type": "clip",
10
+ "pad_token_id": 49407,
11
+ "projection_dim": 512,
12
+ "text_config": {
13
+ "attention_dropout": 0.0,
14
+ "bos_token_id": 0,
15
+ "dropout": 0.0,
16
+ "eos_token_id": 2,
17
+ "hidden_act": "quick_gelu",
18
+ "hidden_size": 512,
19
+ "initializer_factor": 1.0,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 2048,
22
+ "layer_norm_eps": 1e-05,
23
+ "max_position_embeddings": 77,
24
+ "model_type": "clip_text_model",
25
+ "num_attention_heads": 8,
26
+ "num_hidden_layers": 12,
27
+ "projection_dim": 512,
28
+ "torch_dtype": "float32",
29
+ "vocab_size": 49408
30
+ },
31
+ "torch_dtype": "bfloat16",
32
+ "transformers_version": "4.56.0.dev0",
33
+ "vision_config": {
34
+ "attention_dropout": 0.0,
35
+ "dropout": 0.0,
36
+ "hidden_act": "quick_gelu",
37
+ "hidden_size": 768,
38
+ "image_size": 224,
39
+ "initializer_factor": 1.0,
40
+ "initializer_range": 0.02,
41
+ "intermediate_size": 3072,
42
+ "layer_norm_eps": 1e-05,
43
+ "model_type": "clip_vision_model",
44
+ "num_attention_heads": 12,
45
+ "num_channels": 3,
46
+ "num_hidden_layers": 12,
47
+ "patch_size": 16,
48
+ "projection_dim": 512,
49
+ "torch_dtype": "float32"
50
+ }
51
+ }
checkpoint-400/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-400/preprocessor_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "processor_class": "CLIPProcessor",
23
+ "resample": 3,
24
+ "rescale_factor": 0.00392156862745098,
25
+ "size": {
26
+ "shortest_edge": 224
27
+ }
28
+ }
checkpoint-400/special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|startoftext|>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<|endoftext|>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<|endoftext|>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
checkpoint-400/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-400/tokenizer_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "49406": {
5
+ "content": "<|startoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "49407": {
13
+ "content": "<|endoftext|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ }
20
+ },
21
+ "bos_token": "<|startoftext|>",
22
+ "clean_up_tokenization_spaces": false,
23
+ "do_lower_case": true,
24
+ "eos_token": "<|endoftext|>",
25
+ "errors": "replace",
26
+ "extra_special_tokens": {},
27
+ "model_max_length": 77,
28
+ "pad_token": "<|endoftext|>",
29
+ "processor_class": "CLIPProcessor",
30
+ "tokenizer_class": "CLIPTokenizer",
31
+ "unk_token": "<|endoftext|>"
32
+ }
checkpoint-400/trainer_state.json ADDED
@@ -0,0 +1,2834 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.3114754098360657,
6
+ "eval_steps": 500,
7
+ "global_step": 400,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.003278688524590164,
14
+ "grad_norm": 58.999969482421875,
15
+ "learning_rate": 0.0,
16
+ "loss": 3.8125,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.006557377049180328,
21
+ "grad_norm": 56.5999870300293,
22
+ "learning_rate": 3.5714285714285716e-07,
23
+ "loss": 3.8984,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.009836065573770493,
28
+ "grad_norm": 58.442317962646484,
29
+ "learning_rate": 7.142857142857143e-07,
30
+ "loss": 3.8359,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.013114754098360656,
35
+ "grad_norm": 59.13441467285156,
36
+ "learning_rate": 1.0714285714285714e-06,
37
+ "loss": 3.7422,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.01639344262295082,
42
+ "grad_norm": 59.55280685424805,
43
+ "learning_rate": 1.4285714285714286e-06,
44
+ "loss": 3.7383,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.019672131147540985,
49
+ "grad_norm": 57.113956451416016,
50
+ "learning_rate": 1.7857142857142859e-06,
51
+ "loss": 3.8281,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.022950819672131147,
56
+ "grad_norm": 44.69753646850586,
57
+ "learning_rate": 2.1428571428571427e-06,
58
+ "loss": 3.6562,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.02622950819672131,
63
+ "grad_norm": 42.74599075317383,
64
+ "learning_rate": 2.5e-06,
65
+ "loss": 3.668,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.029508196721311476,
70
+ "grad_norm": 27.557815551757812,
71
+ "learning_rate": 2.8571428571428573e-06,
72
+ "loss": 3.3438,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.03278688524590164,
77
+ "grad_norm": 25.87570571899414,
78
+ "learning_rate": 3.2142857142857147e-06,
79
+ "loss": 3.3633,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.036065573770491806,
84
+ "grad_norm": 25.204158782958984,
85
+ "learning_rate": 3.5714285714285718e-06,
86
+ "loss": 3.2852,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.03934426229508197,
91
+ "grad_norm": 25.082277297973633,
92
+ "learning_rate": 3.928571428571429e-06,
93
+ "loss": 3.2188,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.04262295081967213,
98
+ "grad_norm": 21.52399253845215,
99
+ "learning_rate": 4.2857142857142855e-06,
100
+ "loss": 3.1016,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.04590163934426229,
105
+ "grad_norm": 21.302213668823242,
106
+ "learning_rate": 4.642857142857144e-06,
107
+ "loss": 3.0938,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.04918032786885246,
112
+ "grad_norm": 17.48920249938965,
113
+ "learning_rate": 5e-06,
114
+ "loss": 3.0703,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.05245901639344262,
119
+ "grad_norm": 14.723689079284668,
120
+ "learning_rate": 5.357142857142857e-06,
121
+ "loss": 2.7812,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.05573770491803279,
126
+ "grad_norm": 13.425204277038574,
127
+ "learning_rate": 5.7142857142857145e-06,
128
+ "loss": 2.8398,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.05901639344262295,
133
+ "grad_norm": 12.121846199035645,
134
+ "learning_rate": 6.071428571428571e-06,
135
+ "loss": 2.7539,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.06229508196721312,
140
+ "grad_norm": 12.97847843170166,
141
+ "learning_rate": 6.4285714285714295e-06,
142
+ "loss": 2.6719,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.06557377049180328,
147
+ "grad_norm": 12.588919639587402,
148
+ "learning_rate": 6.785714285714287e-06,
149
+ "loss": 2.6758,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.06885245901639345,
154
+ "grad_norm": 10.492659568786621,
155
+ "learning_rate": 7.1428571428571436e-06,
156
+ "loss": 2.6562,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.07213114754098361,
161
+ "grad_norm": 9.676924705505371,
162
+ "learning_rate": 7.500000000000001e-06,
163
+ "loss": 2.6094,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.07540983606557378,
168
+ "grad_norm": 10.222829818725586,
169
+ "learning_rate": 7.857142857142858e-06,
170
+ "loss": 2.6211,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.07868852459016394,
175
+ "grad_norm": 10.020594596862793,
176
+ "learning_rate": 8.214285714285714e-06,
177
+ "loss": 2.5273,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.08196721311475409,
182
+ "grad_norm": 9.404995918273926,
183
+ "learning_rate": 8.571428571428571e-06,
184
+ "loss": 2.5352,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.08524590163934426,
189
+ "grad_norm": 10.194731712341309,
190
+ "learning_rate": 8.92857142857143e-06,
191
+ "loss": 2.543,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.08852459016393442,
196
+ "grad_norm": 9.045639991760254,
197
+ "learning_rate": 9.285714285714288e-06,
198
+ "loss": 2.4961,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.09180327868852459,
203
+ "grad_norm": 9.108716011047363,
204
+ "learning_rate": 9.642857142857144e-06,
205
+ "loss": 2.4336,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.09508196721311475,
210
+ "grad_norm": 9.336297988891602,
211
+ "learning_rate": 1e-05,
212
+ "loss": 2.457,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.09836065573770492,
217
+ "grad_norm": 8.170738220214844,
218
+ "learning_rate": 9.999968638844173e-06,
219
+ "loss": 2.4688,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.10163934426229508,
224
+ "grad_norm": 8.507989883422852,
225
+ "learning_rate": 9.999874555770099e-06,
226
+ "loss": 2.4062,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.10491803278688525,
231
+ "grad_norm": 8.38835334777832,
232
+ "learning_rate": 9.999717751958002e-06,
233
+ "loss": 2.3711,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.10819672131147541,
238
+ "grad_norm": 8.36460018157959,
239
+ "learning_rate": 9.9994982293749e-06,
240
+ "loss": 2.4805,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.11147540983606558,
245
+ "grad_norm": 8.547989845275879,
246
+ "learning_rate": 9.999215990774584e-06,
247
+ "loss": 2.3906,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.11475409836065574,
252
+ "grad_norm": 8.938812255859375,
253
+ "learning_rate": 9.998871039697589e-06,
254
+ "loss": 2.3906,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.1180327868852459,
259
+ "grad_norm": 8.230469703674316,
260
+ "learning_rate": 9.998463380471138e-06,
261
+ "loss": 2.3164,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.12131147540983607,
266
+ "grad_norm": 7.883755683898926,
267
+ "learning_rate": 9.997993018209098e-06,
268
+ "loss": 2.3633,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.12459016393442623,
273
+ "grad_norm": 8.220748901367188,
274
+ "learning_rate": 9.997459958811911e-06,
275
+ "loss": 2.3516,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.12786885245901639,
280
+ "grad_norm": 8.738327980041504,
281
+ "learning_rate": 9.996864208966522e-06,
282
+ "loss": 2.3398,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.13114754098360656,
287
+ "grad_norm": 7.944454669952393,
288
+ "learning_rate": 9.996205776146288e-06,
289
+ "loss": 2.3867,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.13442622950819672,
294
+ "grad_norm": 7.631208419799805,
295
+ "learning_rate": 9.995484668610897e-06,
296
+ "loss": 2.3242,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.1377049180327869,
301
+ "grad_norm": 9.301526069641113,
302
+ "learning_rate": 9.994700895406258e-06,
303
+ "loss": 2.3203,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.14098360655737704,
308
+ "grad_norm": 7.625834941864014,
309
+ "learning_rate": 9.99385446636438e-06,
310
+ "loss": 2.3398,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.14426229508196722,
315
+ "grad_norm": 7.908705234527588,
316
+ "learning_rate": 9.992945392103264e-06,
317
+ "loss": 2.3477,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.14754098360655737,
322
+ "grad_norm": 8.002193450927734,
323
+ "learning_rate": 9.991973684026755e-06,
324
+ "loss": 2.2852,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.15081967213114755,
329
+ "grad_norm": 8.058524131774902,
330
+ "learning_rate": 9.99093935432441e-06,
331
+ "loss": 2.3711,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.1540983606557377,
336
+ "grad_norm": 8.803773880004883,
337
+ "learning_rate": 9.989842415971338e-06,
338
+ "loss": 2.2344,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.15737704918032788,
343
+ "grad_norm": 7.851852893829346,
344
+ "learning_rate": 9.988682882728043e-06,
345
+ "loss": 2.3047,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.16065573770491803,
350
+ "grad_norm": 8.29660415649414,
351
+ "learning_rate": 9.987460769140242e-06,
352
+ "loss": 2.3047,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.16393442622950818,
357
+ "grad_norm": 7.779457092285156,
358
+ "learning_rate": 9.986176090538697e-06,
359
+ "loss": 2.1992,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.16721311475409836,
364
+ "grad_norm": 8.602660179138184,
365
+ "learning_rate": 9.984828863039008e-06,
366
+ "loss": 2.25,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.17049180327868851,
371
+ "grad_norm": 8.079516410827637,
372
+ "learning_rate": 9.983419103541421e-06,
373
+ "loss": 2.3203,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.1737704918032787,
378
+ "grad_norm": 8.443635940551758,
379
+ "learning_rate": 9.981946829730611e-06,
380
+ "loss": 2.3242,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.17704918032786884,
385
+ "grad_norm": 7.873276710510254,
386
+ "learning_rate": 9.980412060075459e-06,
387
+ "loss": 2.2656,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.18032786885245902,
392
+ "grad_norm": 8.633429527282715,
393
+ "learning_rate": 9.978814813828827e-06,
394
+ "loss": 2.2227,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.18360655737704917,
399
+ "grad_norm": 9.135279655456543,
400
+ "learning_rate": 9.97715511102731e-06,
401
+ "loss": 2.2148,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.18688524590163935,
406
+ "grad_norm": 7.584745407104492,
407
+ "learning_rate": 9.975432972490985e-06,
408
+ "loss": 2.2773,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.1901639344262295,
413
+ "grad_norm": 9.086216926574707,
414
+ "learning_rate": 9.973648419823161e-06,
415
+ "loss": 2.2656,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.19344262295081968,
420
+ "grad_norm": 8.223559379577637,
421
+ "learning_rate": 9.971801475410084e-06,
422
+ "loss": 2.2773,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.19672131147540983,
427
+ "grad_norm": 8.726387023925781,
428
+ "learning_rate": 9.969892162420682e-06,
429
+ "loss": 2.3125,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.2,
434
+ "grad_norm": 8.304938316345215,
435
+ "learning_rate": 9.96792050480626e-06,
436
+ "loss": 2.168,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.20327868852459016,
441
+ "grad_norm": 8.20266342163086,
442
+ "learning_rate": 9.965886527300201e-06,
443
+ "loss": 2.25,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.20655737704918034,
448
+ "grad_norm": 7.833676815032959,
449
+ "learning_rate": 9.963790255417663e-06,
450
+ "loss": 2.2422,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.2098360655737705,
455
+ "grad_norm": 7.891449928283691,
456
+ "learning_rate": 9.961631715455245e-06,
457
+ "loss": 2.1758,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.21311475409836064,
462
+ "grad_norm": 8.999512672424316,
463
+ "learning_rate": 9.959410934490673e-06,
464
+ "loss": 2.1484,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.21639344262295082,
469
+ "grad_norm": 7.718410015106201,
470
+ "learning_rate": 9.95712794038245e-06,
471
+ "loss": 2.1172,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.21967213114754097,
476
+ "grad_norm": 8.272541999816895,
477
+ "learning_rate": 9.954782761769509e-06,
478
+ "loss": 2.25,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.22295081967213115,
483
+ "grad_norm": 9.607975006103516,
484
+ "learning_rate": 9.952375428070853e-06,
485
+ "loss": 2.2812,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.2262295081967213,
490
+ "grad_norm": 8.297865867614746,
491
+ "learning_rate": 9.949905969485192e-06,
492
+ "loss": 2.1406,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.22950819672131148,
497
+ "grad_norm": 8.192915916442871,
498
+ "learning_rate": 9.947374416990554e-06,
499
+ "loss": 2.2109,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.23278688524590163,
504
+ "grad_norm": 8.489333152770996,
505
+ "learning_rate": 9.944780802343906e-06,
506
+ "loss": 2.1914,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.2360655737704918,
511
+ "grad_norm": 8.244175910949707,
512
+ "learning_rate": 9.942125158080747e-06,
513
+ "loss": 2.1797,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.23934426229508196,
518
+ "grad_norm": 8.009477615356445,
519
+ "learning_rate": 9.939407517514709e-06,
520
+ "loss": 2.2578,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.24262295081967214,
525
+ "grad_norm": 8.284704208374023,
526
+ "learning_rate": 9.936627914737129e-06,
527
+ "loss": 2.2305,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.2459016393442623,
532
+ "grad_norm": 9.479103088378906,
533
+ "learning_rate": 9.933786384616631e-06,
534
+ "loss": 2.1406,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.24918032786885247,
539
+ "grad_norm": 8.558723449707031,
540
+ "learning_rate": 9.930882962798683e-06,
541
+ "loss": 2.2812,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.25245901639344265,
546
+ "grad_norm": 8.608718872070312,
547
+ "learning_rate": 9.927917685705148e-06,
548
+ "loss": 2.2461,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.25573770491803277,
553
+ "grad_norm": 7.863804817199707,
554
+ "learning_rate": 9.924890590533837e-06,
555
+ "loss": 2.1914,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.25901639344262295,
560
+ "grad_norm": 8.842366218566895,
561
+ "learning_rate": 9.921801715258027e-06,
562
+ "loss": 2.0996,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.26229508196721313,
567
+ "grad_norm": 8.273311614990234,
568
+ "learning_rate": 9.918651098626e-06,
569
+ "loss": 2.1641,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.26557377049180325,
574
+ "grad_norm": 8.011392593383789,
575
+ "learning_rate": 9.915438780160547e-06,
576
+ "loss": 2.168,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.26885245901639343,
581
+ "grad_norm": 7.946706771850586,
582
+ "learning_rate": 9.912164800158474e-06,
583
+ "loss": 2.2227,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.2721311475409836,
588
+ "grad_norm": 9.032291412353516,
589
+ "learning_rate": 9.908829199690106e-06,
590
+ "loss": 2.2891,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.2754098360655738,
595
+ "grad_norm": 8.33333969116211,
596
+ "learning_rate": 9.905432020598751e-06,
597
+ "loss": 2.2344,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.2786885245901639,
602
+ "grad_norm": 7.967901229858398,
603
+ "learning_rate": 9.901973305500197e-06,
604
+ "loss": 2.1172,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.2819672131147541,
609
+ "grad_norm": 8.24918270111084,
610
+ "learning_rate": 9.898453097782164e-06,
611
+ "loss": 2.1602,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.28524590163934427,
616
+ "grad_norm": 8.877111434936523,
617
+ "learning_rate": 9.894871441603766e-06,
618
+ "loss": 2.2383,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.28852459016393445,
623
+ "grad_norm": 8.32304573059082,
624
+ "learning_rate": 9.891228381894954e-06,
625
+ "loss": 2.1719,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.29180327868852457,
630
+ "grad_norm": 9.091806411743164,
631
+ "learning_rate": 9.887523964355953e-06,
632
+ "loss": 2.168,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.29508196721311475,
637
+ "grad_norm": 8.54736614227295,
638
+ "learning_rate": 9.88375823545669e-06,
639
+ "loss": 2.1211,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.2983606557377049,
644
+ "grad_norm": 9.3049955368042,
645
+ "learning_rate": 9.879931242436208e-06,
646
+ "loss": 2.1719,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.3016393442622951,
651
+ "grad_norm": 8.523479461669922,
652
+ "learning_rate": 9.876043033302079e-06,
653
+ "loss": 2.1719,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.30491803278688523,
658
+ "grad_norm": 7.741846561431885,
659
+ "learning_rate": 9.872093656829792e-06,
660
+ "loss": 2.0391,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.3081967213114754,
665
+ "grad_norm": 8.36119270324707,
666
+ "learning_rate": 9.868083162562155e-06,
667
+ "loss": 2.0938,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.3114754098360656,
672
+ "grad_norm": 8.163023948669434,
673
+ "learning_rate": 9.864011600808663e-06,
674
+ "loss": 2.0938,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.31475409836065577,
679
+ "grad_norm": 8.827863693237305,
680
+ "learning_rate": 9.859879022644865e-06,
681
+ "loss": 2.1406,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.3180327868852459,
686
+ "grad_norm": 8.434041023254395,
687
+ "learning_rate": 9.855685479911736e-06,
688
+ "loss": 2.168,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.32131147540983607,
693
+ "grad_norm": 8.144036293029785,
694
+ "learning_rate": 9.851431025215012e-06,
695
+ "loss": 2.0742,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.32459016393442625,
700
+ "grad_norm": 8.843683242797852,
701
+ "learning_rate": 9.847115711924542e-06,
702
+ "loss": 2.1016,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.32786885245901637,
707
+ "grad_norm": 8.300407409667969,
708
+ "learning_rate": 9.842739594173607e-06,
709
+ "loss": 2.1953,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 0.33114754098360655,
714
+ "grad_norm": 8.825713157653809,
715
+ "learning_rate": 9.838302726858258e-06,
716
+ "loss": 2.1406,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 0.3344262295081967,
721
+ "grad_norm": 8.763128280639648,
722
+ "learning_rate": 9.833805165636603e-06,
723
+ "loss": 2.1523,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 0.3377049180327869,
728
+ "grad_norm": 8.53940486907959,
729
+ "learning_rate": 9.829246966928135e-06,
730
+ "loss": 2.0703,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 0.34098360655737703,
735
+ "grad_norm": 8.830872535705566,
736
+ "learning_rate": 9.824628187913001e-06,
737
+ "loss": 2.1602,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 0.3442622950819672,
742
+ "grad_norm": 8.062509536743164,
743
+ "learning_rate": 9.819948886531305e-06,
744
+ "loss": 2.1211,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 0.3475409836065574,
749
+ "grad_norm": 8.8229398727417,
750
+ "learning_rate": 9.815209121482363e-06,
751
+ "loss": 2.1172,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 0.35081967213114756,
756
+ "grad_norm": 9.78320598602295,
757
+ "learning_rate": 9.810408952223982e-06,
758
+ "loss": 2.0586,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 0.3540983606557377,
763
+ "grad_norm": 8.267208099365234,
764
+ "learning_rate": 9.805548438971702e-06,
765
+ "loss": 2.1406,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 0.35737704918032787,
770
+ "grad_norm": 10.9806489944458,
771
+ "learning_rate": 9.80062764269805e-06,
772
+ "loss": 2.1035,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 0.36065573770491804,
777
+ "grad_norm": 9.699831008911133,
778
+ "learning_rate": 9.795646625131771e-06,
779
+ "loss": 2.1797,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 0.3639344262295082,
784
+ "grad_norm": 10.373594284057617,
785
+ "learning_rate": 9.790605448757049e-06,
786
+ "loss": 2.1445,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 0.36721311475409835,
791
+ "grad_norm": 8.51201343536377,
792
+ "learning_rate": 9.785504176812733e-06,
793
+ "loss": 2.1875,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 0.3704918032786885,
798
+ "grad_norm": 10.090012550354004,
799
+ "learning_rate": 9.780342873291535e-06,
800
+ "loss": 2.2109,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 0.3737704918032787,
805
+ "grad_norm": 8.939981460571289,
806
+ "learning_rate": 9.775121602939234e-06,
807
+ "loss": 2.0273,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 0.3770491803278688,
812
+ "grad_norm": 8.556815147399902,
813
+ "learning_rate": 9.76984043125386e-06,
814
+ "loss": 2.002,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 0.380327868852459,
819
+ "grad_norm": 9.271822929382324,
820
+ "learning_rate": 9.764499424484873e-06,
821
+ "loss": 2.0742,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 0.3836065573770492,
826
+ "grad_norm": 9.027301788330078,
827
+ "learning_rate": 9.759098649632326e-06,
828
+ "loss": 2.1133,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 0.38688524590163936,
833
+ "grad_norm": 9.81689453125,
834
+ "learning_rate": 9.753638174446042e-06,
835
+ "loss": 2.0645,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 0.3901639344262295,
840
+ "grad_norm": 9.42857551574707,
841
+ "learning_rate": 9.748118067424742e-06,
842
+ "loss": 2.1172,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 0.39344262295081966,
847
+ "grad_norm": 8.636916160583496,
848
+ "learning_rate": 9.742538397815203e-06,
849
+ "loss": 2.0859,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 0.39672131147540984,
854
+ "grad_norm": 9.01211166381836,
855
+ "learning_rate": 9.73689923561138e-06,
856
+ "loss": 2.0605,
857
+ "step": 121
858
+ },
859
+ {
860
+ "epoch": 0.4,
861
+ "grad_norm": 8.320965766906738,
862
+ "learning_rate": 9.73120065155353e-06,
863
+ "loss": 2.0352,
864
+ "step": 122
865
+ },
866
+ {
867
+ "epoch": 0.40327868852459015,
868
+ "grad_norm": 8.969964027404785,
869
+ "learning_rate": 9.725442717127324e-06,
870
+ "loss": 2.1055,
871
+ "step": 123
872
+ },
873
+ {
874
+ "epoch": 0.4065573770491803,
875
+ "grad_norm": 7.825037002563477,
876
+ "learning_rate": 9.719625504562959e-06,
877
+ "loss": 2.1133,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 0.4098360655737705,
882
+ "grad_norm": 9.594252586364746,
883
+ "learning_rate": 9.713749086834234e-06,
884
+ "loss": 2.1367,
885
+ "step": 125
886
+ },
887
+ {
888
+ "epoch": 0.4131147540983607,
889
+ "grad_norm": 9.286437034606934,
890
+ "learning_rate": 9.707813537657652e-06,
891
+ "loss": 2.0293,
892
+ "step": 126
893
+ },
894
+ {
895
+ "epoch": 0.4163934426229508,
896
+ "grad_norm": 11.219290733337402,
897
+ "learning_rate": 9.701818931491485e-06,
898
+ "loss": 2.1992,
899
+ "step": 127
900
+ },
901
+ {
902
+ "epoch": 0.419672131147541,
903
+ "grad_norm": 8.125706672668457,
904
+ "learning_rate": 9.695765343534846e-06,
905
+ "loss": 2.0273,
906
+ "step": 128
907
+ },
908
+ {
909
+ "epoch": 0.42295081967213116,
910
+ "grad_norm": 9.294842720031738,
911
+ "learning_rate": 9.689652849726739e-06,
912
+ "loss": 2.082,
913
+ "step": 129
914
+ },
915
+ {
916
+ "epoch": 0.4262295081967213,
917
+ "grad_norm": 8.297966957092285,
918
+ "learning_rate": 9.683481526745114e-06,
919
+ "loss": 2.084,
920
+ "step": 130
921
+ },
922
+ {
923
+ "epoch": 0.42950819672131146,
924
+ "grad_norm": 9.68375015258789,
925
+ "learning_rate": 9.677251452005897e-06,
926
+ "loss": 2.0996,
927
+ "step": 131
928
+ },
929
+ {
930
+ "epoch": 0.43278688524590164,
931
+ "grad_norm": 8.27182388305664,
932
+ "learning_rate": 9.670962703662032e-06,
933
+ "loss": 2.0312,
934
+ "step": 132
935
+ },
936
+ {
937
+ "epoch": 0.4360655737704918,
938
+ "grad_norm": 9.681833267211914,
939
+ "learning_rate": 9.664615360602478e-06,
940
+ "loss": 2.0547,
941
+ "step": 133
942
+ },
943
+ {
944
+ "epoch": 0.43934426229508194,
945
+ "grad_norm": 8.213553428649902,
946
+ "learning_rate": 9.658209502451245e-06,
947
+ "loss": 1.9746,
948
+ "step": 134
949
+ },
950
+ {
951
+ "epoch": 0.4426229508196721,
952
+ "grad_norm": 9.513401985168457,
953
+ "learning_rate": 9.651745209566379e-06,
954
+ "loss": 2.0625,
955
+ "step": 135
956
+ },
957
+ {
958
+ "epoch": 0.4459016393442623,
959
+ "grad_norm": 9.364225387573242,
960
+ "learning_rate": 9.645222563038959e-06,
961
+ "loss": 2.0078,
962
+ "step": 136
963
+ },
964
+ {
965
+ "epoch": 0.4491803278688525,
966
+ "grad_norm": 8.659600257873535,
967
+ "learning_rate": 9.638641644692077e-06,
968
+ "loss": 2.0469,
969
+ "step": 137
970
+ },
971
+ {
972
+ "epoch": 0.4524590163934426,
973
+ "grad_norm": 9.638226509094238,
974
+ "learning_rate": 9.632002537079816e-06,
975
+ "loss": 2.0391,
976
+ "step": 138
977
+ },
978
+ {
979
+ "epoch": 0.4557377049180328,
980
+ "grad_norm": 9.299017906188965,
981
+ "learning_rate": 9.625305323486214e-06,
982
+ "loss": 2.0508,
983
+ "step": 139
984
+ },
985
+ {
986
+ "epoch": 0.45901639344262296,
987
+ "grad_norm": 9.940561294555664,
988
+ "learning_rate": 9.61855008792421e-06,
989
+ "loss": 2.082,
990
+ "step": 140
991
+ },
992
+ {
993
+ "epoch": 0.46229508196721314,
994
+ "grad_norm": 9.618109703063965,
995
+ "learning_rate": 9.611736915134605e-06,
996
+ "loss": 1.998,
997
+ "step": 141
998
+ },
999
+ {
1000
+ "epoch": 0.46557377049180326,
1001
+ "grad_norm": 8.827125549316406,
1002
+ "learning_rate": 9.604865890584987e-06,
1003
+ "loss": 2.1055,
1004
+ "step": 142
1005
+ },
1006
+ {
1007
+ "epoch": 0.46885245901639344,
1008
+ "grad_norm": 10.079780578613281,
1009
+ "learning_rate": 9.597937100468668e-06,
1010
+ "loss": 2.0859,
1011
+ "step": 143
1012
+ },
1013
+ {
1014
+ "epoch": 0.4721311475409836,
1015
+ "grad_norm": 9.007691383361816,
1016
+ "learning_rate": 9.590950631703589e-06,
1017
+ "loss": 2.125,
1018
+ "step": 144
1019
+ },
1020
+ {
1021
+ "epoch": 0.47540983606557374,
1022
+ "grad_norm": 9.498886108398438,
1023
+ "learning_rate": 9.583906571931248e-06,
1024
+ "loss": 2.1172,
1025
+ "step": 145
1026
+ },
1027
+ {
1028
+ "epoch": 0.4786885245901639,
1029
+ "grad_norm": 10.918264389038086,
1030
+ "learning_rate": 9.576805009515588e-06,
1031
+ "loss": 2.0508,
1032
+ "step": 146
1033
+ },
1034
+ {
1035
+ "epoch": 0.4819672131147541,
1036
+ "grad_norm": 8.792062759399414,
1037
+ "learning_rate": 9.569646033541888e-06,
1038
+ "loss": 2.1484,
1039
+ "step": 147
1040
+ },
1041
+ {
1042
+ "epoch": 0.4852459016393443,
1043
+ "grad_norm": 9.490158081054688,
1044
+ "learning_rate": 9.562429733815655e-06,
1045
+ "loss": 2.1133,
1046
+ "step": 148
1047
+ },
1048
+ {
1049
+ "epoch": 0.4885245901639344,
1050
+ "grad_norm": 9.344221115112305,
1051
+ "learning_rate": 9.55515620086149e-06,
1052
+ "loss": 2.0684,
1053
+ "step": 149
1054
+ },
1055
+ {
1056
+ "epoch": 0.4918032786885246,
1057
+ "grad_norm": 10.885297775268555,
1058
+ "learning_rate": 9.54782552592195e-06,
1059
+ "loss": 2.0957,
1060
+ "step": 150
1061
+ },
1062
+ {
1063
+ "epoch": 0.49508196721311476,
1064
+ "grad_norm": 8.593646049499512,
1065
+ "learning_rate": 9.540437800956412e-06,
1066
+ "loss": 2.0625,
1067
+ "step": 151
1068
+ },
1069
+ {
1070
+ "epoch": 0.49836065573770494,
1071
+ "grad_norm": 8.966227531433105,
1072
+ "learning_rate": 9.532993118639915e-06,
1073
+ "loss": 2.0,
1074
+ "step": 152
1075
+ },
1076
+ {
1077
+ "epoch": 0.5016393442622951,
1078
+ "grad_norm": 10.044122695922852,
1079
+ "learning_rate": 9.525491572361995e-06,
1080
+ "loss": 2.0176,
1081
+ "step": 153
1082
+ },
1083
+ {
1084
+ "epoch": 0.5049180327868853,
1085
+ "grad_norm": 8.53621768951416,
1086
+ "learning_rate": 9.517933256225516e-06,
1087
+ "loss": 2.1094,
1088
+ "step": 154
1089
+ },
1090
+ {
1091
+ "epoch": 0.5081967213114754,
1092
+ "grad_norm": 10.106084823608398,
1093
+ "learning_rate": 9.510318265045488e-06,
1094
+ "loss": 2.0195,
1095
+ "step": 155
1096
+ },
1097
+ {
1098
+ "epoch": 0.5114754098360655,
1099
+ "grad_norm": 9.872212409973145,
1100
+ "learning_rate": 9.502646694347886e-06,
1101
+ "loss": 2.1055,
1102
+ "step": 156
1103
+ },
1104
+ {
1105
+ "epoch": 0.5147540983606558,
1106
+ "grad_norm": 8.695504188537598,
1107
+ "learning_rate": 9.494918640368435e-06,
1108
+ "loss": 1.9961,
1109
+ "step": 157
1110
+ },
1111
+ {
1112
+ "epoch": 0.5180327868852459,
1113
+ "grad_norm": 9.583698272705078,
1114
+ "learning_rate": 9.487134200051422e-06,
1115
+ "loss": 2.0801,
1116
+ "step": 158
1117
+ },
1118
+ {
1119
+ "epoch": 0.521311475409836,
1120
+ "grad_norm": 9.81774616241455,
1121
+ "learning_rate": 9.479293471048461e-06,
1122
+ "loss": 2.043,
1123
+ "step": 159
1124
+ },
1125
+ {
1126
+ "epoch": 0.5245901639344263,
1127
+ "grad_norm": 8.542972564697266,
1128
+ "learning_rate": 9.471396551717283e-06,
1129
+ "loss": 1.9922,
1130
+ "step": 160
1131
+ },
1132
+ {
1133
+ "epoch": 0.5278688524590164,
1134
+ "grad_norm": 9.894759178161621,
1135
+ "learning_rate": 9.463443541120496e-06,
1136
+ "loss": 2.0469,
1137
+ "step": 161
1138
+ },
1139
+ {
1140
+ "epoch": 0.5311475409836065,
1141
+ "grad_norm": 9.396486282348633,
1142
+ "learning_rate": 9.45543453902434e-06,
1143
+ "loss": 1.9688,
1144
+ "step": 162
1145
+ },
1146
+ {
1147
+ "epoch": 0.5344262295081967,
1148
+ "grad_norm": 9.620316505432129,
1149
+ "learning_rate": 9.447369645897441e-06,
1150
+ "loss": 2.0391,
1151
+ "step": 163
1152
+ },
1153
+ {
1154
+ "epoch": 0.5377049180327869,
1155
+ "grad_norm": 10.236593246459961,
1156
+ "learning_rate": 9.439248962909547e-06,
1157
+ "loss": 1.957,
1158
+ "step": 164
1159
+ },
1160
+ {
1161
+ "epoch": 0.5409836065573771,
1162
+ "grad_norm": 11.980149269104004,
1163
+ "learning_rate": 9.431072591930261e-06,
1164
+ "loss": 2.0293,
1165
+ "step": 165
1166
+ },
1167
+ {
1168
+ "epoch": 0.5442622950819672,
1169
+ "grad_norm": 14.74476146697998,
1170
+ "learning_rate": 9.42284063552776e-06,
1171
+ "loss": 1.9844,
1172
+ "step": 166
1173
+ },
1174
+ {
1175
+ "epoch": 0.5475409836065573,
1176
+ "grad_norm": 9.34021282196045,
1177
+ "learning_rate": 9.414553196967508e-06,
1178
+ "loss": 2.0703,
1179
+ "step": 167
1180
+ },
1181
+ {
1182
+ "epoch": 0.5508196721311476,
1183
+ "grad_norm": 12.805119514465332,
1184
+ "learning_rate": 9.406210380210971e-06,
1185
+ "loss": 2.0117,
1186
+ "step": 168
1187
+ },
1188
+ {
1189
+ "epoch": 0.5540983606557377,
1190
+ "grad_norm": 9.839883804321289,
1191
+ "learning_rate": 9.397812289914297e-06,
1192
+ "loss": 1.9727,
1193
+ "step": 169
1194
+ },
1195
+ {
1196
+ "epoch": 0.5573770491803278,
1197
+ "grad_norm": 14.596024513244629,
1198
+ "learning_rate": 9.389359031427013e-06,
1199
+ "loss": 2.0742,
1200
+ "step": 170
1201
+ },
1202
+ {
1203
+ "epoch": 0.5606557377049181,
1204
+ "grad_norm": 13.544656753540039,
1205
+ "learning_rate": 9.380850710790701e-06,
1206
+ "loss": 2.0742,
1207
+ "step": 171
1208
+ },
1209
+ {
1210
+ "epoch": 0.5639344262295082,
1211
+ "grad_norm": 9.45289421081543,
1212
+ "learning_rate": 9.372287434737673e-06,
1213
+ "loss": 2.0586,
1214
+ "step": 172
1215
+ },
1216
+ {
1217
+ "epoch": 0.5672131147540984,
1218
+ "grad_norm": 10.174077987670898,
1219
+ "learning_rate": 9.363669310689617e-06,
1220
+ "loss": 2.1367,
1221
+ "step": 173
1222
+ },
1223
+ {
1224
+ "epoch": 0.5704918032786885,
1225
+ "grad_norm": 12.690667152404785,
1226
+ "learning_rate": 9.354996446756268e-06,
1227
+ "loss": 1.9395,
1228
+ "step": 174
1229
+ },
1230
+ {
1231
+ "epoch": 0.5737704918032787,
1232
+ "grad_norm": 10.625876426696777,
1233
+ "learning_rate": 9.346268951734042e-06,
1234
+ "loss": 2.0215,
1235
+ "step": 175
1236
+ },
1237
+ {
1238
+ "epoch": 0.5770491803278689,
1239
+ "grad_norm": 11.856694221496582,
1240
+ "learning_rate": 9.337486935104672e-06,
1241
+ "loss": 1.9766,
1242
+ "step": 176
1243
+ },
1244
+ {
1245
+ "epoch": 0.580327868852459,
1246
+ "grad_norm": 10.2452974319458,
1247
+ "learning_rate": 9.328650507033832e-06,
1248
+ "loss": 2.0586,
1249
+ "step": 177
1250
+ },
1251
+ {
1252
+ "epoch": 0.5836065573770491,
1253
+ "grad_norm": 12.020367622375488,
1254
+ "learning_rate": 9.319759778369764e-06,
1255
+ "loss": 2.0352,
1256
+ "step": 178
1257
+ },
1258
+ {
1259
+ "epoch": 0.5868852459016394,
1260
+ "grad_norm": 11.364824295043945,
1261
+ "learning_rate": 9.310814860641874e-06,
1262
+ "loss": 2.0352,
1263
+ "step": 179
1264
+ },
1265
+ {
1266
+ "epoch": 0.5901639344262295,
1267
+ "grad_norm": 12.439066886901855,
1268
+ "learning_rate": 9.301815866059353e-06,
1269
+ "loss": 2.0703,
1270
+ "step": 180
1271
+ },
1272
+ {
1273
+ "epoch": 0.5934426229508196,
1274
+ "grad_norm": 12.66345500946045,
1275
+ "learning_rate": 9.292762907509742e-06,
1276
+ "loss": 1.9922,
1277
+ "step": 181
1278
+ },
1279
+ {
1280
+ "epoch": 0.5967213114754099,
1281
+ "grad_norm": 11.868645668029785,
1282
+ "learning_rate": 9.283656098557545e-06,
1283
+ "loss": 1.9863,
1284
+ "step": 182
1285
+ },
1286
+ {
1287
+ "epoch": 0.6,
1288
+ "grad_norm": 11.235689163208008,
1289
+ "learning_rate": 9.27449555344278e-06,
1290
+ "loss": 2.0195,
1291
+ "step": 183
1292
+ },
1293
+ {
1294
+ "epoch": 0.6032786885245902,
1295
+ "grad_norm": 10.719315528869629,
1296
+ "learning_rate": 9.26528138707956e-06,
1297
+ "loss": 1.9883,
1298
+ "step": 184
1299
+ },
1300
+ {
1301
+ "epoch": 0.6065573770491803,
1302
+ "grad_norm": 11.960482597351074,
1303
+ "learning_rate": 9.256013715054648e-06,
1304
+ "loss": 1.9082,
1305
+ "step": 185
1306
+ },
1307
+ {
1308
+ "epoch": 0.6098360655737705,
1309
+ "grad_norm": 9.54014778137207,
1310
+ "learning_rate": 9.24669265362601e-06,
1311
+ "loss": 2.0156,
1312
+ "step": 186
1313
+ },
1314
+ {
1315
+ "epoch": 0.6131147540983607,
1316
+ "grad_norm": 10.753348350524902,
1317
+ "learning_rate": 9.237318319721346e-06,
1318
+ "loss": 1.9746,
1319
+ "step": 187
1320
+ },
1321
+ {
1322
+ "epoch": 0.6163934426229508,
1323
+ "grad_norm": 9.75783634185791,
1324
+ "learning_rate": 9.227890830936634e-06,
1325
+ "loss": 2.0195,
1326
+ "step": 188
1327
+ },
1328
+ {
1329
+ "epoch": 0.6196721311475409,
1330
+ "grad_norm": 15.201295852661133,
1331
+ "learning_rate": 9.218410305534657e-06,
1332
+ "loss": 2.0117,
1333
+ "step": 189
1334
+ },
1335
+ {
1336
+ "epoch": 0.6229508196721312,
1337
+ "grad_norm": 10.927349090576172,
1338
+ "learning_rate": 9.208876862443504e-06,
1339
+ "loss": 1.9316,
1340
+ "step": 190
1341
+ },
1342
+ {
1343
+ "epoch": 0.6262295081967213,
1344
+ "grad_norm": 13.822660446166992,
1345
+ "learning_rate": 9.199290621255095e-06,
1346
+ "loss": 2.0371,
1347
+ "step": 191
1348
+ },
1349
+ {
1350
+ "epoch": 0.6295081967213115,
1351
+ "grad_norm": 13.426775932312012,
1352
+ "learning_rate": 9.189651702223672e-06,
1353
+ "loss": 2.0273,
1354
+ "step": 192
1355
+ },
1356
+ {
1357
+ "epoch": 0.6327868852459017,
1358
+ "grad_norm": 9.90354061126709,
1359
+ "learning_rate": 9.179960226264288e-06,
1360
+ "loss": 1.957,
1361
+ "step": 193
1362
+ },
1363
+ {
1364
+ "epoch": 0.6360655737704918,
1365
+ "grad_norm": 20.340970993041992,
1366
+ "learning_rate": 9.170216314951301e-06,
1367
+ "loss": 2.0488,
1368
+ "step": 194
1369
+ },
1370
+ {
1371
+ "epoch": 0.639344262295082,
1372
+ "grad_norm": 9.366877555847168,
1373
+ "learning_rate": 9.16042009051684e-06,
1374
+ "loss": 1.9688,
1375
+ "step": 195
1376
+ },
1377
+ {
1378
+ "epoch": 0.6426229508196721,
1379
+ "grad_norm": 14.084966659545898,
1380
+ "learning_rate": 9.150571675849273e-06,
1381
+ "loss": 2.0938,
1382
+ "step": 196
1383
+ },
1384
+ {
1385
+ "epoch": 0.6459016393442623,
1386
+ "grad_norm": 16.520889282226562,
1387
+ "learning_rate": 9.140671194491666e-06,
1388
+ "loss": 2.0508,
1389
+ "step": 197
1390
+ },
1391
+ {
1392
+ "epoch": 0.6491803278688525,
1393
+ "grad_norm": 13.539576530456543,
1394
+ "learning_rate": 9.130718770640235e-06,
1395
+ "loss": 1.9922,
1396
+ "step": 198
1397
+ },
1398
+ {
1399
+ "epoch": 0.6524590163934426,
1400
+ "grad_norm": 14.922486305236816,
1401
+ "learning_rate": 9.120714529142784e-06,
1402
+ "loss": 2.0371,
1403
+ "step": 199
1404
+ },
1405
+ {
1406
+ "epoch": 0.6557377049180327,
1407
+ "grad_norm": 12.108205795288086,
1408
+ "learning_rate": 9.110658595497148e-06,
1409
+ "loss": 1.9902,
1410
+ "step": 200
1411
+ },
1412
+ {
1413
+ "epoch": 0.659016393442623,
1414
+ "grad_norm": 10.225913047790527,
1415
+ "learning_rate": 9.100551095849603e-06,
1416
+ "loss": 1.8574,
1417
+ "step": 201
1418
+ },
1419
+ {
1420
+ "epoch": 0.6622950819672131,
1421
+ "grad_norm": 10.973288536071777,
1422
+ "learning_rate": 9.090392156993301e-06,
1423
+ "loss": 1.9902,
1424
+ "step": 202
1425
+ },
1426
+ {
1427
+ "epoch": 0.6655737704918033,
1428
+ "grad_norm": 12.73608684539795,
1429
+ "learning_rate": 9.080181906366668e-06,
1430
+ "loss": 1.9551,
1431
+ "step": 203
1432
+ },
1433
+ {
1434
+ "epoch": 0.6688524590163935,
1435
+ "grad_norm": 11.886429786682129,
1436
+ "learning_rate": 9.069920472051804e-06,
1437
+ "loss": 1.957,
1438
+ "step": 204
1439
+ },
1440
+ {
1441
+ "epoch": 0.6721311475409836,
1442
+ "grad_norm": 13.453667640686035,
1443
+ "learning_rate": 9.05960798277289e-06,
1444
+ "loss": 1.9512,
1445
+ "step": 205
1446
+ },
1447
+ {
1448
+ "epoch": 0.6754098360655738,
1449
+ "grad_norm": 19.135602951049805,
1450
+ "learning_rate": 9.04924456789456e-06,
1451
+ "loss": 1.9688,
1452
+ "step": 206
1453
+ },
1454
+ {
1455
+ "epoch": 0.6786885245901639,
1456
+ "grad_norm": 16.149311065673828,
1457
+ "learning_rate": 9.038830357420276e-06,
1458
+ "loss": 1.9688,
1459
+ "step": 207
1460
+ },
1461
+ {
1462
+ "epoch": 0.6819672131147541,
1463
+ "grad_norm": 14.444845199584961,
1464
+ "learning_rate": 9.028365481990713e-06,
1465
+ "loss": 2.0625,
1466
+ "step": 208
1467
+ },
1468
+ {
1469
+ "epoch": 0.6852459016393443,
1470
+ "grad_norm": 12.08676528930664,
1471
+ "learning_rate": 9.017850072882107e-06,
1472
+ "loss": 1.9121,
1473
+ "step": 209
1474
+ },
1475
+ {
1476
+ "epoch": 0.6885245901639344,
1477
+ "grad_norm": 11.756418228149414,
1478
+ "learning_rate": 9.007284262004608e-06,
1479
+ "loss": 1.9102,
1480
+ "step": 210
1481
+ },
1482
+ {
1483
+ "epoch": 0.6918032786885245,
1484
+ "grad_norm": 12.56296443939209,
1485
+ "learning_rate": 8.996668181900636e-06,
1486
+ "loss": 2.0312,
1487
+ "step": 211
1488
+ },
1489
+ {
1490
+ "epoch": 0.6950819672131148,
1491
+ "grad_norm": 10.218338966369629,
1492
+ "learning_rate": 8.986001965743205e-06,
1493
+ "loss": 1.9629,
1494
+ "step": 212
1495
+ },
1496
+ {
1497
+ "epoch": 0.6983606557377049,
1498
+ "grad_norm": 9.9441556930542,
1499
+ "learning_rate": 8.975285747334267e-06,
1500
+ "loss": 1.9844,
1501
+ "step": 213
1502
+ },
1503
+ {
1504
+ "epoch": 0.7016393442622951,
1505
+ "grad_norm": 10.612693786621094,
1506
+ "learning_rate": 8.964519661103014e-06,
1507
+ "loss": 1.9746,
1508
+ "step": 214
1509
+ },
1510
+ {
1511
+ "epoch": 0.7049180327868853,
1512
+ "grad_norm": 10.07176399230957,
1513
+ "learning_rate": 8.953703842104212e-06,
1514
+ "loss": 1.9531,
1515
+ "step": 215
1516
+ },
1517
+ {
1518
+ "epoch": 0.7081967213114754,
1519
+ "grad_norm": 9.502864837646484,
1520
+ "learning_rate": 8.942838426016496e-06,
1521
+ "loss": 1.9492,
1522
+ "step": 216
1523
+ },
1524
+ {
1525
+ "epoch": 0.7114754098360656,
1526
+ "grad_norm": 9.91391372680664,
1527
+ "learning_rate": 8.931923549140668e-06,
1528
+ "loss": 1.9844,
1529
+ "step": 217
1530
+ },
1531
+ {
1532
+ "epoch": 0.7147540983606557,
1533
+ "grad_norm": 9.780139923095703,
1534
+ "learning_rate": 8.920959348397988e-06,
1535
+ "loss": 1.9473,
1536
+ "step": 218
1537
+ },
1538
+ {
1539
+ "epoch": 0.7180327868852459,
1540
+ "grad_norm": 15.030618667602539,
1541
+ "learning_rate": 8.909945961328462e-06,
1542
+ "loss": 1.9922,
1543
+ "step": 219
1544
+ },
1545
+ {
1546
+ "epoch": 0.7213114754098361,
1547
+ "grad_norm": 10.924345016479492,
1548
+ "learning_rate": 8.898883526089108e-06,
1549
+ "loss": 2.0195,
1550
+ "step": 220
1551
+ },
1552
+ {
1553
+ "epoch": 0.7245901639344262,
1554
+ "grad_norm": 10.839616775512695,
1555
+ "learning_rate": 8.887772181452228e-06,
1556
+ "loss": 2.0234,
1557
+ "step": 221
1558
+ },
1559
+ {
1560
+ "epoch": 0.7278688524590164,
1561
+ "grad_norm": 11.212786674499512,
1562
+ "learning_rate": 8.876612066803666e-06,
1563
+ "loss": 1.9727,
1564
+ "step": 222
1565
+ },
1566
+ {
1567
+ "epoch": 0.7311475409836066,
1568
+ "grad_norm": 9.742143630981445,
1569
+ "learning_rate": 8.86540332214106e-06,
1570
+ "loss": 1.9844,
1571
+ "step": 223
1572
+ },
1573
+ {
1574
+ "epoch": 0.7344262295081967,
1575
+ "grad_norm": 12.224595069885254,
1576
+ "learning_rate": 8.854146088072087e-06,
1577
+ "loss": 2.0566,
1578
+ "step": 224
1579
+ },
1580
+ {
1581
+ "epoch": 0.7377049180327869,
1582
+ "grad_norm": 9.907976150512695,
1583
+ "learning_rate": 8.842840505812691e-06,
1584
+ "loss": 2.0215,
1585
+ "step": 225
1586
+ },
1587
+ {
1588
+ "epoch": 0.740983606557377,
1589
+ "grad_norm": 10.605637550354004,
1590
+ "learning_rate": 8.83148671718533e-06,
1591
+ "loss": 1.959,
1592
+ "step": 226
1593
+ },
1594
+ {
1595
+ "epoch": 0.7442622950819672,
1596
+ "grad_norm": 10.371489524841309,
1597
+ "learning_rate": 8.82008486461717e-06,
1598
+ "loss": 1.9492,
1599
+ "step": 227
1600
+ },
1601
+ {
1602
+ "epoch": 0.7475409836065574,
1603
+ "grad_norm": 13.147061347961426,
1604
+ "learning_rate": 8.808635091138327e-06,
1605
+ "loss": 1.9375,
1606
+ "step": 228
1607
+ },
1608
+ {
1609
+ "epoch": 0.7508196721311475,
1610
+ "grad_norm": 13.442245483398438,
1611
+ "learning_rate": 8.79713754038005e-06,
1612
+ "loss": 1.9219,
1613
+ "step": 229
1614
+ },
1615
+ {
1616
+ "epoch": 0.7540983606557377,
1617
+ "grad_norm": 10.87454891204834,
1618
+ "learning_rate": 8.785592356572931e-06,
1619
+ "loss": 1.9609,
1620
+ "step": 230
1621
+ },
1622
+ {
1623
+ "epoch": 0.7573770491803279,
1624
+ "grad_norm": 12.752360343933105,
1625
+ "learning_rate": 8.773999684545096e-06,
1626
+ "loss": 1.9414,
1627
+ "step": 231
1628
+ },
1629
+ {
1630
+ "epoch": 0.760655737704918,
1631
+ "grad_norm": 10.660215377807617,
1632
+ "learning_rate": 8.76235966972038e-06,
1633
+ "loss": 1.9395,
1634
+ "step": 232
1635
+ },
1636
+ {
1637
+ "epoch": 0.7639344262295082,
1638
+ "grad_norm": 13.825894355773926,
1639
+ "learning_rate": 8.750672458116513e-06,
1640
+ "loss": 1.9102,
1641
+ "step": 233
1642
+ },
1643
+ {
1644
+ "epoch": 0.7672131147540984,
1645
+ "grad_norm": 11.399506568908691,
1646
+ "learning_rate": 8.738938196343278e-06,
1647
+ "loss": 2.002,
1648
+ "step": 234
1649
+ },
1650
+ {
1651
+ "epoch": 0.7704918032786885,
1652
+ "grad_norm": 10.732869148254395,
1653
+ "learning_rate": 8.72715703160068e-06,
1654
+ "loss": 1.8887,
1655
+ "step": 235
1656
+ },
1657
+ {
1658
+ "epoch": 0.7737704918032787,
1659
+ "grad_norm": 13.854918479919434,
1660
+ "learning_rate": 8.715329111677102e-06,
1661
+ "loss": 1.9922,
1662
+ "step": 236
1663
+ },
1664
+ {
1665
+ "epoch": 0.7770491803278688,
1666
+ "grad_norm": 12.330432891845703,
1667
+ "learning_rate": 8.703454584947433e-06,
1668
+ "loss": 1.9395,
1669
+ "step": 237
1670
+ },
1671
+ {
1672
+ "epoch": 0.780327868852459,
1673
+ "grad_norm": 9.50846004486084,
1674
+ "learning_rate": 8.691533600371227e-06,
1675
+ "loss": 1.8965,
1676
+ "step": 238
1677
+ },
1678
+ {
1679
+ "epoch": 0.7836065573770492,
1680
+ "grad_norm": 10.240988731384277,
1681
+ "learning_rate": 8.67956630749083e-06,
1682
+ "loss": 1.9121,
1683
+ "step": 239
1684
+ },
1685
+ {
1686
+ "epoch": 0.7868852459016393,
1687
+ "grad_norm": 10.392232894897461,
1688
+ "learning_rate": 8.667552856429494e-06,
1689
+ "loss": 1.8809,
1690
+ "step": 240
1691
+ },
1692
+ {
1693
+ "epoch": 0.7901639344262295,
1694
+ "grad_norm": 10.917051315307617,
1695
+ "learning_rate": 8.655493397889503e-06,
1696
+ "loss": 1.9805,
1697
+ "step": 241
1698
+ },
1699
+ {
1700
+ "epoch": 0.7934426229508197,
1701
+ "grad_norm": 10.163880348205566,
1702
+ "learning_rate": 8.64338808315028e-06,
1703
+ "loss": 1.8711,
1704
+ "step": 242
1705
+ },
1706
+ {
1707
+ "epoch": 0.7967213114754098,
1708
+ "grad_norm": 11.161616325378418,
1709
+ "learning_rate": 8.631237064066491e-06,
1710
+ "loss": 1.9102,
1711
+ "step": 243
1712
+ },
1713
+ {
1714
+ "epoch": 0.8,
1715
+ "grad_norm": 11.414627075195312,
1716
+ "learning_rate": 8.61904049306614e-06,
1717
+ "loss": 1.9727,
1718
+ "step": 244
1719
+ },
1720
+ {
1721
+ "epoch": 0.8032786885245902,
1722
+ "grad_norm": 10.282430648803711,
1723
+ "learning_rate": 8.606798523148648e-06,
1724
+ "loss": 1.9414,
1725
+ "step": 245
1726
+ },
1727
+ {
1728
+ "epoch": 0.8065573770491803,
1729
+ "grad_norm": 12.212276458740234,
1730
+ "learning_rate": 8.594511307882945e-06,
1731
+ "loss": 1.8496,
1732
+ "step": 246
1733
+ },
1734
+ {
1735
+ "epoch": 0.8098360655737705,
1736
+ "grad_norm": 14.458395957946777,
1737
+ "learning_rate": 8.582179001405544e-06,
1738
+ "loss": 1.9414,
1739
+ "step": 247
1740
+ },
1741
+ {
1742
+ "epoch": 0.8131147540983606,
1743
+ "grad_norm": 10.135568618774414,
1744
+ "learning_rate": 8.569801758418596e-06,
1745
+ "loss": 1.8691,
1746
+ "step": 248
1747
+ },
1748
+ {
1749
+ "epoch": 0.8163934426229508,
1750
+ "grad_norm": 14.193105697631836,
1751
+ "learning_rate": 8.55737973418796e-06,
1752
+ "loss": 1.834,
1753
+ "step": 249
1754
+ },
1755
+ {
1756
+ "epoch": 0.819672131147541,
1757
+ "grad_norm": 10.561907768249512,
1758
+ "learning_rate": 8.544913084541253e-06,
1759
+ "loss": 2.043,
1760
+ "step": 250
1761
+ },
1762
+ {
1763
+ "epoch": 0.8229508196721311,
1764
+ "grad_norm": 12.753805160522461,
1765
+ "learning_rate": 8.532401965865889e-06,
1766
+ "loss": 1.9141,
1767
+ "step": 251
1768
+ },
1769
+ {
1770
+ "epoch": 0.8262295081967214,
1771
+ "grad_norm": 11.008598327636719,
1772
+ "learning_rate": 8.519846535107129e-06,
1773
+ "loss": 1.9258,
1774
+ "step": 252
1775
+ },
1776
+ {
1777
+ "epoch": 0.8295081967213115,
1778
+ "grad_norm": 10.186516761779785,
1779
+ "learning_rate": 8.507246949766095e-06,
1780
+ "loss": 1.9062,
1781
+ "step": 253
1782
+ },
1783
+ {
1784
+ "epoch": 0.8327868852459016,
1785
+ "grad_norm": 12.843867301940918,
1786
+ "learning_rate": 8.494603367897817e-06,
1787
+ "loss": 1.8516,
1788
+ "step": 254
1789
+ },
1790
+ {
1791
+ "epoch": 0.8360655737704918,
1792
+ "grad_norm": 11.588765144348145,
1793
+ "learning_rate": 8.481915948109226e-06,
1794
+ "loss": 1.9121,
1795
+ "step": 255
1796
+ },
1797
+ {
1798
+ "epoch": 0.839344262295082,
1799
+ "grad_norm": 12.169210433959961,
1800
+ "learning_rate": 8.469184849557188e-06,
1801
+ "loss": 1.9238,
1802
+ "step": 256
1803
+ },
1804
+ {
1805
+ "epoch": 0.8426229508196721,
1806
+ "grad_norm": 17.143991470336914,
1807
+ "learning_rate": 8.456410231946484e-06,
1808
+ "loss": 1.8887,
1809
+ "step": 257
1810
+ },
1811
+ {
1812
+ "epoch": 0.8459016393442623,
1813
+ "grad_norm": 9.951960563659668,
1814
+ "learning_rate": 8.443592255527824e-06,
1815
+ "loss": 1.9355,
1816
+ "step": 258
1817
+ },
1818
+ {
1819
+ "epoch": 0.8491803278688524,
1820
+ "grad_norm": 11.594623565673828,
1821
+ "learning_rate": 8.430731081095832e-06,
1822
+ "loss": 1.9512,
1823
+ "step": 259
1824
+ },
1825
+ {
1826
+ "epoch": 0.8524590163934426,
1827
+ "grad_norm": 10.37341022491455,
1828
+ "learning_rate": 8.417826869987026e-06,
1829
+ "loss": 1.9883,
1830
+ "step": 260
1831
+ },
1832
+ {
1833
+ "epoch": 0.8557377049180328,
1834
+ "grad_norm": 13.815895080566406,
1835
+ "learning_rate": 8.404879784077793e-06,
1836
+ "loss": 1.9746,
1837
+ "step": 261
1838
+ },
1839
+ {
1840
+ "epoch": 0.8590163934426229,
1841
+ "grad_norm": 10.756362915039062,
1842
+ "learning_rate": 8.39188998578237e-06,
1843
+ "loss": 1.9023,
1844
+ "step": 262
1845
+ },
1846
+ {
1847
+ "epoch": 0.8622950819672132,
1848
+ "grad_norm": 11.125969886779785,
1849
+ "learning_rate": 8.37885763805079e-06,
1850
+ "loss": 1.873,
1851
+ "step": 263
1852
+ },
1853
+ {
1854
+ "epoch": 0.8655737704918033,
1855
+ "grad_norm": 10.486440658569336,
1856
+ "learning_rate": 8.365782904366845e-06,
1857
+ "loss": 1.8789,
1858
+ "step": 264
1859
+ },
1860
+ {
1861
+ "epoch": 0.8688524590163934,
1862
+ "grad_norm": 10.452739715576172,
1863
+ "learning_rate": 8.352665948746044e-06,
1864
+ "loss": 1.9453,
1865
+ "step": 265
1866
+ },
1867
+ {
1868
+ "epoch": 0.8721311475409836,
1869
+ "grad_norm": 10.094730377197266,
1870
+ "learning_rate": 8.33950693573354e-06,
1871
+ "loss": 1.8516,
1872
+ "step": 266
1873
+ },
1874
+ {
1875
+ "epoch": 0.8754098360655738,
1876
+ "grad_norm": 11.559701919555664,
1877
+ "learning_rate": 8.326306030402076e-06,
1878
+ "loss": 1.9844,
1879
+ "step": 267
1880
+ },
1881
+ {
1882
+ "epoch": 0.8786885245901639,
1883
+ "grad_norm": 11.006271362304688,
1884
+ "learning_rate": 8.313063398349914e-06,
1885
+ "loss": 1.8887,
1886
+ "step": 268
1887
+ },
1888
+ {
1889
+ "epoch": 0.8819672131147541,
1890
+ "grad_norm": 12.529583930969238,
1891
+ "learning_rate": 8.29977920569875e-06,
1892
+ "loss": 1.877,
1893
+ "step": 269
1894
+ },
1895
+ {
1896
+ "epoch": 0.8852459016393442,
1897
+ "grad_norm": 10.77204704284668,
1898
+ "learning_rate": 8.28645361909164e-06,
1899
+ "loss": 1.998,
1900
+ "step": 270
1901
+ },
1902
+ {
1903
+ "epoch": 0.8885245901639345,
1904
+ "grad_norm": 13.109503746032715,
1905
+ "learning_rate": 8.273086805690903e-06,
1906
+ "loss": 1.8711,
1907
+ "step": 271
1908
+ },
1909
+ {
1910
+ "epoch": 0.8918032786885246,
1911
+ "grad_norm": 15.495031356811523,
1912
+ "learning_rate": 8.259678933176025e-06,
1913
+ "loss": 1.916,
1914
+ "step": 272
1915
+ },
1916
+ {
1917
+ "epoch": 0.8950819672131147,
1918
+ "grad_norm": 11.081769943237305,
1919
+ "learning_rate": 8.24623016974156e-06,
1920
+ "loss": 1.8809,
1921
+ "step": 273
1922
+ },
1923
+ {
1924
+ "epoch": 0.898360655737705,
1925
+ "grad_norm": 10.445962905883789,
1926
+ "learning_rate": 8.232740684095012e-06,
1927
+ "loss": 1.8301,
1928
+ "step": 274
1929
+ },
1930
+ {
1931
+ "epoch": 0.9016393442622951,
1932
+ "grad_norm": 11.152101516723633,
1933
+ "learning_rate": 8.219210645454727e-06,
1934
+ "loss": 1.9258,
1935
+ "step": 275
1936
+ },
1937
+ {
1938
+ "epoch": 0.9049180327868852,
1939
+ "grad_norm": 10.69294261932373,
1940
+ "learning_rate": 8.205640223547766e-06,
1941
+ "loss": 1.8242,
1942
+ "step": 276
1943
+ },
1944
+ {
1945
+ "epoch": 0.9081967213114754,
1946
+ "grad_norm": 13.176959991455078,
1947
+ "learning_rate": 8.192029588607772e-06,
1948
+ "loss": 1.9453,
1949
+ "step": 277
1950
+ },
1951
+ {
1952
+ "epoch": 0.9114754098360656,
1953
+ "grad_norm": 12.394086837768555,
1954
+ "learning_rate": 8.178378911372846e-06,
1955
+ "loss": 1.9434,
1956
+ "step": 278
1957
+ },
1958
+ {
1959
+ "epoch": 0.9147540983606557,
1960
+ "grad_norm": 13.449362754821777,
1961
+ "learning_rate": 8.16468836308339e-06,
1962
+ "loss": 1.9961,
1963
+ "step": 279
1964
+ },
1965
+ {
1966
+ "epoch": 0.9180327868852459,
1967
+ "grad_norm": 10.922250747680664,
1968
+ "learning_rate": 8.150958115479976e-06,
1969
+ "loss": 1.9531,
1970
+ "step": 280
1971
+ },
1972
+ {
1973
+ "epoch": 0.921311475409836,
1974
+ "grad_norm": 14.1903657913208,
1975
+ "learning_rate": 8.137188340801177e-06,
1976
+ "loss": 1.9336,
1977
+ "step": 281
1978
+ },
1979
+ {
1980
+ "epoch": 0.9245901639344263,
1981
+ "grad_norm": 13.266737937927246,
1982
+ "learning_rate": 8.12337921178141e-06,
1983
+ "loss": 1.9688,
1984
+ "step": 282
1985
+ },
1986
+ {
1987
+ "epoch": 0.9278688524590164,
1988
+ "grad_norm": 12.7962064743042,
1989
+ "learning_rate": 8.109530901648776e-06,
1990
+ "loss": 1.875,
1991
+ "step": 283
1992
+ },
1993
+ {
1994
+ "epoch": 0.9311475409836065,
1995
+ "grad_norm": 12.458824157714844,
1996
+ "learning_rate": 8.095643584122879e-06,
1997
+ "loss": 1.9141,
1998
+ "step": 284
1999
+ },
2000
+ {
2001
+ "epoch": 0.9344262295081968,
2002
+ "grad_norm": 10.26793384552002,
2003
+ "learning_rate": 8.08171743341265e-06,
2004
+ "loss": 1.8633,
2005
+ "step": 285
2006
+ },
2007
+ {
2008
+ "epoch": 0.9377049180327869,
2009
+ "grad_norm": 12.272446632385254,
2010
+ "learning_rate": 8.067752624214164e-06,
2011
+ "loss": 1.8945,
2012
+ "step": 286
2013
+ },
2014
+ {
2015
+ "epoch": 0.940983606557377,
2016
+ "grad_norm": 11.974457740783691,
2017
+ "learning_rate": 8.053749331708444e-06,
2018
+ "loss": 1.877,
2019
+ "step": 287
2020
+ },
2021
+ {
2022
+ "epoch": 0.9442622950819672,
2023
+ "grad_norm": 10.955528259277344,
2024
+ "learning_rate": 8.039707731559262e-06,
2025
+ "loss": 1.8867,
2026
+ "step": 288
2027
+ },
2028
+ {
2029
+ "epoch": 0.9475409836065574,
2030
+ "grad_norm": 12.32547378540039,
2031
+ "learning_rate": 8.025627999910946e-06,
2032
+ "loss": 1.9629,
2033
+ "step": 289
2034
+ },
2035
+ {
2036
+ "epoch": 0.9508196721311475,
2037
+ "grad_norm": 12.567078590393066,
2038
+ "learning_rate": 8.011510313386156e-06,
2039
+ "loss": 1.877,
2040
+ "step": 290
2041
+ },
2042
+ {
2043
+ "epoch": 0.9540983606557377,
2044
+ "grad_norm": 10.757590293884277,
2045
+ "learning_rate": 7.997354849083682e-06,
2046
+ "loss": 1.8926,
2047
+ "step": 291
2048
+ },
2049
+ {
2050
+ "epoch": 0.9573770491803278,
2051
+ "grad_norm": 12.629182815551758,
2052
+ "learning_rate": 7.98316178457621e-06,
2053
+ "loss": 1.9199,
2054
+ "step": 292
2055
+ },
2056
+ {
2057
+ "epoch": 0.9606557377049181,
2058
+ "grad_norm": 11.572930335998535,
2059
+ "learning_rate": 7.968931297908105e-06,
2060
+ "loss": 1.9492,
2061
+ "step": 293
2062
+ },
2063
+ {
2064
+ "epoch": 0.9639344262295082,
2065
+ "grad_norm": 13.813660621643066,
2066
+ "learning_rate": 7.954663567593169e-06,
2067
+ "loss": 1.793,
2068
+ "step": 294
2069
+ },
2070
+ {
2071
+ "epoch": 0.9672131147540983,
2072
+ "grad_norm": 10.30826473236084,
2073
+ "learning_rate": 7.940358772612408e-06,
2074
+ "loss": 1.8809,
2075
+ "step": 295
2076
+ },
2077
+ {
2078
+ "epoch": 0.9704918032786886,
2079
+ "grad_norm": 11.040604591369629,
2080
+ "learning_rate": 7.926017092411786e-06,
2081
+ "loss": 1.8438,
2082
+ "step": 296
2083
+ },
2084
+ {
2085
+ "epoch": 0.9737704918032787,
2086
+ "grad_norm": 11.625853538513184,
2087
+ "learning_rate": 7.911638706899968e-06,
2088
+ "loss": 1.8652,
2089
+ "step": 297
2090
+ },
2091
+ {
2092
+ "epoch": 0.9770491803278688,
2093
+ "grad_norm": 12.791728973388672,
2094
+ "learning_rate": 7.89722379644607e-06,
2095
+ "loss": 1.8555,
2096
+ "step": 298
2097
+ },
2098
+ {
2099
+ "epoch": 0.980327868852459,
2100
+ "grad_norm": 15.870569229125977,
2101
+ "learning_rate": 7.882772541877392e-06,
2102
+ "loss": 1.9043,
2103
+ "step": 299
2104
+ },
2105
+ {
2106
+ "epoch": 0.9836065573770492,
2107
+ "grad_norm": 11.703765869140625,
2108
+ "learning_rate": 7.868285124477156e-06,
2109
+ "loss": 1.8613,
2110
+ "step": 300
2111
+ },
2112
+ {
2113
+ "epoch": 0.9868852459016394,
2114
+ "grad_norm": 11.708532333374023,
2115
+ "learning_rate": 7.853761725982221e-06,
2116
+ "loss": 1.8574,
2117
+ "step": 301
2118
+ },
2119
+ {
2120
+ "epoch": 0.9901639344262295,
2121
+ "grad_norm": 10.677489280700684,
2122
+ "learning_rate": 7.839202528580812e-06,
2123
+ "loss": 1.9004,
2124
+ "step": 302
2125
+ },
2126
+ {
2127
+ "epoch": 0.9934426229508196,
2128
+ "grad_norm": 11.307282447814941,
2129
+ "learning_rate": 7.824607714910233e-06,
2130
+ "loss": 1.7969,
2131
+ "step": 303
2132
+ },
2133
+ {
2134
+ "epoch": 0.9967213114754099,
2135
+ "grad_norm": 14.72847843170166,
2136
+ "learning_rate": 7.809977468054574e-06,
2137
+ "loss": 1.8672,
2138
+ "step": 304
2139
+ },
2140
+ {
2141
+ "epoch": 1.0,
2142
+ "grad_norm": 14.625303268432617,
2143
+ "learning_rate": 7.795311971542418e-06,
2144
+ "loss": 1.8945,
2145
+ "step": 305
2146
+ },
2147
+ {
2148
+ "epoch": 1.0032786885245901,
2149
+ "grad_norm": 10.49122142791748,
2150
+ "learning_rate": 7.78061140934453e-06,
2151
+ "loss": 1.709,
2152
+ "step": 306
2153
+ },
2154
+ {
2155
+ "epoch": 1.0065573770491802,
2156
+ "grad_norm": 11.381379127502441,
2157
+ "learning_rate": 7.765875965871561e-06,
2158
+ "loss": 1.6719,
2159
+ "step": 307
2160
+ },
2161
+ {
2162
+ "epoch": 1.0098360655737706,
2163
+ "grad_norm": 13.590991973876953,
2164
+ "learning_rate": 7.751105825971727e-06,
2165
+ "loss": 1.7441,
2166
+ "step": 308
2167
+ },
2168
+ {
2169
+ "epoch": 1.0131147540983607,
2170
+ "grad_norm": 10.184123992919922,
2171
+ "learning_rate": 7.736301174928489e-06,
2172
+ "loss": 1.7246,
2173
+ "step": 309
2174
+ },
2175
+ {
2176
+ "epoch": 1.0163934426229508,
2177
+ "grad_norm": 11.995527267456055,
2178
+ "learning_rate": 7.721462198458235e-06,
2179
+ "loss": 1.6797,
2180
+ "step": 310
2181
+ },
2182
+ {
2183
+ "epoch": 1.019672131147541,
2184
+ "grad_norm": 17.523670196533203,
2185
+ "learning_rate": 7.706589082707951e-06,
2186
+ "loss": 1.6953,
2187
+ "step": 311
2188
+ },
2189
+ {
2190
+ "epoch": 1.022950819672131,
2191
+ "grad_norm": 11.37983226776123,
2192
+ "learning_rate": 7.69168201425287e-06,
2193
+ "loss": 1.6387,
2194
+ "step": 312
2195
+ },
2196
+ {
2197
+ "epoch": 1.0262295081967212,
2198
+ "grad_norm": 16.715530395507812,
2199
+ "learning_rate": 7.676741180094156e-06,
2200
+ "loss": 1.6797,
2201
+ "step": 313
2202
+ },
2203
+ {
2204
+ "epoch": 1.0295081967213116,
2205
+ "grad_norm": 13.554855346679688,
2206
+ "learning_rate": 7.661766767656538e-06,
2207
+ "loss": 1.6738,
2208
+ "step": 314
2209
+ },
2210
+ {
2211
+ "epoch": 1.0327868852459017,
2212
+ "grad_norm": 20.694581985473633,
2213
+ "learning_rate": 7.64675896478597e-06,
2214
+ "loss": 1.6992,
2215
+ "step": 315
2216
+ },
2217
+ {
2218
+ "epoch": 1.0360655737704918,
2219
+ "grad_norm": 16.622581481933594,
2220
+ "learning_rate": 7.631717959747269e-06,
2221
+ "loss": 1.7305,
2222
+ "step": 316
2223
+ },
2224
+ {
2225
+ "epoch": 1.039344262295082,
2226
+ "grad_norm": 19.643783569335938,
2227
+ "learning_rate": 7.616643941221755e-06,
2228
+ "loss": 1.752,
2229
+ "step": 317
2230
+ },
2231
+ {
2232
+ "epoch": 1.042622950819672,
2233
+ "grad_norm": 14.100711822509766,
2234
+ "learning_rate": 7.6015370983048875e-06,
2235
+ "loss": 1.7285,
2236
+ "step": 318
2237
+ },
2238
+ {
2239
+ "epoch": 1.0459016393442624,
2240
+ "grad_norm": 19.153236389160156,
2241
+ "learning_rate": 7.586397620503887e-06,
2242
+ "loss": 1.6953,
2243
+ "step": 319
2244
+ },
2245
+ {
2246
+ "epoch": 1.0491803278688525,
2247
+ "grad_norm": 11.576682090759277,
2248
+ "learning_rate": 7.571225697735365e-06,
2249
+ "loss": 1.668,
2250
+ "step": 320
2251
+ },
2252
+ {
2253
+ "epoch": 1.0524590163934426,
2254
+ "grad_norm": 11.654809951782227,
2255
+ "learning_rate": 7.556021520322932e-06,
2256
+ "loss": 1.7539,
2257
+ "step": 321
2258
+ },
2259
+ {
2260
+ "epoch": 1.0557377049180328,
2261
+ "grad_norm": 14.915346145629883,
2262
+ "learning_rate": 7.54078527899482e-06,
2263
+ "loss": 1.6797,
2264
+ "step": 322
2265
+ },
2266
+ {
2267
+ "epoch": 1.0590163934426229,
2268
+ "grad_norm": 16.05927276611328,
2269
+ "learning_rate": 7.525517164881486e-06,
2270
+ "loss": 1.5918,
2271
+ "step": 323
2272
+ },
2273
+ {
2274
+ "epoch": 1.0622950819672132,
2275
+ "grad_norm": 12.948592185974121,
2276
+ "learning_rate": 7.51021736951321e-06,
2277
+ "loss": 1.5879,
2278
+ "step": 324
2279
+ },
2280
+ {
2281
+ "epoch": 1.0655737704918034,
2282
+ "grad_norm": 14.317193031311035,
2283
+ "learning_rate": 7.4948860848176984e-06,
2284
+ "loss": 1.709,
2285
+ "step": 325
2286
+ },
2287
+ {
2288
+ "epoch": 1.0688524590163935,
2289
+ "grad_norm": 14.903904914855957,
2290
+ "learning_rate": 7.479523503117676e-06,
2291
+ "loss": 1.6738,
2292
+ "step": 326
2293
+ },
2294
+ {
2295
+ "epoch": 1.0721311475409836,
2296
+ "grad_norm": 19.48590087890625,
2297
+ "learning_rate": 7.4641298171284715e-06,
2298
+ "loss": 1.6777,
2299
+ "step": 327
2300
+ },
2301
+ {
2302
+ "epoch": 1.0754098360655737,
2303
+ "grad_norm": 12.375781059265137,
2304
+ "learning_rate": 7.448705219955596e-06,
2305
+ "loss": 1.6719,
2306
+ "step": 328
2307
+ },
2308
+ {
2309
+ "epoch": 1.0786885245901638,
2310
+ "grad_norm": 13.3538179397583,
2311
+ "learning_rate": 7.433249905092331e-06,
2312
+ "loss": 1.707,
2313
+ "step": 329
2314
+ },
2315
+ {
2316
+ "epoch": 1.0819672131147542,
2317
+ "grad_norm": 14.281739234924316,
2318
+ "learning_rate": 7.417764066417289e-06,
2319
+ "loss": 1.7148,
2320
+ "step": 330
2321
+ },
2322
+ {
2323
+ "epoch": 1.0852459016393443,
2324
+ "grad_norm": 12.570676803588867,
2325
+ "learning_rate": 7.4022478981919884e-06,
2326
+ "loss": 1.6445,
2327
+ "step": 331
2328
+ },
2329
+ {
2330
+ "epoch": 1.0885245901639344,
2331
+ "grad_norm": 14.846644401550293,
2332
+ "learning_rate": 7.386701595058422e-06,
2333
+ "loss": 1.6309,
2334
+ "step": 332
2335
+ },
2336
+ {
2337
+ "epoch": 1.0918032786885246,
2338
+ "grad_norm": 12.106386184692383,
2339
+ "learning_rate": 7.3711253520365995e-06,
2340
+ "loss": 1.6641,
2341
+ "step": 333
2342
+ },
2343
+ {
2344
+ "epoch": 1.0950819672131147,
2345
+ "grad_norm": 14.234429359436035,
2346
+ "learning_rate": 7.355519364522117e-06,
2347
+ "loss": 1.6738,
2348
+ "step": 334
2349
+ },
2350
+ {
2351
+ "epoch": 1.098360655737705,
2352
+ "grad_norm": 13.406672477722168,
2353
+ "learning_rate": 7.339883828283696e-06,
2354
+ "loss": 1.6035,
2355
+ "step": 335
2356
+ },
2357
+ {
2358
+ "epoch": 1.1016393442622952,
2359
+ "grad_norm": 12.718591690063477,
2360
+ "learning_rate": 7.324218939460733e-06,
2361
+ "loss": 1.6172,
2362
+ "step": 336
2363
+ },
2364
+ {
2365
+ "epoch": 1.1049180327868853,
2366
+ "grad_norm": 12.047405242919922,
2367
+ "learning_rate": 7.308524894560835e-06,
2368
+ "loss": 1.6328,
2369
+ "step": 337
2370
+ },
2371
+ {
2372
+ "epoch": 1.1081967213114754,
2373
+ "grad_norm": 10.482436180114746,
2374
+ "learning_rate": 7.292801890457357e-06,
2375
+ "loss": 1.5742,
2376
+ "step": 338
2377
+ },
2378
+ {
2379
+ "epoch": 1.1114754098360655,
2380
+ "grad_norm": 13.475817680358887,
2381
+ "learning_rate": 7.2770501243869315e-06,
2382
+ "loss": 1.7402,
2383
+ "step": 339
2384
+ },
2385
+ {
2386
+ "epoch": 1.1147540983606556,
2387
+ "grad_norm": 12.393983840942383,
2388
+ "learning_rate": 7.261269793946994e-06,
2389
+ "loss": 1.6562,
2390
+ "step": 340
2391
+ },
2392
+ {
2393
+ "epoch": 1.118032786885246,
2394
+ "grad_norm": 12.331396102905273,
2395
+ "learning_rate": 7.245461097093307e-06,
2396
+ "loss": 1.6758,
2397
+ "step": 341
2398
+ },
2399
+ {
2400
+ "epoch": 1.1213114754098361,
2401
+ "grad_norm": 14.81454086303711,
2402
+ "learning_rate": 7.229624232137472e-06,
2403
+ "loss": 1.6152,
2404
+ "step": 342
2405
+ },
2406
+ {
2407
+ "epoch": 1.1245901639344262,
2408
+ "grad_norm": 15.760224342346191,
2409
+ "learning_rate": 7.213759397744446e-06,
2410
+ "loss": 1.6465,
2411
+ "step": 343
2412
+ },
2413
+ {
2414
+ "epoch": 1.1278688524590164,
2415
+ "grad_norm": 12.658391952514648,
2416
+ "learning_rate": 7.197866792930044e-06,
2417
+ "loss": 1.6309,
2418
+ "step": 344
2419
+ },
2420
+ {
2421
+ "epoch": 1.1311475409836065,
2422
+ "grad_norm": 13.215618133544922,
2423
+ "learning_rate": 7.1819466170584485e-06,
2424
+ "loss": 1.582,
2425
+ "step": 345
2426
+ },
2427
+ {
2428
+ "epoch": 1.1344262295081968,
2429
+ "grad_norm": 17.812692642211914,
2430
+ "learning_rate": 7.165999069839708e-06,
2431
+ "loss": 1.6035,
2432
+ "step": 346
2433
+ },
2434
+ {
2435
+ "epoch": 1.137704918032787,
2436
+ "grad_norm": 12.584436416625977,
2437
+ "learning_rate": 7.1500243513272275e-06,
2438
+ "loss": 1.6875,
2439
+ "step": 347
2440
+ },
2441
+ {
2442
+ "epoch": 1.140983606557377,
2443
+ "grad_norm": 18.275434494018555,
2444
+ "learning_rate": 7.1340226619152595e-06,
2445
+ "loss": 1.6914,
2446
+ "step": 348
2447
+ },
2448
+ {
2449
+ "epoch": 1.1442622950819672,
2450
+ "grad_norm": 13.600272178649902,
2451
+ "learning_rate": 7.117994202336398e-06,
2452
+ "loss": 1.6367,
2453
+ "step": 349
2454
+ },
2455
+ {
2456
+ "epoch": 1.1475409836065573,
2457
+ "grad_norm": 16.630714416503906,
2458
+ "learning_rate": 7.101939173659045e-06,
2459
+ "loss": 1.6191,
2460
+ "step": 350
2461
+ },
2462
+ {
2463
+ "epoch": 1.1508196721311474,
2464
+ "grad_norm": 15.959205627441406,
2465
+ "learning_rate": 7.085857777284908e-06,
2466
+ "loss": 1.7305,
2467
+ "step": 351
2468
+ },
2469
+ {
2470
+ "epoch": 1.1540983606557378,
2471
+ "grad_norm": 13.133319854736328,
2472
+ "learning_rate": 7.069750214946458e-06,
2473
+ "loss": 1.6641,
2474
+ "step": 352
2475
+ },
2476
+ {
2477
+ "epoch": 1.157377049180328,
2478
+ "grad_norm": 20.936429977416992,
2479
+ "learning_rate": 7.053616688704399e-06,
2480
+ "loss": 1.6738,
2481
+ "step": 353
2482
+ },
2483
+ {
2484
+ "epoch": 1.160655737704918,
2485
+ "grad_norm": 12.605098724365234,
2486
+ "learning_rate": 7.037457400945147e-06,
2487
+ "loss": 1.6953,
2488
+ "step": 354
2489
+ },
2490
+ {
2491
+ "epoch": 1.1639344262295082,
2492
+ "grad_norm": 14.746296882629395,
2493
+ "learning_rate": 7.021272554378279e-06,
2494
+ "loss": 1.75,
2495
+ "step": 355
2496
+ },
2497
+ {
2498
+ "epoch": 1.1672131147540983,
2499
+ "grad_norm": 14.727060317993164,
2500
+ "learning_rate": 7.005062352033991e-06,
2501
+ "loss": 1.6406,
2502
+ "step": 356
2503
+ },
2504
+ {
2505
+ "epoch": 1.1704918032786886,
2506
+ "grad_norm": 11.865753173828125,
2507
+ "learning_rate": 6.988826997260556e-06,
2508
+ "loss": 1.584,
2509
+ "step": 357
2510
+ },
2511
+ {
2512
+ "epoch": 1.1737704918032787,
2513
+ "grad_norm": 12.896044731140137,
2514
+ "learning_rate": 6.972566693721772e-06,
2515
+ "loss": 1.6797,
2516
+ "step": 358
2517
+ },
2518
+ {
2519
+ "epoch": 1.1770491803278689,
2520
+ "grad_norm": 12.144120216369629,
2521
+ "learning_rate": 6.956281645394404e-06,
2522
+ "loss": 1.6055,
2523
+ "step": 359
2524
+ },
2525
+ {
2526
+ "epoch": 1.180327868852459,
2527
+ "grad_norm": 12.519413948059082,
2528
+ "learning_rate": 6.939972056565626e-06,
2529
+ "loss": 1.6895,
2530
+ "step": 360
2531
+ },
2532
+ {
2533
+ "epoch": 1.1836065573770491,
2534
+ "grad_norm": 14.610289573669434,
2535
+ "learning_rate": 6.923638131830461e-06,
2536
+ "loss": 1.6582,
2537
+ "step": 361
2538
+ },
2539
+ {
2540
+ "epoch": 1.1868852459016392,
2541
+ "grad_norm": 11.816085815429688,
2542
+ "learning_rate": 6.907280076089213e-06,
2543
+ "loss": 1.5996,
2544
+ "step": 362
2545
+ },
2546
+ {
2547
+ "epoch": 1.1901639344262296,
2548
+ "grad_norm": 13.615991592407227,
2549
+ "learning_rate": 6.890898094544894e-06,
2550
+ "loss": 1.6504,
2551
+ "step": 363
2552
+ },
2553
+ {
2554
+ "epoch": 1.1934426229508197,
2555
+ "grad_norm": 16.203214645385742,
2556
+ "learning_rate": 6.874492392700659e-06,
2557
+ "loss": 1.6758,
2558
+ "step": 364
2559
+ },
2560
+ {
2561
+ "epoch": 1.1967213114754098,
2562
+ "grad_norm": 12.530990600585938,
2563
+ "learning_rate": 6.858063176357214e-06,
2564
+ "loss": 1.6738,
2565
+ "step": 365
2566
+ },
2567
+ {
2568
+ "epoch": 1.2,
2569
+ "grad_norm": 13.96877384185791,
2570
+ "learning_rate": 6.841610651610243e-06,
2571
+ "loss": 1.6953,
2572
+ "step": 366
2573
+ },
2574
+ {
2575
+ "epoch": 1.20327868852459,
2576
+ "grad_norm": 13.824159622192383,
2577
+ "learning_rate": 6.825135024847825e-06,
2578
+ "loss": 1.6543,
2579
+ "step": 367
2580
+ },
2581
+ {
2582
+ "epoch": 1.2065573770491804,
2583
+ "grad_norm": 12.192745208740234,
2584
+ "learning_rate": 6.808636502747841e-06,
2585
+ "loss": 1.6445,
2586
+ "step": 368
2587
+ },
2588
+ {
2589
+ "epoch": 1.2098360655737705,
2590
+ "grad_norm": 12.605790138244629,
2591
+ "learning_rate": 6.792115292275374e-06,
2592
+ "loss": 1.668,
2593
+ "step": 369
2594
+ },
2595
+ {
2596
+ "epoch": 1.2131147540983607,
2597
+ "grad_norm": 15.496692657470703,
2598
+ "learning_rate": 6.775571600680133e-06,
2599
+ "loss": 1.6426,
2600
+ "step": 370
2601
+ },
2602
+ {
2603
+ "epoch": 1.2163934426229508,
2604
+ "grad_norm": 13.692412376403809,
2605
+ "learning_rate": 6.759005635493831e-06,
2606
+ "loss": 1.7207,
2607
+ "step": 371
2608
+ },
2609
+ {
2610
+ "epoch": 1.219672131147541,
2611
+ "grad_norm": 17.969615936279297,
2612
+ "learning_rate": 6.742417604527594e-06,
2613
+ "loss": 1.5938,
2614
+ "step": 372
2615
+ },
2616
+ {
2617
+ "epoch": 1.222950819672131,
2618
+ "grad_norm": 18.373065948486328,
2619
+ "learning_rate": 6.725807715869352e-06,
2620
+ "loss": 1.6875,
2621
+ "step": 373
2622
+ },
2623
+ {
2624
+ "epoch": 1.2262295081967214,
2625
+ "grad_norm": 13.26612663269043,
2626
+ "learning_rate": 6.709176177881228e-06,
2627
+ "loss": 1.6602,
2628
+ "step": 374
2629
+ },
2630
+ {
2631
+ "epoch": 1.2295081967213115,
2632
+ "grad_norm": 21.26793098449707,
2633
+ "learning_rate": 6.692523199196926e-06,
2634
+ "loss": 1.7227,
2635
+ "step": 375
2636
+ },
2637
+ {
2638
+ "epoch": 1.2327868852459016,
2639
+ "grad_norm": 12.17033576965332,
2640
+ "learning_rate": 6.675848988719102e-06,
2641
+ "loss": 1.6465,
2642
+ "step": 376
2643
+ },
2644
+ {
2645
+ "epoch": 1.2360655737704918,
2646
+ "grad_norm": 21.347976684570312,
2647
+ "learning_rate": 6.659153755616771e-06,
2648
+ "loss": 1.6289,
2649
+ "step": 377
2650
+ },
2651
+ {
2652
+ "epoch": 1.2393442622950819,
2653
+ "grad_norm": 19.131080627441406,
2654
+ "learning_rate": 6.64243770932265e-06,
2655
+ "loss": 1.6406,
2656
+ "step": 378
2657
+ },
2658
+ {
2659
+ "epoch": 1.2426229508196722,
2660
+ "grad_norm": 15.474773406982422,
2661
+ "learning_rate": 6.625701059530552e-06,
2662
+ "loss": 1.668,
2663
+ "step": 379
2664
+ },
2665
+ {
2666
+ "epoch": 1.2459016393442623,
2667
+ "grad_norm": 12.6391019821167,
2668
+ "learning_rate": 6.608944016192754e-06,
2669
+ "loss": 1.623,
2670
+ "step": 380
2671
+ },
2672
+ {
2673
+ "epoch": 1.2491803278688525,
2674
+ "grad_norm": 12.770636558532715,
2675
+ "learning_rate": 6.5921667895173515e-06,
2676
+ "loss": 1.6523,
2677
+ "step": 381
2678
+ },
2679
+ {
2680
+ "epoch": 1.2524590163934426,
2681
+ "grad_norm": 18.565126419067383,
2682
+ "learning_rate": 6.575369589965632e-06,
2683
+ "loss": 1.6641,
2684
+ "step": 382
2685
+ },
2686
+ {
2687
+ "epoch": 1.2557377049180327,
2688
+ "grad_norm": 13.343095779418945,
2689
+ "learning_rate": 6.558552628249435e-06,
2690
+ "loss": 1.6426,
2691
+ "step": 383
2692
+ },
2693
+ {
2694
+ "epoch": 1.2590163934426228,
2695
+ "grad_norm": 12.588839530944824,
2696
+ "learning_rate": 6.541716115328503e-06,
2697
+ "loss": 1.6172,
2698
+ "step": 384
2699
+ },
2700
+ {
2701
+ "epoch": 1.2622950819672132,
2702
+ "grad_norm": 13.01188850402832,
2703
+ "learning_rate": 6.524860262407835e-06,
2704
+ "loss": 1.5957,
2705
+ "step": 385
2706
+ },
2707
+ {
2708
+ "epoch": 1.2655737704918033,
2709
+ "grad_norm": 11.886181831359863,
2710
+ "learning_rate": 6.507985280935047e-06,
2711
+ "loss": 1.6445,
2712
+ "step": 386
2713
+ },
2714
+ {
2715
+ "epoch": 1.2688524590163934,
2716
+ "grad_norm": 18.22747039794922,
2717
+ "learning_rate": 6.491091382597705e-06,
2718
+ "loss": 1.5781,
2719
+ "step": 387
2720
+ },
2721
+ {
2722
+ "epoch": 1.2721311475409836,
2723
+ "grad_norm": 11.928472518920898,
2724
+ "learning_rate": 6.4741787793206824e-06,
2725
+ "loss": 1.5645,
2726
+ "step": 388
2727
+ },
2728
+ {
2729
+ "epoch": 1.275409836065574,
2730
+ "grad_norm": 15.313958168029785,
2731
+ "learning_rate": 6.457247683263494e-06,
2732
+ "loss": 1.6289,
2733
+ "step": 389
2734
+ },
2735
+ {
2736
+ "epoch": 1.278688524590164,
2737
+ "grad_norm": 14.239816665649414,
2738
+ "learning_rate": 6.440298306817637e-06,
2739
+ "loss": 1.6094,
2740
+ "step": 390
2741
+ },
2742
+ {
2743
+ "epoch": 1.2819672131147541,
2744
+ "grad_norm": 21.43814468383789,
2745
+ "learning_rate": 6.423330862603924e-06,
2746
+ "loss": 1.6113,
2747
+ "step": 391
2748
+ },
2749
+ {
2750
+ "epoch": 1.2852459016393443,
2751
+ "grad_norm": 14.304780006408691,
2752
+ "learning_rate": 6.406345563469819e-06,
2753
+ "loss": 1.625,
2754
+ "step": 392
2755
+ },
2756
+ {
2757
+ "epoch": 1.2885245901639344,
2758
+ "grad_norm": 15.105829238891602,
2759
+ "learning_rate": 6.38934262248677e-06,
2760
+ "loss": 1.5078,
2761
+ "step": 393
2762
+ },
2763
+ {
2764
+ "epoch": 1.2918032786885245,
2765
+ "grad_norm": 13.64173412322998,
2766
+ "learning_rate": 6.372322252947529e-06,
2767
+ "loss": 1.6094,
2768
+ "step": 394
2769
+ },
2770
+ {
2771
+ "epoch": 1.2950819672131146,
2772
+ "grad_norm": 12.446194648742676,
2773
+ "learning_rate": 6.355284668363479e-06,
2774
+ "loss": 1.6172,
2775
+ "step": 395
2776
+ },
2777
+ {
2778
+ "epoch": 1.298360655737705,
2779
+ "grad_norm": 12.818573951721191,
2780
+ "learning_rate": 6.3382300824619584e-06,
2781
+ "loss": 1.623,
2782
+ "step": 396
2783
+ },
2784
+ {
2785
+ "epoch": 1.301639344262295,
2786
+ "grad_norm": 16.565000534057617,
2787
+ "learning_rate": 6.321158709183579e-06,
2788
+ "loss": 1.6016,
2789
+ "step": 397
2790
+ },
2791
+ {
2792
+ "epoch": 1.3049180327868852,
2793
+ "grad_norm": 12.29269027709961,
2794
+ "learning_rate": 6.30407076267954e-06,
2795
+ "loss": 1.666,
2796
+ "step": 398
2797
+ },
2798
+ {
2799
+ "epoch": 1.3081967213114754,
2800
+ "grad_norm": 13.712056159973145,
2801
+ "learning_rate": 6.286966457308941e-06,
2802
+ "loss": 1.6289,
2803
+ "step": 399
2804
+ },
2805
+ {
2806
+ "epoch": 1.3114754098360657,
2807
+ "grad_norm": 12.996943473815918,
2808
+ "learning_rate": 6.2698460076360966e-06,
2809
+ "loss": 1.6035,
2810
+ "step": 400
2811
+ }
2812
+ ],
2813
+ "logging_steps": 1,
2814
+ "max_steps": 915,
2815
+ "num_input_tokens_seen": 0,
2816
+ "num_train_epochs": 3,
2817
+ "save_steps": 100,
2818
+ "stateful_callbacks": {
2819
+ "TrainerControl": {
2820
+ "args": {
2821
+ "should_epoch_stop": false,
2822
+ "should_evaluate": false,
2823
+ "should_log": false,
2824
+ "should_save": true,
2825
+ "should_training_stop": false
2826
+ },
2827
+ "attributes": {}
2828
+ }
2829
+ },
2830
+ "total_flos": 4.693001428061389e+16,
2831
+ "train_batch_size": 512,
2832
+ "trial_name": null,
2833
+ "trial_params": null
2834
+ }
checkpoint-400/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-500/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CLIPModel"
4
+ ],
5
+ "bos_token_id": 49406,
6
+ "eos_token_id": 49407,
7
+ "initializer_factor": 1.0,
8
+ "logit_scale_init_value": 2.6592,
9
+ "model_type": "clip",
10
+ "pad_token_id": 49407,
11
+ "projection_dim": 512,
12
+ "text_config": {
13
+ "attention_dropout": 0.0,
14
+ "bos_token_id": 0,
15
+ "dropout": 0.0,
16
+ "eos_token_id": 2,
17
+ "hidden_act": "quick_gelu",
18
+ "hidden_size": 512,
19
+ "initializer_factor": 1.0,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 2048,
22
+ "layer_norm_eps": 1e-05,
23
+ "max_position_embeddings": 77,
24
+ "model_type": "clip_text_model",
25
+ "num_attention_heads": 8,
26
+ "num_hidden_layers": 12,
27
+ "projection_dim": 512,
28
+ "torch_dtype": "float32",
29
+ "vocab_size": 49408
30
+ },
31
+ "torch_dtype": "bfloat16",
32
+ "transformers_version": "4.56.0.dev0",
33
+ "vision_config": {
34
+ "attention_dropout": 0.0,
35
+ "dropout": 0.0,
36
+ "hidden_act": "quick_gelu",
37
+ "hidden_size": 768,
38
+ "image_size": 224,
39
+ "initializer_factor": 1.0,
40
+ "initializer_range": 0.02,
41
+ "intermediate_size": 3072,
42
+ "layer_norm_eps": 1e-05,
43
+ "model_type": "clip_vision_model",
44
+ "num_attention_heads": 12,
45
+ "num_channels": 3,
46
+ "num_hidden_layers": 12,
47
+ "patch_size": 16,
48
+ "projection_dim": 512,
49
+ "torch_dtype": "float32"
50
+ }
51
+ }
checkpoint-500/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-500/preprocessor_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "processor_class": "CLIPProcessor",
23
+ "resample": 3,
24
+ "rescale_factor": 0.00392156862745098,
25
+ "size": {
26
+ "shortest_edge": 224
27
+ }
28
+ }
checkpoint-500/special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|startoftext|>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<|endoftext|>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<|endoftext|>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
checkpoint-500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-500/tokenizer_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "49406": {
5
+ "content": "<|startoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "49407": {
13
+ "content": "<|endoftext|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ }
20
+ },
21
+ "bos_token": "<|startoftext|>",
22
+ "clean_up_tokenization_spaces": false,
23
+ "do_lower_case": true,
24
+ "eos_token": "<|endoftext|>",
25
+ "errors": "replace",
26
+ "extra_special_tokens": {},
27
+ "model_max_length": 77,
28
+ "pad_token": "<|endoftext|>",
29
+ "processor_class": "CLIPProcessor",
30
+ "tokenizer_class": "CLIPTokenizer",
31
+ "unk_token": "<|endoftext|>"
32
+ }
checkpoint-500/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-600/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CLIPModel"
4
+ ],
5
+ "bos_token_id": 49406,
6
+ "eos_token_id": 49407,
7
+ "initializer_factor": 1.0,
8
+ "logit_scale_init_value": 2.6592,
9
+ "model_type": "clip",
10
+ "pad_token_id": 49407,
11
+ "projection_dim": 512,
12
+ "text_config": {
13
+ "attention_dropout": 0.0,
14
+ "bos_token_id": 0,
15
+ "dropout": 0.0,
16
+ "eos_token_id": 2,
17
+ "hidden_act": "quick_gelu",
18
+ "hidden_size": 512,
19
+ "initializer_factor": 1.0,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 2048,
22
+ "layer_norm_eps": 1e-05,
23
+ "max_position_embeddings": 77,
24
+ "model_type": "clip_text_model",
25
+ "num_attention_heads": 8,
26
+ "num_hidden_layers": 12,
27
+ "projection_dim": 512,
28
+ "torch_dtype": "float32",
29
+ "vocab_size": 49408
30
+ },
31
+ "torch_dtype": "bfloat16",
32
+ "transformers_version": "4.56.0.dev0",
33
+ "vision_config": {
34
+ "attention_dropout": 0.0,
35
+ "dropout": 0.0,
36
+ "hidden_act": "quick_gelu",
37
+ "hidden_size": 768,
38
+ "image_size": 224,
39
+ "initializer_factor": 1.0,
40
+ "initializer_range": 0.02,
41
+ "intermediate_size": 3072,
42
+ "layer_norm_eps": 1e-05,
43
+ "model_type": "clip_vision_model",
44
+ "num_attention_heads": 12,
45
+ "num_channels": 3,
46
+ "num_hidden_layers": 12,
47
+ "patch_size": 16,
48
+ "projection_dim": 512,
49
+ "torch_dtype": "float32"
50
+ }
51
+ }
checkpoint-600/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-600/special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|startoftext|>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<|endoftext|>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<|endoftext|>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
checkpoint-600/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-700/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CLIPModel"
4
+ ],
5
+ "bos_token_id": 49406,
6
+ "eos_token_id": 49407,
7
+ "initializer_factor": 1.0,
8
+ "logit_scale_init_value": 2.6592,
9
+ "model_type": "clip",
10
+ "pad_token_id": 49407,
11
+ "projection_dim": 512,
12
+ "text_config": {
13
+ "attention_dropout": 0.0,
14
+ "bos_token_id": 0,
15
+ "dropout": 0.0,
16
+ "eos_token_id": 2,
17
+ "hidden_act": "quick_gelu",
18
+ "hidden_size": 512,
19
+ "initializer_factor": 1.0,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 2048,
22
+ "layer_norm_eps": 1e-05,
23
+ "max_position_embeddings": 77,
24
+ "model_type": "clip_text_model",
25
+ "num_attention_heads": 8,
26
+ "num_hidden_layers": 12,
27
+ "projection_dim": 512,
28
+ "torch_dtype": "float32",
29
+ "vocab_size": 49408
30
+ },
31
+ "torch_dtype": "bfloat16",
32
+ "transformers_version": "4.56.0.dev0",
33
+ "vision_config": {
34
+ "attention_dropout": 0.0,
35
+ "dropout": 0.0,
36
+ "hidden_act": "quick_gelu",
37
+ "hidden_size": 768,
38
+ "image_size": 224,
39
+ "initializer_factor": 1.0,
40
+ "initializer_range": 0.02,
41
+ "intermediate_size": 3072,
42
+ "layer_norm_eps": 1e-05,
43
+ "model_type": "clip_vision_model",
44
+ "num_attention_heads": 12,
45
+ "num_channels": 3,
46
+ "num_hidden_layers": 12,
47
+ "patch_size": 16,
48
+ "projection_dim": 512,
49
+ "torch_dtype": "float32"
50
+ }
51
+ }
checkpoint-700/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-700/preprocessor_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "processor_class": "CLIPProcessor",
23
+ "resample": 3,
24
+ "rescale_factor": 0.00392156862745098,
25
+ "size": {
26
+ "shortest_edge": 224
27
+ }
28
+ }
checkpoint-700/special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|startoftext|>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<|endoftext|>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<|endoftext|>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
checkpoint-700/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-700/tokenizer_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "49406": {
5
+ "content": "<|startoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "49407": {
13
+ "content": "<|endoftext|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ }
20
+ },
21
+ "bos_token": "<|startoftext|>",
22
+ "clean_up_tokenization_spaces": false,
23
+ "do_lower_case": true,
24
+ "eos_token": "<|endoftext|>",
25
+ "errors": "replace",
26
+ "extra_special_tokens": {},
27
+ "model_max_length": 77,
28
+ "pad_token": "<|endoftext|>",
29
+ "processor_class": "CLIPProcessor",
30
+ "tokenizer_class": "CLIPTokenizer",
31
+ "unk_token": "<|endoftext|>"
32
+ }
checkpoint-700/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-700/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-800/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-900/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CLIPModel"
4
+ ],
5
+ "bos_token_id": 49406,
6
+ "eos_token_id": 49407,
7
+ "initializer_factor": 1.0,
8
+ "logit_scale_init_value": 2.6592,
9
+ "model_type": "clip",
10
+ "pad_token_id": 49407,
11
+ "projection_dim": 512,
12
+ "text_config": {
13
+ "attention_dropout": 0.0,
14
+ "bos_token_id": 0,
15
+ "dropout": 0.0,
16
+ "eos_token_id": 2,
17
+ "hidden_act": "quick_gelu",
18
+ "hidden_size": 512,
19
+ "initializer_factor": 1.0,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 2048,
22
+ "layer_norm_eps": 1e-05,
23
+ "max_position_embeddings": 77,
24
+ "model_type": "clip_text_model",
25
+ "num_attention_heads": 8,
26
+ "num_hidden_layers": 12,
27
+ "projection_dim": 512,
28
+ "torch_dtype": "float32",
29
+ "vocab_size": 49408
30
+ },
31
+ "torch_dtype": "bfloat16",
32
+ "transformers_version": "4.56.0.dev0",
33
+ "vision_config": {
34
+ "attention_dropout": 0.0,
35
+ "dropout": 0.0,
36
+ "hidden_act": "quick_gelu",
37
+ "hidden_size": 768,
38
+ "image_size": 224,
39
+ "initializer_factor": 1.0,
40
+ "initializer_range": 0.02,
41
+ "intermediate_size": 3072,
42
+ "layer_norm_eps": 1e-05,
43
+ "model_type": "clip_vision_model",
44
+ "num_attention_heads": 12,
45
+ "num_channels": 3,
46
+ "num_hidden_layers": 12,
47
+ "patch_size": 16,
48
+ "projection_dim": 512,
49
+ "torch_dtype": "float32"
50
+ }
51
+ }
checkpoint-900/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-900/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-900/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CLIPModel"
4
+ ],
5
+ "bos_token_id": 49406,
6
+ "eos_token_id": 49407,
7
+ "initializer_factor": 1.0,
8
+ "logit_scale_init_value": 2.6592,
9
+ "model_type": "clip",
10
+ "pad_token_id": 49407,
11
+ "projection_dim": 512,
12
+ "text_config": {
13
+ "attention_dropout": 0.0,
14
+ "bos_token_id": 0,
15
+ "dropout": 0.0,
16
+ "eos_token_id": 2,
17
+ "hidden_act": "quick_gelu",
18
+ "hidden_size": 512,
19
+ "initializer_factor": 1.0,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 2048,
22
+ "layer_norm_eps": 1e-05,
23
+ "max_position_embeddings": 77,
24
+ "model_type": "clip_text_model",
25
+ "num_attention_heads": 8,
26
+ "num_hidden_layers": 12,
27
+ "projection_dim": 512,
28
+ "torch_dtype": "float32",
29
+ "vocab_size": 49408
30
+ },
31
+ "torch_dtype": "bfloat16",
32
+ "transformers_version": "4.56.0.dev0",
33
+ "vision_config": {
34
+ "attention_dropout": 0.0,
35
+ "dropout": 0.0,
36
+ "hidden_act": "quick_gelu",
37
+ "hidden_size": 768,
38
+ "image_size": 224,
39
+ "initializer_factor": 1.0,
40
+ "initializer_range": 0.02,
41
+ "intermediate_size": 3072,
42
+ "layer_norm_eps": 1e-05,
43
+ "model_type": "clip_vision_model",
44
+ "num_attention_heads": 12,
45
+ "num_channels": 3,
46
+ "num_hidden_layers": 12,
47
+ "patch_size": 16,
48
+ "projection_dim": 512,
49
+ "torch_dtype": "float32"
50
+ }
51
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff