Hugging Face's logo

alexue4
/
text-normalization-ru-new

Text2Text Generation
Transformers
PyTorch
TensorBoard
Safetensors
Russian
t5
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Metrics Community
alexue4 commited on
Commit
627343b
•
1 Parent(s): e06c4d5

End of training

Browse files
Files changed (4) hide show
  1. README.md +55 -42
  2. pytorch_model.bin +1 -1
  3. trainer_state.json +965 -1531
  4. training_args.bin +1 -1
README.md CHANGED
@@ -15,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [cointegrated/rut5-small](https://huggingface.co/cointegrated/rut5-small) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
- - Loss: 0.0442
19
  - Mean Distance: 0
20
- - Max Distance: 25
21
 
22
  ## Model description
23
 
@@ -37,55 +37,68 @@ More information needed
37
 
38
  The following hyperparameters were used during training:
39
  - learning_rate: 0.001
40
- - train_batch_size: 15
41
- - eval_batch_size: 15
42
  - seed: 42
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
  - lr_scheduler_warmup_ratio: 0.1
46
- - num_epochs: 30
47
 
48
  ### Training results
49
 
50
  | Training Loss | Epoch | Step | Validation Loss | Max Distance | Mean Distance |
51
  |:-------------:|:-----:|:------:|:---------------:|:------------:|:-------------:|
52
- | 0.199 | 1.0 | 11526 | 0.8173 | 167 | 17 |
53
- | 0.1286 | 2.0 | 23052 | 0.5453 | 158 | 14 |
54
- | 0.0891 | 3.0 | 34578 | 0.3629 | 122 | 10 |
55
- | 0.0711 | 4.0 | 46104 | 0.4011 | 114 | 12 |
56
- | 0.0566 | 5.0 | 57630 | 0.2997 | 100 | 7 |
57
- | 0.0402 | 6.0 | 69156 | 0.1552 | 75 | 4 |
58
- | 0.0348 | 7.0 | 80682 | 0.1513 | 79 | 3 |
59
- | 0.0302 | 8.0 | 92208 | 0.1452 | 76 | 3 |
60
- | 0.0223 | 9.0 | 103734 | 0.0866 | 76 | 1 |
61
- | 0.0202 | 10.0 | 115260 | 0.1091 | 71 | 2 |
62
- | 0.0175 | 11.0 | 126786 | 0.0655 | 66 | 1 |
63
- | 0.014 | 12.0 | 138312 | 0.0474 | 44 | 0 |
64
- | 0.0122 | 13.0 | 149838 | 0.0515 | 42 | 0 |
65
- | 0.0117 | 14.0 | 161364 | 0.0479 | 30 | 0 |
66
- | 0.0093 | 15.0 | 172890 | 0.0565 | 56 | 0 |
67
- | 0.0085 | 16.0 | 184416 | 0.0472 | 34 | 0 |
68
- | 0.0075 | 17.0 | 195942 | 0.0420 | 28 | 0 |
69
- | 0.0059 | 18.0 | 207468 | 0.0415 | 32 | 0 |
70
- | 0.0054 | 19.0 | 218994 | 0.0406 | 28 | 0 |
71
- | 0.0046 | 20.0 | 230520 | 0.0393 | 24 | 0 |
72
- | 0.004 | 21.0 | 242046 | 0.0417 | 24 | 0 |
73
- | 0.0034 | 22.0 | 253572 | 0.0403 | 18 | 0 |
74
- | 0.0029 | 23.0 | 265098 | 0.0422 | 21 | 0 |
75
- | 0.0024 | 24.0 | 276624 | 0.0410 | 21 | 0 |
76
- | 0.002 | 25.0 | 288150 | 0.0435 | 15 | 0 |
77
- | 0.0016 | 26.0 | 299676 | 0.0452 | 15 | 0 |
78
- | 0.0013 | 27.0 | 311202 | 0.0414 | 14 | 0 |
79
- | 0.0012 | 28.0 | 322728 | 0.0439 | 14 | 0 |
80
- | 0.001 | 29.0 | 334254 | 0.0444 | 15 | 0 |
81
- | 0.0026 | 30.0 | 345780 | 0.0427 | 19 | 0 |
82
- | 0.0077 | 24.0 | 368808 | 0.0495 | 27 | 0 |
83
- | 0.0083 | 25.0 | 384175 | 0.0446 | 37 | 0 |
84
- | 0.0078 | 26.0 | 399542 | 0.0481 | 47 | 0 |
85
- | 0.006 | 27.0 | 414909 | 0.0424 | 37 | 0 |
86
- | 0.0056 | 28.0 | 430276 | 0.0439 | 22 | 0 |
87
- | 0.0054 | 29.0 | 445643 | 0.0481 | 23 | 0 |
88
- | 0.004 | 30.0 | 461010 | 0.0442 | 0 | 25 |
 
 
 
 
 
 
 
 
 
 
 
 
 
89
 
90
 
91
  ### Framework versions
 
15
 
16
  This model is a fine-tuned version of [cointegrated/rut5-small](https://huggingface.co/cointegrated/rut5-small) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.0664
19
  - Mean Distance: 0
20
+ - Max Distance: 18
21
 
22
  ## Model description
23
 
 
37
 
38
  The following hyperparameters were used during training:
39
  - learning_rate: 0.001
40
+ - train_batch_size: 20
41
+ - eval_batch_size: 20
42
  - seed: 42
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
  - lr_scheduler_warmup_ratio: 0.1
46
+ - num_epochs: 50
47
 
48
  ### Training results
49
 
50
  | Training Loss | Epoch | Step | Validation Loss | Max Distance | Mean Distance |
51
  |:-------------:|:-----:|:------:|:---------------:|:------------:|:-------------:|
52
+ | 0.2198 | 1.0 | 15301 | 0.7378 | 206 | 16 |
53
+ | 0.1431 | 2.0 | 23102 | 0.9838 | 183 | 22 |
54
+ | 0.1014 | 3.0 | 34653 | 0.5183 | 122 | 13 |
55
+ | 0.0818 | 4.0 | 46204 | 0.3151 | 94 | 7 |
56
+ | 0.07 | 5.0 | 57755 | 0.4898 | 100 | 12 |
57
+ | 0.0555 | 6.0 | 69306 | 0.3159 | 99 | 8 |
58
+ | 0.0442 | 7.0 | 80857 | 0.2628 | 91 | 6 |
59
+ | 0.0358 | 8.0 | 92408 | 0.2342 | 69 | 5 |
60
+ | 0.0299 | 9.0 | 103959 | 0.1100 | 63 | 2 |
61
+ | 0.0252 | 10.0 | 115510 | 0.0934 | 51 | 1 |
62
+ | 0.022 | 11.0 | 127061 | 0.1330 | 53 | 3 |
63
+ | 0.019 | 12.0 | 138612 | 0.1217 | 51 | 2 |
64
+ | 0.0163 | 13.0 | 150163 | 0.0772 | 37 | 1 |
65
+ | 0.0145 | 14.0 | 161714 | 0.1268 | 65 | 2 |
66
+ | 0.0131 | 15.0 | 173265 | 0.0883 | 34 | 1 |
67
+ | 0.0115 | 16.0 | 184816 | 0.0709 | 30 | 1 |
68
+ | 0.0104 | 17.0 | 196367 | 0.0669 | 38 | 1 |
69
+ | 0.0094 | 18.0 | 207918 | 0.0667 | 29 | 0 |
70
+ | 0.0086 | 19.0 | 219469 | 0.0641 | 27 | 0 |
71
+ | 0.0078 | 20.0 | 231020 | 0.0651 | 0 | 22 |
72
+ | 0.007 | 21.0 | 242571 | 0.0651 | 0 | 33 |
73
+ | 0.0064 | 22.0 | 254122 | 0.0605 | 0 | 31 |
74
+ | 0.0059 | 23.0 | 265673 | 0.0561 | 0 | 25 |
75
+ | 0.0053 | 24.0 | 277224 | 0.0567 | 0 | 36 |
76
+ | 0.0049 | 25.0 | 288775 | 0.0557 | 0 | 25 |
77
+ | 0.0045 | 26.0 | 300326 | 0.0523 | 0 | 26 |
78
+ | 0.0041 | 27.0 | 311877 | 0.0577 | 0 | 25 |
79
+ | 0.0038 | 28.0 | 323428 | 0.0577 | 0 | 32 |
80
+ | 0.0033 | 29.0 | 334979 | 0.0575 | 0 | 20 |
81
+ | 0.0031 | 30.0 | 346530 | 0.0562 | 0 | 16 |
82
+ | 0.0026 | 31.0 | 358081 | 0.0596 | 0 | 15 |
83
+ | 0.0026 | 32.0 | 369632 | 0.0617 | 0 | 30 |
84
+ | 0.0024 | 33.0 | 381183 | 0.0572 | 0 | 20 |
85
+ | 0.0022 | 34.0 | 392734 | 0.0557 | 0 | 19 |
86
+ | 0.002 | 35.0 | 404285 | 0.0612 | 0 | 17 |
87
+ | 0.0017 | 36.0 | 415836 | 0.0576 | 0 | 14 |
88
+ | 0.0016 | 37.0 | 427387 | 0.0606 | 0 | 16 |
89
+ | 0.0015 | 38.0 | 438938 | 0.0570 | 0 | 24 |
90
+ | 0.0013 | 39.0 | 450489 | 0.0594 | 0 | 24 |
91
+ | 0.0012 | 40.0 | 462040 | 0.0603 | 0 | 24 |
92
+ | 0.001 | 41.0 | 473591 | 0.0642 | 0 | 20 |
93
+ | 0.0009 | 42.0 | 485142 | 0.0638 | 0 | 29 |
94
+ | 0.0008 | 43.0 | 496693 | 0.0659 | 0 | 29 |
95
+ | 0.0007 | 44.0 | 508244 | 0.0672 | 0 | 29 |
96
+ | 0.0007 | 45.0 | 519795 | 0.0662 | 0 | 19 |
97
+ | 0.0006 | 46.0 | 531346 | 0.0666 | 0 | 24 |
98
+ | 0.0006 | 47.0 | 542897 | 0.0690 | 0 | 22 |
99
+ | 0.0005 | 48.0 | 554448 | 0.0670 | 0 | 15 |
100
+ | 0.0005 | 49.0 | 565999 | 0.0662 | 0 | 18 |
101
+ | 0.0004 | 50.0 | 577550 | 0.0664 | 0 | 18 |
102
 
103
 
104
  ### Framework versions
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1af9c44d78cc87dce6c9af177a92980e3657aa89417f862a6decf4575d013140
3
  size 258643461
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cdc958f1006ef7ccf3b2642547782a3b93fe5e10a96ecc3ad1689478196ee825
3
  size 258643461
trainer_state.json CHANGED
@@ -1,2000 +1,1434 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 30.0,
5
  "eval_steps": 500,
6
- "global_step": 461010,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 0.0,
13
- "learning_rate": 2.8920122621319915e-08,
14
- "loss": 14.3504,
15
  "step": 1
16
  },
17
  {
18
- "epoch": 0.15,
19
- "learning_rate": 5.000289201226213e-05,
20
- "loss": 2.1001,
21
- "step": 1729
22
  },
23
  {
24
- "epoch": 0.3,
25
- "learning_rate": 0.00010000578402452426,
26
- "loss": 0.3756,
27
- "step": 3458
28
- },
29
- {
30
- "epoch": 0.45,
31
- "learning_rate": 0.0001500086760367864,
32
- "loss": 0.2527,
33
- "step": 5187
34
- },
35
- {
36
- "epoch": 0.6,
37
- "learning_rate": 0.00020001156804904852,
38
- "loss": 0.2076,
39
- "step": 6916
40
  },
41
  {
42
  "epoch": 0.75,
43
- "learning_rate": 0.00025001446006131067,
44
- "loss": 0.1864,
45
- "step": 8645
46
- },
47
- {
48
- "epoch": 0.9,
49
- "learning_rate": 0.0003000173520735728,
50
- "loss": 0.199,
51
- "step": 10374
52
  },
53
  {
54
  "epoch": 1.0,
55
- "eval_loss": 0.8173184990882874,
56
- "eval_max_distance": 167,
57
- "eval_mean_distance": 17,
58
- "eval_runtime": 64.0638,
59
- "eval_samples_per_second": 15.609,
60
- "eval_steps_per_second": 0.78,
61
- "step": 11526
62
- },
63
- {
64
- "epoch": 1.05,
65
- "learning_rate": 0.0003500202440858349,
66
- "loss": 0.2481,
67
- "step": 12103
68
  },
69
  {
70
- "epoch": 1.2,
71
- "learning_rate": 0.00040002313609809704,
72
- "loss": 0.1244,
73
- "step": 13832
74
  },
75
  {
76
- "epoch": 1.35,
77
- "learning_rate": 0.0004500260281103592,
78
- "loss": 0.1055,
79
- "step": 15561
80
  },
81
  {
82
- "epoch": 1.5,
83
- "learning_rate": 0.0005000289201226213,
84
- "loss": 0.102,
85
- "step": 17290
86
- },
87
- {
88
- "epoch": 1.65,
89
- "learning_rate": 0.0005500318121348835,
90
- "loss": 0.102,
91
- "step": 19019
92
- },
93
- {
94
- "epoch": 1.8,
95
- "learning_rate": 0.0006000347041471456,
96
- "loss": 0.1083,
97
- "step": 20748
98
- },
99
- {
100
- "epoch": 1.95,
101
- "learning_rate": 0.0006500375961594078,
102
- "loss": 0.1286,
103
- "step": 22477
104
  },
105
  {
106
  "epoch": 2.0,
107
- "eval_loss": 0.5452634692192078,
108
- "eval_max_distance": 158,
109
- "eval_mean_distance": 14,
110
- "eval_runtime": 30.4268,
111
- "eval_samples_per_second": 32.866,
112
- "eval_steps_per_second": 1.643,
113
- "step": 23052
114
- },
115
- {
116
- "epoch": 2.1,
117
- "learning_rate": 0.0007000404881716698,
118
- "loss": 0.1449,
119
- "step": 24206
120
  },
121
  {
122
- "epoch": 2.25,
123
- "learning_rate": 0.000750043380183932,
124
- "loss": 0.0747,
125
- "step": 25935
126
  },
127
  {
128
- "epoch": 2.4,
129
- "learning_rate": 0.0008000462721961941,
130
- "loss": 0.0744,
131
- "step": 27664
132
  },
133
  {
134
- "epoch": 2.55,
135
- "learning_rate": 0.0008500491642084563,
136
- "loss": 0.0742,
137
- "step": 29393
138
- },
139
- {
140
- "epoch": 2.7,
141
- "learning_rate": 0.0009000520562207184,
142
- "loss": 0.0792,
143
- "step": 31122
144
- },
145
- {
146
- "epoch": 2.85,
147
- "learning_rate": 0.0009500549482329805,
148
- "loss": 0.0891,
149
- "step": 32851
150
  },
151
  {
152
  "epoch": 3.0,
153
- "eval_loss": 0.3629104495048523,
154
  "eval_max_distance": 122,
155
- "eval_mean_distance": 10,
156
- "eval_runtime": 35.1149,
157
- "eval_samples_per_second": 28.478,
158
- "eval_steps_per_second": 1.424,
159
- "step": 34578
160
  },
161
  {
162
- "epoch": 3.0,
163
- "learning_rate": 0.0009999935733060843,
164
- "loss": 0.1079,
165
- "step": 34580
166
- },
167
- {
168
- "epoch": 3.15,
169
- "learning_rate": 0.000994437696415833,
170
- "loss": 0.0962,
171
- "step": 36309
172
  },
173
  {
174
- "epoch": 3.3,
175
- "learning_rate": 0.0009888818195255813,
176
- "loss": 0.059,
177
- "step": 38038
178
  },
179
  {
180
- "epoch": 3.45,
181
- "learning_rate": 0.0009833259426353302,
182
- "loss": 0.0576,
183
- "step": 39767
184
- },
185
- {
186
- "epoch": 3.6,
187
- "learning_rate": 0.0009777700657450789,
188
- "loss": 0.058,
189
- "step": 41496
190
- },
191
- {
192
- "epoch": 3.75,
193
- "learning_rate": 0.0009722141888548275,
194
- "loss": 0.0611,
195
- "step": 43225
196
- },
197
- {
198
- "epoch": 3.9,
199
- "learning_rate": 0.0009666583119645761,
200
- "loss": 0.0711,
201
- "step": 44954
202
  },
203
  {
204
  "epoch": 4.0,
205
- "eval_loss": 0.4011004865169525,
206
- "eval_max_distance": 114,
207
- "eval_mean_distance": 12,
208
- "eval_runtime": 24.9859,
209
- "eval_samples_per_second": 40.022,
210
- "eval_steps_per_second": 2.001,
211
- "step": 46104
212
- },
213
- {
214
- "epoch": 4.05,
215
- "learning_rate": 0.0009611024350743247,
216
- "loss": 0.0991,
217
- "step": 46683
218
- },
219
- {
220
- "epoch": 4.2,
221
- "learning_rate": 0.0009555465581840734,
222
- "loss": 0.0481,
223
- "step": 48412
224
- },
225
- {
226
- "epoch": 4.35,
227
- "learning_rate": 0.0009499906812938221,
228
- "loss": 0.044,
229
- "step": 50141
230
- },
231
- {
232
- "epoch": 4.5,
233
- "learning_rate": 0.0009444348044035706,
234
- "loss": 0.0435,
235
- "step": 51870
236
  },
237
  {
238
- "epoch": 4.65,
239
- "learning_rate": 0.0009388789275133194,
240
- "loss": 0.0454,
241
- "step": 53599
242
  },
243
  {
244
- "epoch": 4.8,
245
- "learning_rate": 0.000933323050623068,
246
- "loss": 0.0483,
247
- "step": 55328
248
  },
249
  {
250
- "epoch": 4.95,
251
- "learning_rate": 0.0009277671737328166,
252
- "loss": 0.0566,
253
- "step": 57057
254
  },
255
  {
256
  "epoch": 5.0,
257
- "eval_loss": 0.29974234104156494,
258
  "eval_max_distance": 100,
259
- "eval_mean_distance": 7,
260
- "eval_runtime": 25.609,
261
- "eval_samples_per_second": 39.049,
262
- "eval_steps_per_second": 1.952,
263
- "step": 57630
264
- },
265
- {
266
- "epoch": 5.1,
267
- "learning_rate": 0.0009222112968425653,
268
- "loss": 0.0724,
269
- "step": 58786
270
- },
271
- {
272
- "epoch": 5.25,
273
- "learning_rate": 0.000916655419952314,
274
- "loss": 0.0364,
275
- "step": 60515
276
- },
277
- {
278
- "epoch": 5.4,
279
- "learning_rate": 0.0009110995430620625,
280
- "loss": 0.035,
281
- "step": 62244
282
- },
283
- {
284
- "epoch": 5.55,
285
- "learning_rate": 0.0009055436661718113,
286
- "loss": 0.0353,
287
- "step": 63973
288
  },
289
  {
290
- "epoch": 5.7,
291
- "learning_rate": 0.0008999877892815599,
292
- "loss": 0.0364,
293
- "step": 65702
294
  },
295
  {
296
- "epoch": 5.85,
297
- "learning_rate": 0.0008944319123913086,
298
- "loss": 0.0402,
299
- "step": 67431
300
  },
301
  {
302
- "epoch": 6.0,
303
- "eval_loss": 0.15522713959217072,
304
- "eval_max_distance": 75,
305
- "eval_mean_distance": 4,
306
- "eval_runtime": 24.8618,
307
- "eval_samples_per_second": 40.222,
308
- "eval_steps_per_second": 2.011,
309
- "step": 69156
310
  },
311
  {
312
  "epoch": 6.0,
313
- "learning_rate": 0.0008888760355010572,
314
- "loss": 0.0506,
315
- "step": 69160
316
- },
317
- {
318
- "epoch": 6.15,
319
- "learning_rate": 0.0008833201586108059,
320
- "loss": 0.0514,
321
- "step": 70889
322
- },
323
- {
324
- "epoch": 6.3,
325
- "learning_rate": 0.0008777642817205546,
326
- "loss": 0.0298,
327
- "step": 72618
328
  },
329
  {
330
- "epoch": 6.45,
331
- "learning_rate": 0.0008722084048303031,
332
- "loss": 0.0287,
333
- "step": 74347
334
  },
335
  {
336
- "epoch": 6.6,
337
- "learning_rate": 0.0008666525279400518,
338
- "loss": 0.0289,
339
- "step": 76076
340
  },
341
  {
342
- "epoch": 6.75,
343
- "learning_rate": 0.0008610966510498005,
344
- "loss": 0.0303,
345
- "step": 77805
346
- },
347
- {
348
- "epoch": 6.9,
349
- "learning_rate": 0.0008555407741595491,
350
- "loss": 0.0348,
351
- "step": 79534
352
  },
353
  {
354
  "epoch": 7.0,
355
- "eval_loss": 0.15130603313446045,
356
- "eval_max_distance": 79,
357
- "eval_mean_distance": 3,
358
- "eval_runtime": 25.5455,
359
- "eval_samples_per_second": 39.146,
360
- "eval_steps_per_second": 1.957,
361
- "step": 80682
362
- },
363
- {
364
- "epoch": 7.05,
365
- "learning_rate": 0.0008499848972692977,
366
- "loss": 0.0527,
367
- "step": 81263
368
- },
369
- {
370
- "epoch": 7.2,
371
- "learning_rate": 0.0008444290203790465,
372
- "loss": 0.0293,
373
- "step": 82992
374
- },
375
- {
376
- "epoch": 7.35,
377
- "learning_rate": 0.000838873143488795,
378
- "loss": 0.0247,
379
- "step": 84721
380
- },
381
- {
382
- "epoch": 7.5,
383
- "learning_rate": 0.0008333172665985437,
384
- "loss": 0.024,
385
- "step": 86450
386
  },
387
  {
388
- "epoch": 7.65,
389
- "learning_rate": 0.0008277613897082924,
390
- "loss": 0.0243,
391
- "step": 88179
392
  },
393
  {
394
- "epoch": 7.8,
395
- "learning_rate": 0.000822205512818041,
396
- "loss": 0.0263,
397
- "step": 89908
398
  },
399
  {
400
  "epoch": 7.95,
401
- "learning_rate": 0.0008166496359277897,
402
- "loss": 0.0302,
403
- "step": 91637
404
  },
405
  {
406
  "epoch": 8.0,
407
- "eval_loss": 0.14522188901901245,
408
- "eval_max_distance": 76,
409
- "eval_mean_distance": 3,
410
- "eval_runtime": 25.6271,
411
- "eval_samples_per_second": 39.021,
412
- "eval_steps_per_second": 1.951,
413
- "step": 92208
414
- },
415
- {
416
- "epoch": 8.1,
417
- "learning_rate": 0.0008110937590375384,
418
- "loss": 0.0445,
419
- "step": 93366
420
- },
421
- {
422
- "epoch": 8.25,
423
- "learning_rate": 0.0008055378821472869,
424
- "loss": 0.0229,
425
- "step": 95095
426
  },
427
  {
428
- "epoch": 8.4,
429
- "learning_rate": 0.0007999820052570357,
430
- "loss": 0.0207,
431
- "step": 96824
432
  },
433
  {
434
- "epoch": 8.55,
435
- "learning_rate": 0.0007944261283667843,
436
- "loss": 0.0203,
437
- "step": 98553
438
  },
439
  {
440
- "epoch": 8.7,
441
- "learning_rate": 0.000788870251476533,
442
- "loss": 0.021,
443
- "step": 100282
444
- },
445
- {
446
- "epoch": 8.85,
447
- "learning_rate": 0.0007833143745862816,
448
- "loss": 0.0223,
449
- "step": 102011
450
- },
451
- {
452
- "epoch": 9.0,
453
- "eval_loss": 0.08658243715763092,
454
- "eval_max_distance": 76,
455
- "eval_mean_distance": 1,
456
- "eval_runtime": 23.498,
457
- "eval_samples_per_second": 42.557,
458
- "eval_steps_per_second": 2.128,
459
- "step": 103734
460
  },
461
  {
462
  "epoch": 9.0,
463
- "learning_rate": 0.0007777584976960303,
464
- "loss": 0.0284,
465
- "step": 103740
466
- },
467
- {
468
- "epoch": 9.15,
469
- "learning_rate": 0.0007722026208057789,
470
- "loss": 0.034,
471
- "step": 105469
472
- },
473
- {
474
- "epoch": 9.3,
475
- "learning_rate": 0.0007666467439155275,
476
- "loss": 0.0193,
477
- "step": 107198
478
- },
479
- {
480
- "epoch": 9.45,
481
- "learning_rate": 0.0007610908670252762,
482
- "loss": 0.0175,
483
- "step": 108927
484
- },
485
- {
486
- "epoch": 9.6,
487
- "learning_rate": 0.0007555349901350248,
488
- "loss": 0.0172,
489
- "step": 110656
490
- },
491
- {
492
- "epoch": 9.75,
493
- "learning_rate": 0.0007499791132447735,
494
- "loss": 0.018,
495
- "step": 112385
496
- },
497
- {
498
- "epoch": 9.9,
499
- "learning_rate": 0.0007444232363545221,
500
- "loss": 0.0202,
501
- "step": 114114
502
- },
503
- {
504
- "epoch": 10.0,
505
- "eval_loss": 0.10908353328704834,
506
- "eval_max_distance": 71,
507
  "eval_mean_distance": 2,
508
- "eval_runtime": 25.0909,
509
- "eval_samples_per_second": 39.855,
510
- "eval_steps_per_second": 1.993,
511
- "step": 115260
512
- },
513
- {
514
- "epoch": 10.05,
515
- "learning_rate": 0.0007388673594642709,
516
- "loss": 0.0342,
517
- "step": 115843
518
- },
519
- {
520
- "epoch": 10.2,
521
- "learning_rate": 0.0007333114825740194,
522
- "loss": 0.0202,
523
- "step": 117572
524
- },
525
- {
526
- "epoch": 10.35,
527
- "learning_rate": 0.0007277556056837681,
528
- "loss": 0.0159,
529
- "step": 119301
530
- },
531
- {
532
- "epoch": 10.5,
533
- "learning_rate": 0.0007221997287935168,
534
- "loss": 0.0149,
535
- "step": 121030
536
  },
537
  {
538
- "epoch": 10.65,
539
- "learning_rate": 0.0007166438519032654,
540
- "loss": 0.015,
541
- "step": 122759
542
  },
543
  {
544
- "epoch": 10.8,
545
- "learning_rate": 0.000711087975013014,
546
- "loss": 0.0155,
547
- "step": 124488
548
  },
549
  {
550
- "epoch": 10.95,
551
- "learning_rate": 0.0007055320981227628,
552
- "loss": 0.0175,
553
- "step": 126217
554
  },
555
  {
556
- "epoch": 11.0,
557
- "eval_loss": 0.06553788483142853,
558
- "eval_max_distance": 66,
559
  "eval_mean_distance": 1,
560
- "eval_runtime": 22.988,
561
- "eval_samples_per_second": 43.501,
562
- "eval_steps_per_second": 2.175,
563
- "step": 126786
564
  },
565
  {
566
- "epoch": 11.1,
567
- "learning_rate": 0.0006999762212325114,
568
- "loss": 0.0305,
569
- "step": 127946
570
  },
571
  {
572
- "epoch": 11.25,
573
- "learning_rate": 0.00069442034434226,
574
- "loss": 0.0157,
575
- "step": 129675
576
  },
577
  {
578
- "epoch": 11.4,
579
- "learning_rate": 0.0006888644674520087,
580
- "loss": 0.0134,
581
- "step": 131404
582
  },
583
  {
584
- "epoch": 11.55,
585
- "learning_rate": 0.0006833085905617574,
586
- "loss": 0.0124,
587
- "step": 133133
 
 
 
 
588
  },
589
  {
590
- "epoch": 11.7,
591
- "learning_rate": 0.0006777527136715059,
592
- "loss": 0.0131,
593
- "step": 134862
594
  },
595
  {
596
- "epoch": 11.85,
597
- "learning_rate": 0.0006721968367812547,
598
- "loss": 0.014,
599
- "step": 136591
600
  },
601
  {
602
- "epoch": 12.0,
603
- "eval_loss": 0.04735955968499184,
604
- "eval_max_distance": 44,
605
- "eval_mean_distance": 0,
606
- "eval_runtime": 22.5236,
607
- "eval_samples_per_second": 44.398,
608
- "eval_steps_per_second": 2.22,
609
- "step": 138312
610
  },
611
  {
612
  "epoch": 12.0,
613
- "learning_rate": 0.0006666409598910033,
614
- "loss": 0.0182,
615
- "step": 138320
616
- },
617
- {
618
- "epoch": 12.15,
619
- "learning_rate": 0.000661085083000752,
620
- "loss": 0.025,
621
- "step": 140049
622
- },
623
- {
624
- "epoch": 12.3,
625
- "learning_rate": 0.0006555292061105006,
626
- "loss": 0.0128,
627
- "step": 141778
628
- },
629
- {
630
- "epoch": 12.45,
631
- "learning_rate": 0.0006499733292202492,
632
- "loss": 0.0114,
633
- "step": 143507
634
  },
635
  {
636
- "epoch": 12.6,
637
- "learning_rate": 0.0006444174523299979,
638
- "loss": 0.011,
639
- "step": 145236
640
  },
641
  {
642
- "epoch": 12.75,
643
- "learning_rate": 0.0006388615754397465,
644
- "loss": 0.0117,
645
- "step": 146965
646
  },
647
  {
648
- "epoch": 12.9,
649
- "learning_rate": 0.0006333056985494952,
650
- "loss": 0.0122,
651
- "step": 148694
652
  },
653
  {
654
  "epoch": 13.0,
655
- "eval_loss": 0.05152251571416855,
656
- "eval_max_distance": 42,
657
- "eval_mean_distance": 0,
658
- "eval_runtime": 23.898,
659
- "eval_samples_per_second": 41.845,
660
- "eval_steps_per_second": 2.092,
661
- "step": 149838
662
- },
663
- {
664
- "epoch": 13.05,
665
- "learning_rate": 0.0006277498216592438,
666
- "loss": 0.0241,
667
- "step": 150423
668
- },
669
- {
670
- "epoch": 13.2,
671
- "learning_rate": 0.0006221939447689925,
672
- "loss": 0.0148,
673
- "step": 152152
674
- },
675
- {
676
- "epoch": 13.35,
677
- "learning_rate": 0.0006166380678787411,
678
- "loss": 0.0106,
679
- "step": 153881
680
- },
681
- {
682
- "epoch": 13.5,
683
- "learning_rate": 0.0006110821909884899,
684
- "loss": 0.0096,
685
- "step": 155610
686
  },
687
  {
688
- "epoch": 13.65,
689
- "learning_rate": 0.0006055263140982384,
690
- "loss": 0.0098,
691
- "step": 157339
692
  },
693
  {
694
- "epoch": 13.8,
695
- "learning_rate": 0.0005999704372079872,
696
- "loss": 0.0104,
697
- "step": 159068
698
  },
699
  {
700
- "epoch": 13.95,
701
- "learning_rate": 0.0005944145603177358,
702
- "loss": 0.0117,
703
- "step": 160797
704
  },
705
  {
706
  "epoch": 14.0,
707
- "eval_loss": 0.047906968742609024,
708
- "eval_max_distance": 30,
709
- "eval_mean_distance": 0,
710
- "eval_runtime": 23.3531,
711
- "eval_samples_per_second": 42.821,
712
- "eval_steps_per_second": 2.141,
713
- "step": 161364
714
- },
715
- {
716
- "epoch": 14.1,
717
- "learning_rate": 0.0005888586834274844,
718
- "loss": 0.0224,
719
- "step": 162526
720
- },
721
- {
722
- "epoch": 14.25,
723
- "learning_rate": 0.0005833028065372331,
724
- "loss": 0.0111,
725
- "step": 164255
726
- },
727
- {
728
- "epoch": 14.4,
729
- "learning_rate": 0.0005777469296469818,
730
- "loss": 0.009,
731
- "step": 165984
732
  },
733
  {
734
- "epoch": 14.55,
735
- "learning_rate": 0.0005721910527567303,
736
- "loss": 0.0086,
737
- "step": 167713
738
  },
739
  {
740
- "epoch": 14.7,
741
- "learning_rate": 0.000566635175866479,
742
- "loss": 0.0088,
743
- "step": 169442
744
  },
745
  {
746
- "epoch": 14.85,
747
- "learning_rate": 0.0005610792989762277,
748
- "loss": 0.0093,
749
- "step": 171171
750
- },
751
- {
752
- "epoch": 15.0,
753
- "eval_loss": 0.05651288107037544,
754
- "eval_max_distance": 56,
755
- "eval_mean_distance": 0,
756
- "eval_runtime": 23.6545,
757
- "eval_samples_per_second": 42.275,
758
- "eval_steps_per_second": 2.114,
759
- "step": 172890
760
  },
761
  {
762
  "epoch": 15.0,
763
- "learning_rate": 0.0005555234220859762,
764
- "loss": 0.0124,
765
- "step": 172900
766
- },
767
- {
768
- "epoch": 15.15,
769
- "learning_rate": 0.000549967545195725,
770
- "loss": 0.0181,
771
- "step": 174629
772
- },
773
- {
774
- "epoch": 15.3,
775
- "learning_rate": 0.0005444116683054736,
776
- "loss": 0.0091,
777
- "step": 176358
778
- },
779
- {
780
- "epoch": 15.45,
781
- "learning_rate": 0.0005388557914152222,
782
- "loss": 0.0075,
783
- "step": 178087
784
  },
785
  {
786
- "epoch": 15.6,
787
- "learning_rate": 0.0005332999145249709,
788
- "loss": 0.0075,
789
- "step": 179816
790
  },
791
  {
792
- "epoch": 15.75,
793
- "learning_rate": 0.0005277440376347196,
794
- "loss": 0.0075,
795
- "step": 181545
796
  },
797
  {
798
  "epoch": 15.9,
799
- "learning_rate": 0.0005221881607444683,
800
- "loss": 0.0085,
801
- "step": 183274
802
  },
803
  {
804
  "epoch": 16.0,
805
- "eval_loss": 0.047154366970062256,
806
- "eval_max_distance": 34,
807
- "eval_mean_distance": 0,
808
- "eval_runtime": 23.554,
809
- "eval_samples_per_second": 42.456,
810
- "eval_steps_per_second": 2.123,
811
- "step": 184416
812
- },
813
- {
814
- "epoch": 16.05,
815
- "learning_rate": 0.0005166322838542169,
816
- "loss": 0.0169,
817
- "step": 185003
818
- },
819
- {
820
- "epoch": 16.2,
821
- "learning_rate": 0.0005110764069639655,
822
- "loss": 0.0104,
823
- "step": 186732
824
- },
825
- {
826
- "epoch": 16.35,
827
- "learning_rate": 0.0005055205300737143,
828
- "loss": 0.0072,
829
- "step": 188461
830
- },
831
- {
832
- "epoch": 16.5,
833
- "learning_rate": 0.0004999646531834628,
834
- "loss": 0.0068,
835
- "step": 190190
836
  },
837
  {
838
- "epoch": 16.65,
839
- "learning_rate": 0.0004944087762932115,
840
- "loss": 0.0064,
841
- "step": 191919
842
  },
843
  {
844
- "epoch": 16.8,
845
- "learning_rate": 0.0004888528994029601,
846
- "loss": 0.0068,
847
- "step": 193648
848
  },
849
  {
850
- "epoch": 16.95,
851
- "learning_rate": 0.0004832970225127088,
852
- "loss": 0.0075,
853
- "step": 195377
854
  },
855
  {
856
  "epoch": 17.0,
857
- "eval_loss": 0.04200902581214905,
858
- "eval_max_distance": 28,
859
- "eval_mean_distance": 0,
860
- "eval_runtime": 22.1607,
861
- "eval_samples_per_second": 45.125,
862
- "eval_steps_per_second": 2.256,
863
- "step": 195942
864
- },
865
- {
866
- "epoch": 17.1,
867
- "learning_rate": 0.00047774114562245746,
868
- "loss": 0.0162,
869
- "step": 197106
870
- },
871
- {
872
- "epoch": 17.25,
873
- "learning_rate": 0.00047218526873220605,
874
- "loss": 0.0077,
875
- "step": 198835
876
- },
877
- {
878
- "epoch": 17.4,
879
- "learning_rate": 0.00046662939184195475,
880
- "loss": 0.0063,
881
- "step": 200564
882
  },
883
  {
884
- "epoch": 17.55,
885
- "learning_rate": 0.0004610735149517034,
886
- "loss": 0.0058,
887
- "step": 202293
888
  },
889
  {
890
- "epoch": 17.7,
891
- "learning_rate": 0.0004555176380614521,
892
- "loss": 0.0057,
893
- "step": 204022
894
  },
895
  {
896
- "epoch": 17.85,
897
- "learning_rate": 0.0004499617611712007,
898
- "loss": 0.0059,
899
- "step": 205751
900
  },
901
  {
902
  "epoch": 18.0,
903
- "eval_loss": 0.04149915650486946,
904
- "eval_max_distance": 32,
905
  "eval_mean_distance": 0,
906
- "eval_runtime": 22.9895,
907
- "eval_samples_per_second": 43.498,
908
- "eval_steps_per_second": 2.175,
909
- "step": 207468
910
- },
911
- {
912
- "epoch": 18.0,
913
- "learning_rate": 0.00044440588428094934,
914
- "loss": 0.0082,
915
- "step": 207480
916
- },
917
- {
918
- "epoch": 18.15,
919
- "learning_rate": 0.00043885000739069804,
920
- "loss": 0.0133,
921
- "step": 209209
922
  },
923
  {
924
- "epoch": 18.3,
925
- "learning_rate": 0.00043329413050044663,
926
- "loss": 0.0063,
927
- "step": 210938
928
  },
929
  {
930
- "epoch": 18.45,
931
- "learning_rate": 0.0004277382536101953,
932
- "loss": 0.0051,
933
- "step": 212667
934
- },
935
- {
936
- "epoch": 18.6,
937
- "learning_rate": 0.000422182376719944,
938
- "loss": 0.0051,
939
- "step": 214396
940
- },
941
- {
942
- "epoch": 18.75,
943
- "learning_rate": 0.00041662649982969263,
944
- "loss": 0.0051,
945
- "step": 216125
946
  },
947
  {
948
- "epoch": 18.9,
949
- "learning_rate": 0.0004110706229394413,
950
- "loss": 0.0054,
951
- "step": 217854
952
  },
953
  {
954
  "epoch": 19.0,
955
- "eval_loss": 0.0405677855014801,
956
- "eval_max_distance": 28,
957
  "eval_mean_distance": 0,
958
- "eval_runtime": 21.777,
959
- "eval_samples_per_second": 45.92,
960
- "eval_steps_per_second": 2.296,
961
- "step": 218994
962
- },
963
- {
964
- "epoch": 19.05,
965
- "learning_rate": 0.0004055147460491899,
966
- "loss": 0.0117,
967
- "step": 219583
968
  },
969
  {
970
- "epoch": 19.2,
971
- "learning_rate": 0.00039995886915893857,
972
  "loss": 0.0075,
973
- "step": 221312
974
- },
975
- {
976
- "epoch": 19.35,
977
- "learning_rate": 0.00039440299226868727,
978
- "loss": 0.0051,
979
- "step": 223041
980
  },
981
  {
982
- "epoch": 19.5,
983
- "learning_rate": 0.00038884711537843586,
984
- "loss": 0.0046,
985
- "step": 224770
986
- },
987
- {
988
- "epoch": 19.65,
989
- "learning_rate": 0.0003832912384881845,
990
- "loss": 0.0043,
991
- "step": 226499
992
- },
993
- {
994
- "epoch": 19.8,
995
- "learning_rate": 0.0003777353615979332,
996
- "loss": 0.0044,
997
- "step": 228228
998
  },
999
  {
1000
- "epoch": 19.95,
1001
- "learning_rate": 0.0003721794847076818,
1002
- "loss": 0.0046,
1003
- "step": 229957
1004
  },
1005
  {
1006
  "epoch": 20.0,
1007
- "eval_loss": 0.03926468640565872,
1008
- "eval_max_distance": 24,
1009
  "eval_mean_distance": 0,
1010
- "eval_runtime": 23.1294,
1011
- "eval_samples_per_second": 43.235,
1012
- "eval_steps_per_second": 2.162,
1013
- "step": 230520
1014
- },
1015
- {
1016
- "epoch": 20.1,
1017
- "learning_rate": 0.0003666236078174305,
1018
- "loss": 0.0118,
1019
- "step": 231686
1020
  },
1021
  {
1022
- "epoch": 20.25,
1023
- "learning_rate": 0.00036106773092717915,
1024
- "loss": 0.0056,
1025
- "step": 233415
1026
  },
1027
  {
1028
- "epoch": 20.4,
1029
- "learning_rate": 0.0003555118540369278,
1030
- "loss": 0.0043,
1031
- "step": 235144
1032
- },
1033
- {
1034
- "epoch": 20.55,
1035
- "learning_rate": 0.00034995597714667645,
1036
- "loss": 0.0039,
1037
- "step": 236873
1038
- },
1039
- {
1040
- "epoch": 20.7,
1041
- "learning_rate": 0.0003444001002564251,
1042
- "loss": 0.0037,
1043
- "step": 238602
1044
  },
1045
  {
1046
- "epoch": 20.85,
1047
- "learning_rate": 0.00033884422336617374,
1048
- "loss": 0.004,
1049
- "step": 240331
1050
  },
1051
  {
1052
  "epoch": 21.0,
1053
- "eval_loss": 0.04168349876999855,
1054
- "eval_max_distance": 24,
1055
  "eval_mean_distance": 0,
1056
- "eval_runtime": 22.786,
1057
- "eval_samples_per_second": 43.887,
1058
- "eval_steps_per_second": 2.194,
1059
- "step": 242046
1060
- },
1061
- {
1062
- "epoch": 21.0,
1063
- "learning_rate": 0.0003332883464759224,
1064
- "loss": 0.0053,
1065
- "step": 242060
1066
- },
1067
- {
1068
- "epoch": 21.15,
1069
- "learning_rate": 0.00032773246958567103,
1070
- "loss": 0.0097,
1071
- "step": 243789
1072
  },
1073
  {
1074
- "epoch": 21.3,
1075
- "learning_rate": 0.00032217659269541973,
1076
- "loss": 0.0044,
1077
- "step": 245518
1078
- },
1079
- {
1080
- "epoch": 21.45,
1081
- "learning_rate": 0.0003166207158051684,
1082
- "loss": 0.0036,
1083
- "step": 247247
1084
  },
1085
  {
1086
- "epoch": 21.6,
1087
- "learning_rate": 0.000311064838914917,
1088
- "loss": 0.0031,
1089
- "step": 248976
1090
- },
1091
- {
1092
- "epoch": 21.75,
1093
- "learning_rate": 0.0003055089620246657,
1094
- "loss": 0.0031,
1095
- "step": 250705
1096
  },
1097
  {
1098
- "epoch": 21.9,
1099
- "learning_rate": 0.0002999530851344143,
1100
- "loss": 0.0034,
1101
- "step": 252434
1102
  },
1103
  {
1104
  "epoch": 22.0,
1105
- "eval_loss": 0.040287140756845474,
1106
- "eval_max_distance": 18,
1107
  "eval_mean_distance": 0,
1108
- "eval_runtime": 21.6664,
1109
- "eval_samples_per_second": 46.155,
1110
- "eval_steps_per_second": 2.308,
1111
- "step": 253572
1112
  },
1113
  {
1114
- "epoch": 22.05,
1115
- "learning_rate": 0.0002943972082441629,
1116
- "loss": 0.0082,
1117
- "step": 254163
1118
- },
1119
- {
1120
- "epoch": 22.2,
1121
- "learning_rate": 0.0002888413313539116,
1122
- "loss": 0.0053,
1123
- "step": 255892
1124
- },
1125
- {
1126
- "epoch": 22.35,
1127
- "learning_rate": 0.00028328545446366026,
1128
- "loss": 0.0034,
1129
- "step": 257621
1130
- },
1131
- {
1132
- "epoch": 22.5,
1133
- "learning_rate": 0.00027772957757340896,
1134
- "loss": 0.0031,
1135
- "step": 259350
1136
- },
1137
- {
1138
- "epoch": 22.65,
1139
- "learning_rate": 0.00027217370068315756,
1140
- "loss": 0.0029,
1141
- "step": 261079
1142
  },
1143
  {
1144
- "epoch": 22.8,
1145
- "learning_rate": 0.0002666178237929062,
1146
- "loss": 0.0027,
1147
- "step": 262808
1148
  },
1149
  {
1150
- "epoch": 22.95,
1151
- "learning_rate": 0.0002610619469026549,
1152
- "loss": 0.0029,
1153
- "step": 264537
1154
  },
1155
  {
1156
  "epoch": 23.0,
1157
- "eval_loss": 0.04222797229886055,
1158
- "eval_max_distance": 21,
1159
  "eval_mean_distance": 0,
1160
- "eval_runtime": 22.874,
1161
- "eval_samples_per_second": 43.718,
1162
- "eval_steps_per_second": 2.186,
1163
- "step": 265098
1164
  },
1165
  {
1166
- "epoch": 23.1,
1167
- "learning_rate": 0.0002555060700124035,
1168
- "loss": 0.0081,
1169
- "step": 266266
1170
  },
1171
  {
1172
- "epoch": 23.25,
1173
- "learning_rate": 0.00024995019312215214,
1174
- "loss": 0.0038,
1175
- "step": 267995
1176
- },
1177
- {
1178
- "epoch": 23.4,
1179
- "learning_rate": 0.00024439431623190085,
1180
- "loss": 0.0028,
1181
- "step": 269724
1182
- },
1183
- {
1184
- "epoch": 23.55,
1185
- "learning_rate": 0.00023883843934164947,
1186
- "loss": 0.0025,
1187
- "step": 271453
1188
- },
1189
- {
1190
- "epoch": 23.7,
1191
- "learning_rate": 0.00023328256245139814,
1192
- "loss": 0.0026,
1193
- "step": 273182
1194
  },
1195
  {
1196
  "epoch": 23.85,
1197
- "learning_rate": 0.0002277266855611468,
1198
- "loss": 0.0024,
1199
- "step": 274911
1200
  },
1201
  {
1202
  "epoch": 24.0,
1203
- "eval_loss": 0.04101773351430893,
1204
- "eval_max_distance": 21,
1205
  "eval_mean_distance": 0,
1206
- "eval_runtime": 22.8008,
1207
- "eval_samples_per_second": 43.858,
1208
- "eval_steps_per_second": 2.193,
1209
- "step": 276624
1210
- },
1211
- {
1212
- "epoch": 24.0,
1213
- "learning_rate": 0.00022217080867089543,
1214
- "loss": 0.0033,
1215
- "step": 276640
1216
- },
1217
- {
1218
- "epoch": 24.15,
1219
- "learning_rate": 0.00021661493178064408,
1220
- "loss": 0.0069,
1221
- "step": 278369
1222
- },
1223
- {
1224
- "epoch": 24.3,
1225
- "learning_rate": 0.00021105905489039273,
1226
- "loss": 0.0029,
1227
- "step": 280098
1228
- },
1229
- {
1230
- "epoch": 24.45,
1231
- "learning_rate": 0.0002055031780001414,
1232
- "loss": 0.0024,
1233
- "step": 281827
1234
  },
1235
  {
1236
- "epoch": 24.6,
1237
- "learning_rate": 0.00019994730110989005,
1238
- "loss": 0.0021,
1239
- "step": 283556
1240
  },
1241
  {
1242
- "epoch": 24.75,
1243
- "learning_rate": 0.0001943914242196387,
1244
- "loss": 0.002,
1245
- "step": 285285
1246
  },
1247
  {
1248
- "epoch": 24.9,
1249
- "learning_rate": 0.00018883554732938734,
1250
- "loss": 0.002,
1251
- "step": 287014
1252
  },
1253
  {
1254
  "epoch": 25.0,
1255
- "eval_loss": 0.043479613959789276,
1256
- "eval_max_distance": 15,
1257
  "eval_mean_distance": 0,
1258
- "eval_runtime": 21.481,
1259
- "eval_samples_per_second": 46.553,
1260
- "eval_steps_per_second": 2.328,
1261
- "step": 288150
1262
- },
1263
- {
1264
- "epoch": 25.05,
1265
- "learning_rate": 0.000183279670439136,
1266
- "loss": 0.0055,
1267
- "step": 288743
1268
- },
1269
- {
1270
- "epoch": 25.2,
1271
- "learning_rate": 0.00017772379354888466,
1272
- "loss": 0.0038,
1273
- "step": 290472
1274
  },
1275
  {
1276
- "epoch": 25.35,
1277
- "learning_rate": 0.00017216791665863328,
1278
- "loss": 0.0023,
1279
- "step": 292201
1280
  },
1281
  {
1282
  "epoch": 25.5,
1283
- "learning_rate": 0.00016661203976838196,
1284
- "loss": 0.002,
1285
- "step": 293930
1286
- },
1287
- {
1288
- "epoch": 25.65,
1289
- "learning_rate": 0.0001610561628781306,
1290
- "loss": 0.0017,
1291
- "step": 295659
1292
- },
1293
- {
1294
- "epoch": 25.8,
1295
- "learning_rate": 0.00015550028598787928,
1296
- "loss": 0.0018,
1297
- "step": 297388
1298
  },
1299
  {
1300
- "epoch": 25.95,
1301
- "learning_rate": 0.0001499444090976279,
1302
- "loss": 0.0016,
1303
- "step": 299117
1304
  },
1305
  {
1306
  "epoch": 26.0,
1307
- "eval_loss": 0.04517431557178497,
1308
- "eval_max_distance": 15,
1309
  "eval_mean_distance": 0,
1310
- "eval_runtime": 22.7192,
1311
- "eval_samples_per_second": 44.016,
1312
- "eval_steps_per_second": 2.201,
1313
- "step": 299676
1314
- },
1315
- {
1316
- "epoch": 26.1,
1317
- "learning_rate": 0.00014438853220737654,
1318
- "loss": 0.0054,
1319
- "step": 300846
1320
- },
1321
- {
1322
- "epoch": 26.25,
1323
- "learning_rate": 0.00013883265531712522,
1324
- "loss": 0.0026,
1325
- "step": 302575
1326
  },
1327
  {
1328
- "epoch": 26.4,
1329
- "learning_rate": 0.00013327677842687387,
1330
- "loss": 0.0018,
1331
- "step": 304304
1332
  },
1333
  {
1334
- "epoch": 26.55,
1335
- "learning_rate": 0.0001277209015366225,
1336
- "loss": 0.0016,
1337
- "step": 306033
1338
  },
1339
  {
1340
- "epoch": 26.7,
1341
- "learning_rate": 0.00012216502464637116,
1342
- "loss": 0.0015,
1343
- "step": 307762
1344
- },
1345
- {
1346
- "epoch": 26.85,
1347
- "learning_rate": 0.00011660914775611982,
1348
- "loss": 0.0013,
1349
- "step": 309491
1350
  },
1351
  {
1352
  "epoch": 27.0,
1353
- "eval_loss": 0.04144546017050743,
1354
- "eval_max_distance": 14,
1355
  "eval_mean_distance": 0,
1356
- "eval_runtime": 22.686,
1357
- "eval_samples_per_second": 44.08,
1358
- "eval_steps_per_second": 2.204,
1359
- "step": 311202
1360
  },
1361
  {
1362
- "epoch": 27.0,
1363
- "learning_rate": 0.00011105327086586847,
1364
- "loss": 0.0021,
1365
- "step": 311220
1366
  },
1367
  {
1368
- "epoch": 27.15,
1369
- "learning_rate": 0.00010549739397561713,
1370
- "loss": 0.0046,
1371
- "step": 312949
1372
  },
1373
  {
1374
- "epoch": 27.3,
1375
- "learning_rate": 9.994151708536577e-05,
1376
- "loss": 0.002,
1377
- "step": 314678
1378
  },
1379
  {
1380
- "epoch": 27.45,
1381
- "learning_rate": 9.438564019511443e-05,
1382
- "loss": 0.0016,
1383
- "step": 316407
 
 
 
 
1384
  },
1385
  {
1386
- "epoch": 27.6,
1387
- "learning_rate": 8.882976330486308e-05,
1388
- "loss": 0.0013,
1389
- "step": 318136
1390
  },
1391
  {
1392
- "epoch": 27.75,
1393
- "learning_rate": 8.327388641461173e-05,
1394
- "loss": 0.0013,
1395
- "step": 319865
1396
  },
1397
  {
1398
- "epoch": 27.9,
1399
- "learning_rate": 7.771800952436038e-05,
1400
- "loss": 0.0012,
1401
- "step": 321594
1402
  },
1403
  {
1404
- "epoch": 28.0,
1405
- "eval_loss": 0.04387975484132767,
1406
- "eval_max_distance": 14,
1407
  "eval_mean_distance": 0,
1408
- "eval_runtime": 21.586,
1409
- "eval_samples_per_second": 46.326,
1410
- "eval_steps_per_second": 2.316,
1411
- "step": 322728
1412
  },
1413
  {
1414
- "epoch": 28.05,
1415
- "learning_rate": 7.216213263410904e-05,
1416
- "loss": 0.0037,
1417
- "step": 323323
1418
  },
1419
  {
1420
- "epoch": 28.2,
1421
- "learning_rate": 6.660625574385768e-05,
1422
- "loss": 0.0027,
1423
- "step": 325052
1424
  },
1425
  {
1426
- "epoch": 28.35,
1427
- "learning_rate": 6.105037885360634e-05,
1428
- "loss": 0.0016,
1429
- "step": 326781
1430
  },
1431
  {
1432
- "epoch": 28.5,
1433
- "learning_rate": 5.5494501963355e-05,
1434
- "loss": 0.0013,
1435
- "step": 328510
 
 
 
 
1436
  },
1437
  {
1438
- "epoch": 28.65,
1439
- "learning_rate": 4.9938625073103644e-05,
1440
- "loss": 0.0011,
1441
- "step": 330239
1442
  },
1443
  {
1444
- "epoch": 28.8,
1445
- "learning_rate": 4.43827481828523e-05,
1446
- "loss": 0.001,
1447
- "step": 331968
1448
  },
1449
  {
1450
- "epoch": 28.95,
1451
- "learning_rate": 3.882687129260095e-05,
1452
- "loss": 0.001,
1453
- "step": 333697
1454
  },
1455
  {
1456
- "epoch": 29.0,
1457
- "eval_loss": 0.044395141303539276,
1458
  "eval_max_distance": 15,
1459
  "eval_mean_distance": 0,
1460
- "eval_runtime": 22.6289,
1461
- "eval_samples_per_second": 44.191,
1462
- "eval_steps_per_second": 2.21,
1463
- "step": 334254
1464
- },
1465
- {
1466
- "epoch": 29.1,
1467
- "learning_rate": 3.32709944023496e-05,
1468
- "loss": 0.0039,
1469
- "step": 335426
1470
  },
1471
  {
1472
- "epoch": 29.25,
1473
- "learning_rate": 2.771511751209825e-05,
1474
- "loss": 0.0017,
1475
- "step": 337155
1476
- },
1477
- {
1478
- "epoch": 29.4,
1479
- "learning_rate": 2.2159240621846906e-05,
1480
- "loss": 0.0012,
1481
- "step": 338884
1482
- },
1483
- {
1484
- "epoch": 29.55,
1485
- "learning_rate": 1.6603363731595556e-05,
1486
- "loss": 0.0011,
1487
- "step": 340613
1488
  },
1489
  {
1490
- "epoch": 29.7,
1491
- "learning_rate": 1.1047486841344207e-05,
1492
- "loss": 0.001,
1493
- "step": 342342
1494
  },
1495
  {
1496
- "epoch": 29.85,
1497
- "learning_rate": 0.0004477394104151002,
1498
  "loss": 0.0026,
1499
- "step": 344071
1500
  },
1501
  {
1502
- "epoch": 30.0,
1503
- "eval_loss": 0.042660146951675415,
1504
- "eval_max_distance": 19,
1505
  "eval_mean_distance": 0,
1506
- "eval_runtime": 48.9711,
1507
- "eval_samples_per_second": 20.42,
1508
- "eval_steps_per_second": 1.021,
1509
- "step": 345780
1510
- },
1511
- {
1512
- "epoch": 30.0,
1513
- "learning_rate": 0.00044440588428094934,
1514
- "loss": 0.0045,
1515
- "step": 345800
1516
  },
1517
  {
1518
- "epoch": 30.15,
1519
- "learning_rate": 0.00044107235814679855,
1520
- "loss": 0.0068,
1521
- "step": 347529
1522
  },
1523
  {
1524
- "epoch": 30.3,
1525
- "learning_rate": 0.00043773883201264776,
1526
- "loss": 0.0036,
1527
- "step": 349258
1528
  },
1529
  {
1530
- "epoch": 30.45,
1531
- "learning_rate": 0.0004344053058784969,
1532
- "loss": 0.0032,
1533
- "step": 350987
1534
  },
1535
  {
1536
- "epoch": 30.6,
1537
- "learning_rate": 0.0004310717797443461,
1538
- "loss": 0.0033,
1539
- "step": 352716
 
 
 
 
1540
  },
1541
  {
1542
- "epoch": 30.75,
1543
- "learning_rate": 0.0004277382536101953,
1544
- "loss": 0.0034,
1545
- "step": 354445
1546
  },
1547
  {
1548
- "epoch": 30.9,
1549
- "learning_rate": 0.00042440472747604454,
1550
- "loss": 0.004,
1551
- "step": 356174
1552
  },
1553
  {
1554
- "epoch": 23.29,
1555
- "learning_rate": 0.000593547500777279,
1556
- "loss": 0.0057,
1557
- "step": 357903
1558
  },
1559
  {
1560
- "epoch": 23.4,
1561
- "learning_rate": 0.0005910471934809801,
1562
- "loss": 0.0053,
1563
- "step": 359632
 
 
 
 
1564
  },
1565
  {
1566
- "epoch": 23.52,
1567
- "learning_rate": 0.0005885468861846814,
1568
- "loss": 0.0055,
1569
- "step": 361361
1570
  },
1571
  {
1572
- "epoch": 23.63,
1573
- "learning_rate": 0.0005860465788883828,
1574
- "loss": 0.0056,
1575
- "step": 363090
1576
  },
1577
  {
1578
- "epoch": 23.74,
1579
- "learning_rate": 0.0005835462715920841,
1580
- "loss": 0.006,
1581
- "step": 364819
1582
  },
1583
  {
1584
- "epoch": 23.85,
1585
- "learning_rate": 0.0005810459642957853,
1586
- "loss": 0.0064,
1587
- "step": 366548
 
 
 
 
1588
  },
1589
  {
1590
- "epoch": 23.97,
1591
- "learning_rate": 0.0005785456569994867,
1592
- "loss": 0.0077,
1593
- "step": 368277
1594
  },
1595
  {
1596
- "epoch": 24.0,
1597
- "eval_loss": 0.049542564898729324,
1598
- "eval_max_distance": 27,
1599
- "eval_mean_distance": 0,
1600
- "eval_runtime": 72.4399,
1601
- "eval_samples_per_second": 13.805,
1602
- "eval_steps_per_second": 0.925,
1603
- "step": 368808
1604
  },
1605
  {
1606
- "epoch": 24.08,
1607
- "learning_rate": 0.0005760453497031879,
1608
- "loss": 0.0158,
1609
- "step": 370006
1610
  },
1611
  {
1612
- "epoch": 24.19,
1613
- "learning_rate": 0.0005735450424068892,
1614
- "loss": 0.0082,
1615
- "step": 371735
 
 
 
 
1616
  },
1617
  {
1618
- "epoch": 24.3,
1619
- "learning_rate": 0.0005710447351105905,
1620
- "loss": 0.0057,
1621
- "step": 373464
1622
  },
1623
  {
1624
- "epoch": 24.42,
1625
- "learning_rate": 0.0005685444278142918,
1626
- "loss": 0.0059,
1627
- "step": 375193
1628
  },
1629
  {
1630
- "epoch": 24.53,
1631
- "learning_rate": 0.0005660441205179932,
1632
- "loss": 0.0055,
1633
- "step": 376922
1634
  },
1635
  {
1636
- "epoch": 24.64,
1637
- "learning_rate": 0.0005635438132216944,
1638
- "loss": 0.0062,
1639
- "step": 378651
 
 
 
 
1640
  },
1641
  {
1642
- "epoch": 24.75,
1643
- "learning_rate": 0.0005610435059253957,
1644
- "loss": 0.0059,
1645
- "step": 380380
1646
  },
1647
  {
1648
- "epoch": 24.87,
1649
- "learning_rate": 0.000558543198629097,
1650
- "loss": 0.0062,
1651
- "step": 382109
1652
  },
1653
  {
1654
- "epoch": 24.98,
1655
- "learning_rate": 0.0005560428913327982,
1656
- "loss": 0.0083,
1657
- "step": 383838
1658
  },
1659
  {
1660
- "epoch": 25.0,
1661
- "eval_loss": 0.044624801725149155,
1662
- "eval_max_distance": 37,
1663
  "eval_mean_distance": 0,
1664
- "eval_runtime": 35.859,
1665
- "eval_samples_per_second": 27.887,
1666
- "eval_steps_per_second": 1.868,
1667
- "step": 384175
1668
  },
1669
  {
1670
- "epoch": 25.09,
1671
- "learning_rate": 0.0005535425840364996,
1672
- "loss": 0.0154,
1673
- "step": 385567
1674
- },
1675
- {
1676
- "epoch": 25.2,
1677
- "learning_rate": 0.0005510422767402009,
1678
- "loss": 0.0075,
1679
- "step": 387296
1680
  },
1681
  {
1682
- "epoch": 25.32,
1683
- "learning_rate": 0.0005485419694439023,
1684
- "loss": 0.006,
1685
- "step": 389025
1686
  },
1687
  {
1688
- "epoch": 25.43,
1689
- "learning_rate": 0.0005460416621476034,
1690
- "loss": 0.0057,
1691
- "step": 390754
1692
  },
1693
  {
1694
- "epoch": 25.54,
1695
- "learning_rate": 0.0005435413548513047,
1696
- "loss": 0.0057,
1697
- "step": 392483
 
 
 
 
1698
  },
1699
  {
1700
- "epoch": 25.65,
1701
- "learning_rate": 0.0005410410475550061,
1702
- "loss": 0.0056,
1703
- "step": 394212
1704
  },
1705
  {
1706
- "epoch": 25.77,
1707
- "learning_rate": 0.0005385407402587073,
1708
- "loss": 0.0059,
1709
- "step": 395941
1710
  },
1711
  {
1712
- "epoch": 25.88,
1713
- "learning_rate": 0.0005360404329624087,
1714
- "loss": 0.0067,
1715
- "step": 397670
1716
  },
1717
  {
1718
- "epoch": 25.99,
1719
- "learning_rate": 0.0005335401256661099,
1720
- "loss": 0.0078,
1721
- "step": 399399
 
 
 
 
1722
  },
1723
  {
1724
- "epoch": 26.0,
1725
- "eval_loss": 0.048115409910678864,
1726
- "eval_max_distance": 47,
1727
- "eval_mean_distance": 0,
1728
- "eval_runtime": 35.8872,
1729
- "eval_samples_per_second": 27.865,
1730
- "eval_steps_per_second": 1.867,
1731
- "step": 399542
1732
  },
1733
  {
1734
- "epoch": 26.1,
1735
- "learning_rate": 0.0005310398183698112,
1736
- "loss": 0.0151,
1737
- "step": 401128
1738
  },
1739
  {
1740
- "epoch": 26.22,
1741
- "learning_rate": 0.0005285395110735125,
1742
- "loss": 0.0068,
1743
- "step": 402857
1744
  },
1745
  {
1746
- "epoch": 26.33,
1747
- "learning_rate": 0.0005260392037772138,
1748
- "loss": 0.0055,
1749
- "step": 404586
 
 
 
 
1750
  },
1751
  {
1752
- "epoch": 26.44,
1753
- "learning_rate": 0.0005235388964809152,
1754
- "loss": 0.0052,
1755
- "step": 406315
1756
  },
1757
  {
1758
- "epoch": 26.55,
1759
- "learning_rate": 0.0005210385891846165,
1760
- "loss": 0.0054,
1761
- "step": 408044
1762
  },
1763
  {
1764
- "epoch": 26.67,
1765
- "learning_rate": 0.0005185382818883176,
1766
- "loss": 0.0053,
1767
- "step": 409773
1768
  },
1769
  {
1770
- "epoch": 26.78,
1771
- "learning_rate": 0.000516037974592019,
1772
- "loss": 0.0052,
1773
- "step": 411502
 
 
 
 
1774
  },
1775
  {
1776
- "epoch": 26.89,
1777
- "learning_rate": 0.0005135376672957203,
1778
- "loss": 0.006,
1779
- "step": 413231
1780
  },
1781
  {
1782
- "epoch": 27.0,
1783
- "eval_loss": 0.042408570647239685,
1784
- "eval_max_distance": 37,
1785
- "eval_mean_distance": 0,
1786
- "eval_runtime": 14.3824,
1787
- "eval_samples_per_second": 69.529,
1788
- "eval_steps_per_second": 4.658,
1789
- "step": 414909
1790
  },
1791
  {
1792
- "epoch": 27.0,
1793
- "learning_rate": 0.0005110373599994215,
1794
- "loss": 0.0095,
1795
- "step": 414960
1796
  },
1797
  {
1798
- "epoch": 27.12,
1799
- "learning_rate": 0.0005085370527031229,
1800
- "loss": 0.0115,
1801
- "step": 416689
 
 
 
 
1802
  },
1803
  {
1804
- "epoch": 27.23,
1805
- "learning_rate": 0.0005060367454068242,
1806
- "loss": 0.006,
1807
- "step": 418418
1808
  },
1809
  {
1810
- "epoch": 27.34,
1811
- "learning_rate": 0.0005035364381105254,
1812
- "loss": 0.0051,
1813
- "step": 420147
1814
  },
1815
  {
1816
- "epoch": 27.45,
1817
- "learning_rate": 0.0005010361308142267,
1818
- "loss": 0.0047,
1819
- "step": 421876
1820
  },
1821
  {
1822
- "epoch": 27.57,
1823
- "learning_rate": 0.000498535823517928,
1824
- "loss": 0.0047,
1825
- "step": 423605
 
 
 
 
1826
  },
1827
  {
1828
- "epoch": 27.68,
1829
- "learning_rate": 0.0004960355162216294,
1830
- "loss": 0.0054,
1831
- "step": 425334
1832
  },
1833
  {
1834
- "epoch": 27.79,
1835
- "learning_rate": 0.0004935352089253306,
1836
- "loss": 0.0052,
1837
- "step": 427063
1838
  },
1839
  {
1840
- "epoch": 27.9,
1841
- "learning_rate": 0.000491034901629032,
1842
- "loss": 0.0056,
1843
- "step": 428792
1844
  },
1845
  {
1846
- "epoch": 28.0,
1847
- "eval_loss": 0.04393070191144943,
1848
- "eval_max_distance": 22,
1849
  "eval_mean_distance": 0,
1850
- "eval_runtime": 26.5483,
1851
- "eval_samples_per_second": 37.667,
1852
- "eval_steps_per_second": 2.524,
1853
- "step": 430276
1854
  },
1855
  {
1856
- "epoch": 28.02,
1857
- "learning_rate": 0.0004885345943327332,
1858
- "loss": 0.0105,
1859
- "step": 430521
1860
  },
1861
  {
1862
- "epoch": 28.13,
1863
- "learning_rate": 0.0004860342870364345,
1864
- "loss": 0.0093,
1865
- "step": 432250
1866
  },
1867
  {
1868
- "epoch": 28.24,
1869
- "learning_rate": 0.0004835339797401358,
1870
- "loss": 0.0055,
1871
- "step": 433979
1872
  },
1873
  {
1874
- "epoch": 28.35,
1875
- "learning_rate": 0.0004810336724438371,
1876
- "loss": 0.0049,
1877
- "step": 435708
 
 
 
 
1878
  },
1879
  {
1880
- "epoch": 28.47,
1881
- "learning_rate": 0.00047853336514753835,
1882
- "loss": 0.0044,
1883
- "step": 437437
1884
  },
1885
  {
1886
- "epoch": 28.58,
1887
- "learning_rate": 0.0004760330578512397,
1888
- "loss": 0.0044,
1889
- "step": 439166
1890
  },
1891
  {
1892
- "epoch": 28.69,
1893
- "learning_rate": 0.000473532750554941,
1894
- "loss": 0.0047,
1895
- "step": 440895
1896
  },
1897
  {
1898
- "epoch": 28.8,
1899
- "learning_rate": 0.0004710324432586423,
1900
- "loss": 0.0047,
1901
- "step": 442624
 
 
 
 
1902
  },
1903
  {
1904
- "epoch": 28.92,
1905
- "learning_rate": 0.00046853213596234356,
1906
- "loss": 0.0054,
1907
- "step": 444353
1908
  },
1909
  {
1910
- "epoch": 29.0,
1911
- "eval_loss": 0.04814203828573227,
1912
- "eval_max_distance": 23,
1913
- "eval_mean_distance": 0,
1914
- "eval_runtime": 35.122,
1915
- "eval_samples_per_second": 28.472,
1916
- "eval_steps_per_second": 1.908,
1917
- "step": 445643
1918
  },
1919
  {
1920
- "epoch": 29.03,
1921
- "learning_rate": 0.00046603182866604485,
1922
- "loss": 0.0103,
1923
- "step": 446082
1924
  },
1925
  {
1926
- "epoch": 29.14,
1927
- "learning_rate": 0.00046353152136974614,
1928
- "loss": 0.0076,
1929
- "step": 447811
 
 
 
 
1930
  },
1931
  {
1932
- "epoch": 29.25,
1933
- "learning_rate": 0.00046103121407344743,
1934
- "loss": 0.005,
1935
- "step": 449540
1936
  },
1937
  {
1938
- "epoch": 29.37,
1939
- "learning_rate": 0.00045853090677714877,
1940
- "loss": 0.0043,
1941
- "step": 451269
1942
  },
1943
  {
1944
- "epoch": 29.48,
1945
- "learning_rate": 0.00045603059948085,
1946
- "loss": 0.0043,
1947
- "step": 452998
1948
  },
1949
  {
1950
- "epoch": 29.59,
1951
- "learning_rate": 0.00045353029218455135,
1952
- "loss": 0.004,
1953
- "step": 454727
 
 
 
 
1954
  },
1955
  {
1956
- "epoch": 29.7,
1957
- "learning_rate": 0.0004510299848882526,
1958
- "loss": 0.0042,
1959
- "step": 456456
1960
  },
1961
  {
1962
- "epoch": 29.82,
1963
- "learning_rate": 6.808721912515757e-06,
1964
- "loss": 0.0039,
1965
- "step": 458185
1966
  },
1967
  {
1968
- "epoch": 29.93,
1969
- "learning_rate": 2.641543085351246e-06,
1970
- "loss": 0.004,
1971
- "step": 459914
1972
  },
1973
  {
1974
- "epoch": 30.0,
1975
- "eval_loss": 0.04417673125863075,
1976
- "eval_max_distance": 25,
1977
  "eval_mean_distance": 0,
1978
- "eval_runtime": 68.6071,
1979
- "eval_samples_per_second": 14.576,
1980
- "eval_steps_per_second": 0.977,
1981
- "step": 461010
1982
- },
1983
- {
1984
- "epoch": 30.0,
1985
- "step": 461010,
1986
- "total_flos": 1.1619438964958822e+17,
1987
- "train_loss": 3.931375028864634e-05,
1988
- "train_runtime": 411.3459,
1989
- "train_samples_per_second": 16810.814,
1990
- "train_steps_per_second": 1120.736
1991
  }
1992
  ],
1993
- "logging_steps": 1729,
1994
- "max_steps": 461010,
1995
- "num_train_epochs": 30,
1996
- "save_steps": 3458,
1997
- "total_flos": 1.1619438964958822e+17,
1998
  "trial_name": null,
1999
  "trial_params": null
2000
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 50.0,
5
  "eval_steps": 500,
6
+ "global_step": 577550,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 0.0,
13
+ "learning_rate": 1.3071041108424287e-08,
14
+ "loss": 13.0422,
15
  "step": 1
16
  },
17
  {
18
+ "epoch": 0.25,
19
+ "learning_rate": 5.000980328083132e-05,
20
+ "loss": 1.5626,
21
+ "step": 3826
22
  },
23
  {
24
+ "epoch": 0.5,
25
+ "learning_rate": 0.00010001960656166264,
26
+ "loss": 0.2997,
27
+ "step": 7652
 
 
 
 
 
 
 
 
 
 
 
 
28
  },
29
  {
30
  "epoch": 0.75,
31
+ "learning_rate": 0.00015002940984249396,
32
+ "loss": 0.2198,
33
+ "step": 11478
 
 
 
 
 
 
34
  },
35
  {
36
  "epoch": 1.0,
37
+ "eval_loss": 0.7378385066986084,
38
+ "eval_max_distance": 206,
39
+ "eval_mean_distance": 16,
40
+ "eval_runtime": 214.8198,
41
+ "eval_samples_per_second": 9.31,
42
+ "eval_steps_per_second": 0.624,
43
+ "step": 15301
 
 
 
 
 
 
44
  },
45
  {
46
+ "epoch": 1.32,
47
+ "learning_rate": 0.00026498138689290976,
48
+ "loss": 1.5195,
49
+ "step": 15304
50
  },
51
  {
52
+ "epoch": 1.66,
53
+ "learning_rate": 0.00033122673361613713,
54
+ "loss": 0.1366,
55
+ "step": 19130
56
  },
57
  {
58
+ "epoch": 1.99,
59
+ "learning_rate": 0.00039747208033936456,
60
+ "loss": 0.1431,
61
+ "step": 22956
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
62
  },
63
  {
64
  "epoch": 2.0,
65
+ "eval_loss": 0.983799397945404,
66
+ "eval_max_distance": 183,
67
+ "eval_mean_distance": 22,
68
+ "eval_runtime": 10.3439,
69
+ "eval_samples_per_second": 48.338,
70
+ "eval_steps_per_second": 2.417,
71
+ "step": 23102
 
 
 
 
 
 
72
  },
73
  {
74
+ "epoch": 2.32,
75
+ "learning_rate": 0.000463717427062592,
76
+ "loss": 0.1221,
77
+ "step": 26782
78
  },
79
  {
80
+ "epoch": 2.65,
81
+ "learning_rate": 0.0005299627737858195,
82
+ "loss": 0.0829,
83
+ "step": 30608
84
  },
85
  {
86
+ "epoch": 2.98,
87
+ "learning_rate": 0.0005962081205090468,
88
+ "loss": 0.1014,
89
+ "step": 34434
 
 
 
 
 
 
 
 
 
 
 
 
90
  },
91
  {
92
  "epoch": 3.0,
93
+ "eval_loss": 0.5182881951332092,
94
  "eval_max_distance": 122,
95
+ "eval_mean_distance": 13,
96
+ "eval_runtime": 8.5861,
97
+ "eval_samples_per_second": 58.233,
98
+ "eval_steps_per_second": 2.912,
99
+ "step": 34653
100
  },
101
  {
102
+ "epoch": 3.31,
103
+ "learning_rate": 0.0006624534672322743,
104
+ "loss": 0.0865,
105
+ "step": 38260
 
 
 
 
 
 
106
  },
107
  {
108
+ "epoch": 3.64,
109
+ "learning_rate": 0.0007286988139555017,
110
+ "loss": 0.0645,
111
+ "step": 42086
112
  },
113
  {
114
+ "epoch": 3.97,
115
+ "learning_rate": 0.0007949441606787291,
116
+ "loss": 0.0818,
117
+ "step": 45912
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
118
  },
119
  {
120
  "epoch": 4.0,
121
+ "eval_loss": 0.3150796592235565,
122
+ "eval_max_distance": 94,
123
+ "eval_mean_distance": 7,
124
+ "eval_runtime": 10.1617,
125
+ "eval_samples_per_second": 49.204,
126
+ "eval_steps_per_second": 2.46,
127
+ "step": 46204
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
128
  },
129
  {
130
+ "epoch": 4.31,
131
+ "learning_rate": 0.0008611895074019565,
132
+ "loss": 0.0713,
133
+ "step": 49738
134
  },
135
  {
136
+ "epoch": 4.64,
137
+ "learning_rate": 0.000927434854125184,
138
+ "loss": 0.0547,
139
+ "step": 53564
140
  },
141
  {
142
+ "epoch": 4.97,
143
+ "learning_rate": 0.0009936802008484113,
144
+ "loss": 0.07,
145
+ "step": 57390
146
  },
147
  {
148
  "epoch": 5.0,
149
+ "eval_loss": 0.48977312445640564,
150
  "eval_max_distance": 100,
151
+ "eval_mean_distance": 12,
152
+ "eval_runtime": 8.5226,
153
+ "eval_samples_per_second": 58.667,
154
+ "eval_steps_per_second": 2.933,
155
+ "step": 57755
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
156
  },
157
  {
158
+ "epoch": 5.3,
159
+ "learning_rate": 0.0009933416058253734,
160
+ "loss": 0.0632,
161
+ "step": 61216
162
  },
163
  {
164
+ "epoch": 5.63,
165
+ "learning_rate": 0.0009859810117450148,
166
+ "loss": 0.0462,
167
+ "step": 65042
168
  },
169
  {
170
+ "epoch": 5.96,
171
+ "learning_rate": 0.0009786204176646563,
172
+ "loss": 0.0555,
173
+ "step": 68868
 
 
 
 
174
  },
175
  {
176
  "epoch": 6.0,
177
+ "eval_loss": 0.3158600926399231,
178
+ "eval_max_distance": 99,
179
+ "eval_mean_distance": 8,
180
+ "eval_runtime": 8.627,
181
+ "eval_samples_per_second": 57.958,
182
+ "eval_steps_per_second": 2.898,
183
+ "step": 69306
 
 
 
 
 
 
 
 
184
  },
185
  {
186
+ "epoch": 6.29,
187
+ "learning_rate": 0.0009712598235842977,
188
+ "loss": 0.0532,
189
+ "step": 72694
190
  },
191
  {
192
+ "epoch": 6.62,
193
+ "learning_rate": 0.000963899229503939,
194
+ "loss": 0.0367,
195
+ "step": 76520
196
  },
197
  {
198
+ "epoch": 6.96,
199
+ "learning_rate": 0.0009565386354235804,
200
+ "loss": 0.0442,
201
+ "step": 80346
 
 
 
 
 
 
202
  },
203
  {
204
  "epoch": 7.0,
205
+ "eval_loss": 0.2628127336502075,
206
+ "eval_max_distance": 91,
207
+ "eval_mean_distance": 6,
208
+ "eval_runtime": 8.3481,
209
+ "eval_samples_per_second": 59.894,
210
+ "eval_steps_per_second": 2.995,
211
+ "step": 80857
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
212
  },
213
  {
214
+ "epoch": 7.29,
215
+ "learning_rate": 0.0009491780413432219,
216
+ "loss": 0.0455,
217
+ "step": 84172
218
  },
219
  {
220
+ "epoch": 7.62,
221
+ "learning_rate": 0.0009418174472628632,
222
+ "loss": 0.0307,
223
+ "step": 87998
224
  },
225
  {
226
  "epoch": 7.95,
227
+ "learning_rate": 0.0009344568531825046,
228
+ "loss": 0.0358,
229
+ "step": 91824
230
  },
231
  {
232
  "epoch": 8.0,
233
+ "eval_loss": 0.2341708391904831,
234
+ "eval_max_distance": 69,
235
+ "eval_mean_distance": 5,
236
+ "eval_runtime": 8.2389,
237
+ "eval_samples_per_second": 60.688,
238
+ "eval_steps_per_second": 3.034,
239
+ "step": 92408
 
 
 
 
 
 
 
 
 
 
 
 
240
  },
241
  {
242
+ "epoch": 8.28,
243
+ "learning_rate": 0.000927096259102146,
244
+ "loss": 0.0397,
245
+ "step": 95650
246
  },
247
  {
248
+ "epoch": 8.61,
249
+ "learning_rate": 0.0009197356650217874,
250
+ "loss": 0.0261,
251
+ "step": 99476
252
  },
253
  {
254
+ "epoch": 8.94,
255
+ "learning_rate": 0.0009123750709414288,
256
+ "loss": 0.0299,
257
+ "step": 103302
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
258
  },
259
  {
260
  "epoch": 9.0,
261
+ "eval_loss": 0.11001851409673691,
262
+ "eval_max_distance": 63,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
263
  "eval_mean_distance": 2,
264
+ "eval_runtime": 7.3095,
265
+ "eval_samples_per_second": 68.405,
266
+ "eval_steps_per_second": 3.42,
267
+ "step": 103959
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
268
  },
269
  {
270
+ "epoch": 9.27,
271
+ "learning_rate": 0.0009050144768610703,
272
+ "loss": 0.0346,
273
+ "step": 107128
274
  },
275
  {
276
+ "epoch": 9.61,
277
+ "learning_rate": 0.0008976538827807117,
278
+ "loss": 0.0225,
279
+ "step": 110954
280
  },
281
  {
282
+ "epoch": 9.94,
283
+ "learning_rate": 0.0008902932887003531,
284
+ "loss": 0.0252,
285
+ "step": 114780
286
  },
287
  {
288
+ "epoch": 10.0,
289
+ "eval_loss": 0.09344363212585449,
290
+ "eval_max_distance": 51,
291
  "eval_mean_distance": 1,
292
+ "eval_runtime": 7.6397,
293
+ "eval_samples_per_second": 65.447,
294
+ "eval_steps_per_second": 3.272,
295
+ "step": 115510
296
  },
297
  {
298
+ "epoch": 10.27,
299
+ "learning_rate": 0.0008829326946199945,
300
+ "loss": 0.0308,
301
+ "step": 118606
302
  },
303
  {
304
+ "epoch": 10.6,
305
+ "learning_rate": 0.0008755721005396359,
306
+ "loss": 0.0194,
307
+ "step": 122432
308
  },
309
  {
310
+ "epoch": 10.93,
311
+ "learning_rate": 0.0008682115064592773,
312
+ "loss": 0.022,
313
+ "step": 126258
314
  },
315
  {
316
+ "epoch": 11.0,
317
+ "eval_loss": 0.13300806283950806,
318
+ "eval_max_distance": 53,
319
+ "eval_mean_distance": 3,
320
+ "eval_runtime": 7.8584,
321
+ "eval_samples_per_second": 63.626,
322
+ "eval_steps_per_second": 3.181,
323
+ "step": 127061
324
  },
325
  {
326
+ "epoch": 11.26,
327
+ "learning_rate": 0.0008608509123789186,
328
+ "loss": 0.028,
329
+ "step": 130084
330
  },
331
  {
332
+ "epoch": 11.59,
333
+ "learning_rate": 0.0008534903182985601,
334
+ "loss": 0.017,
335
+ "step": 133910
336
  },
337
  {
338
+ "epoch": 11.92,
339
+ "learning_rate": 0.0008461297242182014,
340
+ "loss": 0.019,
341
+ "step": 137736
 
 
 
 
342
  },
343
  {
344
  "epoch": 12.0,
345
+ "eval_loss": 0.12166010588407516,
346
+ "eval_max_distance": 51,
347
+ "eval_mean_distance": 2,
348
+ "eval_runtime": 7.7916,
349
+ "eval_samples_per_second": 64.172,
350
+ "eval_steps_per_second": 3.209,
351
+ "step": 138612
 
 
 
 
 
 
 
 
 
 
 
 
 
 
352
  },
353
  {
354
+ "epoch": 12.26,
355
+ "learning_rate": 0.0008387691301378428,
356
+ "loss": 0.0252,
357
+ "step": 141562
358
  },
359
  {
360
+ "epoch": 12.59,
361
+ "learning_rate": 0.0008314085360574843,
362
+ "loss": 0.0151,
363
+ "step": 145388
364
  },
365
  {
366
+ "epoch": 12.92,
367
+ "learning_rate": 0.0008240479419771256,
368
+ "loss": 0.0163,
369
+ "step": 149214
370
  },
371
  {
372
  "epoch": 13.0,
373
+ "eval_loss": 0.07717631012201309,
374
+ "eval_max_distance": 37,
375
+ "eval_mean_distance": 1,
376
+ "eval_runtime": 7.0966,
377
+ "eval_samples_per_second": 70.457,
378
+ "eval_steps_per_second": 3.523,
379
+ "step": 150163
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
380
  },
381
  {
382
+ "epoch": 13.25,
383
+ "learning_rate": 0.000816687347896767,
384
+ "loss": 0.0228,
385
+ "step": 153040
386
  },
387
  {
388
+ "epoch": 13.58,
389
+ "learning_rate": 0.0008093267538164085,
390
+ "loss": 0.0132,
391
+ "step": 156866
392
  },
393
  {
394
+ "epoch": 13.91,
395
+ "learning_rate": 0.0008019661597360498,
396
+ "loss": 0.0145,
397
+ "step": 160692
398
  },
399
  {
400
  "epoch": 14.0,
401
+ "eval_loss": 0.12677042186260223,
402
+ "eval_max_distance": 65,
403
+ "eval_mean_distance": 2,
404
+ "eval_runtime": 7.3512,
405
+ "eval_samples_per_second": 68.016,
406
+ "eval_steps_per_second": 3.401,
407
+ "step": 161714
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
408
  },
409
  {
410
+ "epoch": 14.24,
411
+ "learning_rate": 0.0007946055656556912,
412
+ "loss": 0.0211,
413
+ "step": 164518
414
  },
415
  {
416
+ "epoch": 14.57,
417
+ "learning_rate": 0.0007872449715753325,
418
+ "loss": 0.0119,
419
+ "step": 168344
420
  },
421
  {
422
+ "epoch": 14.91,
423
+ "learning_rate": 0.000779884377494974,
424
+ "loss": 0.0131,
425
+ "step": 172170
 
 
 
 
 
 
 
 
 
 
426
  },
427
  {
428
  "epoch": 15.0,
429
+ "eval_loss": 0.08834327012300491,
430
+ "eval_max_distance": 34,
431
+ "eval_mean_distance": 1,
432
+ "eval_runtime": 7.4014,
433
+ "eval_samples_per_second": 67.555,
434
+ "eval_steps_per_second": 3.378,
435
+ "step": 173265
 
 
 
 
 
 
 
 
 
 
 
 
 
 
436
  },
437
  {
438
+ "epoch": 15.24,
439
+ "learning_rate": 0.0007725237834146154,
440
+ "loss": 0.019,
441
+ "step": 175996
442
  },
443
  {
444
+ "epoch": 15.57,
445
+ "learning_rate": 0.0007651631893342567,
446
+ "loss": 0.0109,
447
+ "step": 179822
448
  },
449
  {
450
  "epoch": 15.9,
451
+ "learning_rate": 0.0007578025952538982,
452
+ "loss": 0.0115,
453
+ "step": 183648
454
  },
455
  {
456
  "epoch": 16.0,
457
+ "eval_loss": 0.07092861831188202,
458
+ "eval_max_distance": 30,
459
+ "eval_mean_distance": 1,
460
+ "eval_runtime": 7.346,
461
+ "eval_samples_per_second": 68.064,
462
+ "eval_steps_per_second": 3.403,
463
+ "step": 184816
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
464
  },
465
  {
466
+ "epoch": 16.23,
467
+ "learning_rate": 0.0007504420011735396,
468
+ "loss": 0.0175,
469
+ "step": 187474
470
  },
471
  {
472
+ "epoch": 16.56,
473
+ "learning_rate": 0.0007430814070931809,
474
+ "loss": 0.0097,
475
+ "step": 191300
476
  },
477
  {
478
+ "epoch": 16.89,
479
+ "learning_rate": 0.0007357208130128224,
480
+ "loss": 0.0104,
481
+ "step": 195126
482
  },
483
  {
484
  "epoch": 17.0,
485
+ "eval_loss": 0.06687292456626892,
486
+ "eval_max_distance": 38,
487
+ "eval_mean_distance": 1,
488
+ "eval_runtime": 7.1541,
489
+ "eval_samples_per_second": 69.89,
490
+ "eval_steps_per_second": 3.494,
491
+ "step": 196367
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
492
  },
493
  {
494
+ "epoch": 17.22,
495
+ "learning_rate": 0.0007283602189324637,
496
+ "loss": 0.0161,
497
+ "step": 198952
498
  },
499
  {
500
+ "epoch": 17.56,
501
+ "learning_rate": 0.0007209996248521051,
502
+ "loss": 0.0088,
503
+ "step": 202778
504
  },
505
  {
506
+ "epoch": 17.89,
507
+ "learning_rate": 0.0007136390307717465,
508
+ "loss": 0.0094,
509
+ "step": 206604
510
  },
511
  {
512
  "epoch": 18.0,
513
+ "eval_loss": 0.06669158488512039,
514
+ "eval_max_distance": 29,
515
  "eval_mean_distance": 0,
516
+ "eval_runtime": 6.7997,
517
+ "eval_samples_per_second": 73.532,
518
+ "eval_steps_per_second": 3.677,
519
+ "step": 207918
 
 
 
 
 
 
 
 
 
 
 
 
520
  },
521
  {
522
+ "epoch": 18.22,
523
+ "learning_rate": 0.0007062784366913879,
524
+ "loss": 0.0144,
525
+ "step": 210430
526
  },
527
  {
528
+ "epoch": 18.55,
529
+ "learning_rate": 0.0006989178426110293,
530
+ "loss": 0.0081,
531
+ "step": 214256
 
 
 
 
 
 
 
 
 
 
 
 
532
  },
533
  {
534
+ "epoch": 18.88,
535
+ "learning_rate": 0.0006915572485306708,
536
+ "loss": 0.0086,
537
+ "step": 218082
538
  },
539
  {
540
  "epoch": 19.0,
541
+ "eval_loss": 0.06412819027900696,
542
+ "eval_max_distance": 27,
543
  "eval_mean_distance": 0,
544
+ "eval_runtime": 7.0833,
545
+ "eval_samples_per_second": 70.589,
546
+ "eval_steps_per_second": 3.529,
547
+ "step": 219469
 
 
 
 
 
 
548
  },
549
  {
550
+ "epoch": 19.21,
551
+ "learning_rate": 0.0006841966544503122,
552
  "loss": 0.0075,
553
+ "step": 221908
 
 
 
 
 
 
554
  },
555
  {
556
+ "epoch": 19.54,
557
+ "learning_rate": 0.0006768360603699536,
558
+ "loss": 0.0075,
559
+ "step": 225734
 
 
 
 
 
 
 
 
 
 
 
 
560
  },
561
  {
562
+ "epoch": 19.87,
563
+ "learning_rate": 0.000669475466289595,
564
+ "loss": 0.0078,
565
+ "step": 229560
566
  },
567
  {
568
  "epoch": 20.0,
569
+ "eval_loss": 0.06507006287574768,
570
+ "eval_max_distance": 22,
571
  "eval_mean_distance": 0,
572
+ "eval_runtime": 8.9369,
573
+ "eval_samples_per_second": 55.948,
574
+ "eval_steps_per_second": 2.797,
575
+ "step": 231020
 
 
 
 
 
 
576
  },
577
  {
578
+ "epoch": 20.2,
579
+ "learning_rate": 0.0006621148722092364,
580
+ "loss": 0.0127,
581
+ "step": 233386
582
  },
583
  {
584
+ "epoch": 20.54,
585
+ "learning_rate": 0.0006547542781288778,
586
+ "loss": 0.0068,
587
+ "step": 237212
 
 
 
 
 
 
 
 
 
 
 
 
588
  },
589
  {
590
+ "epoch": 20.87,
591
+ "learning_rate": 0.0006473936840485191,
592
+ "loss": 0.007,
593
+ "step": 241038
594
  },
595
  {
596
  "epoch": 21.0,
597
+ "eval_loss": 0.06505928188562393,
598
+ "eval_max_distance": 33,
599
  "eval_mean_distance": 0,
600
+ "eval_runtime": 6.7396,
601
+ "eval_samples_per_second": 74.188,
602
+ "eval_steps_per_second": 3.709,
603
+ "step": 242571
 
 
 
 
 
 
 
 
 
 
 
 
604
  },
605
  {
606
+ "epoch": 21.2,
607
+ "learning_rate": 0.0006400330899681606,
608
+ "loss": 0.0116,
609
+ "step": 244864
 
 
 
 
 
 
610
  },
611
  {
612
+ "epoch": 21.53,
613
+ "learning_rate": 0.000632672495887802,
614
+ "loss": 0.0064,
615
+ "step": 248690
 
 
 
 
 
 
616
  },
617
  {
618
+ "epoch": 21.86,
619
+ "learning_rate": 0.0006253119018074433,
620
+ "loss": 0.0064,
621
+ "step": 252516
622
  },
623
  {
624
  "epoch": 22.0,
625
+ "eval_loss": 0.06048734113574028,
626
+ "eval_max_distance": 31,
627
  "eval_mean_distance": 0,
628
+ "eval_runtime": 6.4223,
629
+ "eval_samples_per_second": 77.853,
630
+ "eval_steps_per_second": 3.893,
631
+ "step": 254122
632
  },
633
  {
634
+ "epoch": 22.19,
635
+ "learning_rate": 0.0006179513077270847,
636
+ "loss": 0.0105,
637
+ "step": 256342
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
638
  },
639
  {
640
+ "epoch": 22.52,
641
+ "learning_rate": 0.0006105907136467262,
642
+ "loss": 0.0059,
643
+ "step": 260168
644
  },
645
  {
646
+ "epoch": 22.85,
647
+ "learning_rate": 0.0006032301195663675,
648
+ "loss": 0.0059,
649
+ "step": 263994
650
  },
651
  {
652
  "epoch": 23.0,
653
+ "eval_loss": 0.0561179518699646,
654
+ "eval_max_distance": 25,
655
  "eval_mean_distance": 0,
656
+ "eval_runtime": 6.4339,
657
+ "eval_samples_per_second": 77.714,
658
+ "eval_steps_per_second": 3.886,
659
+ "step": 265673
660
  },
661
  {
662
+ "epoch": 23.19,
663
+ "learning_rate": 0.0005958695254860089,
664
+ "loss": 0.01,
665
+ "step": 267820
666
  },
667
  {
668
+ "epoch": 23.52,
669
+ "learning_rate": 0.0005885089314056503,
670
+ "loss": 0.0054,
671
+ "step": 271646
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
672
  },
673
  {
674
  "epoch": 23.85,
675
+ "learning_rate": 0.0005811483373252917,
676
+ "loss": 0.0053,
677
+ "step": 275472
678
  },
679
  {
680
  "epoch": 24.0,
681
+ "eval_loss": 0.05673111602663994,
682
+ "eval_max_distance": 36,
683
  "eval_mean_distance": 0,
684
+ "eval_runtime": 6.5025,
685
+ "eval_samples_per_second": 76.893,
686
+ "eval_steps_per_second": 3.845,
687
+ "step": 277224
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
688
  },
689
  {
690
+ "epoch": 24.18,
691
+ "learning_rate": 0.0005737877432449331,
692
+ "loss": 0.0091,
693
+ "step": 279298
694
  },
695
  {
696
+ "epoch": 24.51,
697
+ "learning_rate": 0.0005664271491645745,
698
+ "loss": 0.005,
699
+ "step": 283124
700
  },
701
  {
702
+ "epoch": 24.84,
703
+ "learning_rate": 0.0005590665550842159,
704
+ "loss": 0.0049,
705
+ "step": 286950
706
  },
707
  {
708
  "epoch": 25.0,
709
+ "eval_loss": 0.0556904599070549,
710
+ "eval_max_distance": 25,
711
  "eval_mean_distance": 0,
712
+ "eval_runtime": 6.3704,
713
+ "eval_samples_per_second": 78.489,
714
+ "eval_steps_per_second": 3.924,
715
+ "step": 288775
 
 
 
 
 
 
 
 
 
 
 
 
716
  },
717
  {
718
+ "epoch": 25.17,
719
+ "learning_rate": 0.0005517059610038573,
720
+ "loss": 0.0086,
721
+ "step": 290776
722
  },
723
  {
724
  "epoch": 25.5,
725
+ "learning_rate": 0.0005443453669234986,
726
+ "loss": 0.0047,
727
+ "step": 294602
 
 
 
 
 
 
 
 
 
 
 
 
728
  },
729
  {
730
+ "epoch": 25.84,
731
+ "learning_rate": 0.0005369847728431401,
732
+ "loss": 0.0045,
733
+ "step": 298428
734
  },
735
  {
736
  "epoch": 26.0,
737
+ "eval_loss": 0.052325230091810226,
738
+ "eval_max_distance": 26,
739
  "eval_mean_distance": 0,
740
+ "eval_runtime": 6.5774,
741
+ "eval_samples_per_second": 76.017,
742
+ "eval_steps_per_second": 3.801,
743
+ "step": 300326
 
 
 
 
 
 
 
 
 
 
 
 
744
  },
745
  {
746
+ "epoch": 26.17,
747
+ "learning_rate": 0.0005296241787627815,
748
+ "loss": 0.008,
749
+ "step": 302254
750
  },
751
  {
752
+ "epoch": 26.5,
753
+ "learning_rate": 0.0005222635846824228,
754
+ "loss": 0.0044,
755
+ "step": 306080
756
  },
757
  {
758
+ "epoch": 26.83,
759
+ "learning_rate": 0.0005149029906020643,
760
+ "loss": 0.0041,
761
+ "step": 309906
 
 
 
 
 
 
762
  },
763
  {
764
  "epoch": 27.0,
765
+ "eval_loss": 0.05767456814646721,
766
+ "eval_max_distance": 25,
767
  "eval_mean_distance": 0,
768
+ "eval_runtime": 6.3552,
769
+ "eval_samples_per_second": 78.676,
770
+ "eval_steps_per_second": 3.934,
771
+ "step": 311877
772
  },
773
  {
774
+ "epoch": 27.16,
775
+ "learning_rate": 0.0005075423965217056,
776
+ "loss": 0.0073,
777
+ "step": 313732
778
  },
779
  {
780
+ "epoch": 27.49,
781
+ "learning_rate": 0.000500181802441347,
782
+ "loss": 0.004,
783
+ "step": 317558
784
  },
785
  {
786
+ "epoch": 27.82,
787
+ "learning_rate": 0.0004928212083609885,
788
+ "loss": 0.0038,
789
+ "step": 321384
790
  },
791
  {
792
+ "epoch": 28.0,
793
+ "eval_loss": 0.057724930346012115,
794
+ "eval_max_distance": 32,
795
+ "eval_mean_distance": 0,
796
+ "eval_runtime": 6.472,
797
+ "eval_samples_per_second": 77.256,
798
+ "eval_steps_per_second": 3.863,
799
+ "step": 323428
800
  },
801
  {
802
+ "epoch": 28.15,
803
+ "learning_rate": 0.0004854606142806299,
804
+ "loss": 0.0066,
805
+ "step": 325210
806
  },
807
  {
808
+ "epoch": 28.49,
809
+ "learning_rate": 0.00047810002020027127,
810
+ "loss": 0.0039,
811
+ "step": 329036
812
  },
813
  {
814
+ "epoch": 28.82,
815
+ "learning_rate": 0.00047073942611991265,
816
+ "loss": 0.0033,
817
+ "step": 332862
818
  },
819
  {
820
+ "epoch": 29.0,
821
+ "eval_loss": 0.05752771720290184,
822
+ "eval_max_distance": 20,
823
  "eval_mean_distance": 0,
824
+ "eval_runtime": 6.3104,
825
+ "eval_samples_per_second": 79.234,
826
+ "eval_steps_per_second": 3.962,
827
+ "step": 334979
828
  },
829
  {
830
+ "epoch": 29.15,
831
+ "learning_rate": 0.00046337883203955403,
832
+ "loss": 0.0061,
833
+ "step": 336688
834
  },
835
  {
836
+ "epoch": 29.48,
837
+ "learning_rate": 0.00045601823795919546,
838
+ "loss": 0.0034,
839
+ "step": 340514
840
  },
841
  {
842
+ "epoch": 29.81,
843
+ "learning_rate": 0.00044865764387883684,
844
+ "loss": 0.0031,
845
+ "step": 344340
846
  },
847
  {
848
+ "epoch": 30.0,
849
+ "eval_loss": 0.05623897165060043,
850
+ "eval_max_distance": 16,
851
+ "eval_mean_distance": 0,
852
+ "eval_runtime": 6.2956,
853
+ "eval_samples_per_second": 79.42,
854
+ "eval_steps_per_second": 3.971,
855
+ "step": 346530
856
  },
857
  {
858
+ "epoch": 30.14,
859
+ "learning_rate": 0.0004412970497984782,
860
+ "loss": 0.0058,
861
+ "step": 348166
862
  },
863
  {
864
+ "epoch": 30.47,
865
+ "learning_rate": 0.00043393645571811965,
866
+ "loss": 0.0032,
867
+ "step": 351992
868
  },
869
  {
870
+ "epoch": 30.8,
871
+ "learning_rate": 0.0004265758616377611,
872
+ "loss": 0.0026,
873
+ "step": 355818
874
  },
875
  {
876
+ "epoch": 31.0,
877
+ "eval_loss": 0.05957825854420662,
878
  "eval_max_distance": 15,
879
  "eval_mean_distance": 0,
880
+ "eval_runtime": 6.4083,
881
+ "eval_samples_per_second": 78.023,
882
+ "eval_steps_per_second": 3.901,
883
+ "step": 358081
 
 
 
 
 
 
884
  },
885
  {
886
+ "epoch": 31.14,
887
+ "learning_rate": 0.00041921526755740247,
888
+ "loss": 0.0051,
889
+ "step": 359644
 
 
 
 
 
 
 
 
 
 
 
 
890
  },
891
  {
892
+ "epoch": 31.47,
893
+ "learning_rate": 0.00041185467347704385,
894
+ "loss": 0.0031,
895
+ "step": 363470
896
  },
897
  {
898
+ "epoch": 31.8,
899
+ "learning_rate": 0.0004044940793966853,
900
  "loss": 0.0026,
901
+ "step": 367296
902
  },
903
  {
904
+ "epoch": 32.0,
905
+ "eval_loss": 0.06173785403370857,
906
+ "eval_max_distance": 30,
907
  "eval_mean_distance": 0,
908
+ "eval_runtime": 6.2839,
909
+ "eval_samples_per_second": 79.568,
910
+ "eval_steps_per_second": 3.978,
911
+ "step": 369632
 
 
 
 
 
 
912
  },
913
  {
914
+ "epoch": 32.13,
915
+ "learning_rate": 0.00039713348531632666,
916
+ "loss": 0.0047,
917
+ "step": 371122
918
  },
919
  {
920
+ "epoch": 32.46,
921
+ "learning_rate": 0.00038977289123596804,
922
+ "loss": 0.0028,
923
+ "step": 374948
924
  },
925
  {
926
+ "epoch": 32.79,
927
+ "learning_rate": 0.0003824122971556094,
928
+ "loss": 0.0024,
929
+ "step": 378774
930
  },
931
  {
932
+ "epoch": 33.0,
933
+ "eval_loss": 0.05716051906347275,
934
+ "eval_max_distance": 20,
935
+ "eval_mean_distance": 0,
936
+ "eval_runtime": 6.221,
937
+ "eval_samples_per_second": 80.373,
938
+ "eval_steps_per_second": 4.019,
939
+ "step": 381183
940
  },
941
  {
942
+ "epoch": 33.12,
943
+ "learning_rate": 0.00037505170307525085,
944
+ "loss": 0.0042,
945
+ "step": 382600
946
  },
947
  {
948
+ "epoch": 33.45,
949
+ "learning_rate": 0.00036769110899489223,
950
+ "loss": 0.0026,
951
+ "step": 386426
952
  },
953
  {
954
+ "epoch": 33.79,
955
+ "learning_rate": 0.0003603305149145336,
956
+ "loss": 0.0022,
957
+ "step": 390252
958
  },
959
  {
960
+ "epoch": 34.0,
961
+ "eval_loss": 0.05569161847233772,
962
+ "eval_max_distance": 19,
963
+ "eval_mean_distance": 0,
964
+ "eval_runtime": 6.2592,
965
+ "eval_samples_per_second": 79.883,
966
+ "eval_steps_per_second": 3.994,
967
+ "step": 392734
968
  },
969
  {
970
+ "epoch": 34.12,
971
+ "learning_rate": 0.000352969920834175,
972
+ "loss": 0.0039,
973
+ "step": 394078
974
  },
975
  {
976
+ "epoch": 34.45,
977
+ "learning_rate": 0.0003456093267538164,
978
+ "loss": 0.0024,
979
+ "step": 397904
980
  },
981
  {
982
+ "epoch": 34.78,
983
+ "learning_rate": 0.0003382487326734578,
984
+ "loss": 0.002,
985
+ "step": 401730
986
  },
987
  {
988
+ "epoch": 35.0,
989
+ "eval_loss": 0.06119859591126442,
990
+ "eval_max_distance": 17,
991
+ "eval_mean_distance": 0,
992
+ "eval_runtime": 6.2535,
993
+ "eval_samples_per_second": 79.955,
994
+ "eval_steps_per_second": 3.998,
995
+ "step": 404285
996
  },
997
  {
998
+ "epoch": 35.11,
999
+ "learning_rate": 0.0003308881385930992,
1000
+ "loss": 0.0035,
1001
+ "step": 405556
1002
  },
1003
  {
1004
+ "epoch": 35.44,
1005
+ "learning_rate": 0.00032352754451274056,
1006
+ "loss": 0.0023,
1007
+ "step": 409382
 
 
 
 
1008
  },
1009
  {
1010
+ "epoch": 35.77,
1011
+ "learning_rate": 0.00031616695043238204,
1012
+ "loss": 0.0017,
1013
+ "step": 413208
1014
  },
1015
  {
1016
+ "epoch": 36.0,
1017
+ "eval_loss": 0.057597871869802475,
1018
+ "eval_max_distance": 14,
1019
+ "eval_mean_distance": 0,
1020
+ "eval_runtime": 6.2578,
1021
+ "eval_samples_per_second": 79.9,
1022
+ "eval_steps_per_second": 3.995,
1023
+ "step": 415836
1024
  },
1025
  {
1026
+ "epoch": 36.1,
1027
+ "learning_rate": 0.0003088063563520234,
1028
+ "loss": 0.0032,
1029
+ "step": 417034
1030
  },
1031
  {
1032
+ "epoch": 36.43,
1033
+ "learning_rate": 0.0003014457622716648,
1034
+ "loss": 0.0021,
1035
+ "step": 420860
1036
  },
1037
  {
1038
+ "epoch": 36.77,
1039
+ "learning_rate": 0.00029408516819130623,
1040
+ "loss": 0.0016,
1041
+ "step": 424686
1042
  },
1043
  {
1044
+ "epoch": 37.0,
1045
+ "eval_loss": 0.060558244585990906,
1046
+ "eval_max_distance": 16,
1047
+ "eval_mean_distance": 0,
1048
+ "eval_runtime": 6.2656,
1049
+ "eval_samples_per_second": 79.801,
1050
+ "eval_steps_per_second": 3.99,
1051
+ "step": 427387
1052
  },
1053
  {
1054
+ "epoch": 37.1,
1055
+ "learning_rate": 0.0002867245741109476,
1056
+ "loss": 0.003,
1057
+ "step": 428512
1058
  },
1059
  {
1060
+ "epoch": 37.43,
1061
+ "learning_rate": 0.000279363980030589,
1062
+ "loss": 0.0019,
1063
+ "step": 432338
1064
  },
1065
  {
1066
+ "epoch": 37.76,
1067
+ "learning_rate": 0.00027200338595023037,
1068
+ "loss": 0.0015,
1069
+ "step": 436164
1070
  },
1071
  {
1072
+ "epoch": 38.0,
1073
+ "eval_loss": 0.057046957314014435,
1074
+ "eval_max_distance": 24,
1075
  "eval_mean_distance": 0,
1076
+ "eval_runtime": 6.266,
1077
+ "eval_samples_per_second": 79.796,
1078
+ "eval_steps_per_second": 3.99,
1079
+ "step": 438938
1080
  },
1081
  {
1082
+ "epoch": 38.09,
1083
+ "learning_rate": 0.0002646427918698718,
1084
+ "loss": 0.0026,
1085
+ "step": 439990
 
 
 
 
 
 
1086
  },
1087
  {
1088
+ "epoch": 38.42,
1089
+ "learning_rate": 0.0002572821977895132,
1090
+ "loss": 0.0018,
1091
+ "step": 443816
1092
  },
1093
  {
1094
+ "epoch": 38.75,
1095
+ "learning_rate": 0.00024992160370915456,
1096
+ "loss": 0.0013,
1097
+ "step": 447642
1098
  },
1099
  {
1100
+ "epoch": 39.0,
1101
+ "eval_loss": 0.059430696070194244,
1102
+ "eval_max_distance": 24,
1103
+ "eval_mean_distance": 0,
1104
+ "eval_runtime": 6.2947,
1105
+ "eval_samples_per_second": 79.432,
1106
+ "eval_steps_per_second": 3.972,
1107
+ "step": 450489
1108
  },
1109
  {
1110
+ "epoch": 39.08,
1111
+ "learning_rate": 0.00024256100962879597,
1112
+ "loss": 0.0024,
1113
+ "step": 451468
1114
  },
1115
  {
1116
+ "epoch": 39.42,
1117
+ "learning_rate": 0.00023520041554843737,
1118
+ "loss": 0.0017,
1119
+ "step": 455294
1120
  },
1121
  {
1122
+ "epoch": 39.75,
1123
+ "learning_rate": 0.00022783982146807878,
1124
+ "loss": 0.0012,
1125
+ "step": 459120
1126
  },
1127
  {
1128
+ "epoch": 40.0,
1129
+ "eval_loss": 0.06027701497077942,
1130
+ "eval_max_distance": 24,
1131
+ "eval_mean_distance": 0,
1132
+ "eval_runtime": 6.2086,
1133
+ "eval_samples_per_second": 80.533,
1134
+ "eval_steps_per_second": 4.027,
1135
+ "step": 462040
1136
  },
1137
  {
1138
+ "epoch": 40.08,
1139
+ "learning_rate": 0.00022047922738772016,
1140
+ "loss": 0.0021,
1141
+ "step": 462946
 
 
 
 
1142
  },
1143
  {
1144
+ "epoch": 40.41,
1145
+ "learning_rate": 0.00021311863330736156,
1146
+ "loss": 0.0015,
1147
+ "step": 466772
1148
  },
1149
  {
1150
+ "epoch": 40.74,
1151
+ "learning_rate": 0.00020575803922700294,
1152
+ "loss": 0.001,
1153
+ "step": 470598
1154
  },
1155
  {
1156
+ "epoch": 41.0,
1157
+ "eval_loss": 0.06419633328914642,
1158
+ "eval_max_distance": 20,
1159
+ "eval_mean_distance": 0,
1160
+ "eval_runtime": 6.2439,
1161
+ "eval_samples_per_second": 80.078,
1162
+ "eval_steps_per_second": 4.004,
1163
+ "step": 473591
1164
  },
1165
  {
1166
+ "epoch": 41.07,
1167
+ "learning_rate": 0.00019839744514664435,
1168
+ "loss": 0.0019,
1169
+ "step": 474424
1170
  },
1171
  {
1172
+ "epoch": 41.4,
1173
+ "learning_rate": 0.00019103685106628573,
1174
+ "loss": 0.0014,
1175
+ "step": 478250
1176
  },
1177
  {
1178
+ "epoch": 41.73,
1179
+ "learning_rate": 0.00018367625698592716,
1180
+ "loss": 0.0009,
1181
+ "step": 482076
1182
  },
1183
  {
1184
+ "epoch": 42.0,
1185
+ "eval_loss": 0.06380578130483627,
1186
+ "eval_max_distance": 29,
1187
+ "eval_mean_distance": 0,
1188
+ "eval_runtime": 6.2294,
1189
+ "eval_samples_per_second": 80.264,
1190
+ "eval_steps_per_second": 4.013,
1191
+ "step": 485142
1192
  },
1193
  {
1194
+ "epoch": 42.07,
1195
+ "learning_rate": 0.00017631566290556854,
1196
+ "loss": 0.0016,
1197
+ "step": 485902
1198
  },
1199
  {
1200
+ "epoch": 42.4,
1201
+ "learning_rate": 0.00016895506882520995,
1202
+ "loss": 0.0013,
1203
+ "step": 489728
 
 
 
 
1204
  },
1205
  {
1206
+ "epoch": 42.73,
1207
+ "learning_rate": 0.00016159447474485133,
1208
+ "loss": 0.0008,
1209
+ "step": 493554
1210
  },
1211
  {
1212
+ "epoch": 43.0,
1213
+ "eval_loss": 0.0659068301320076,
1214
+ "eval_max_distance": 29,
1215
+ "eval_mean_distance": 0,
1216
+ "eval_runtime": 6.2148,
1217
+ "eval_samples_per_second": 80.454,
1218
+ "eval_steps_per_second": 4.023,
1219
+ "step": 496693
1220
  },
1221
  {
1222
+ "epoch": 43.06,
1223
+ "learning_rate": 0.00015423388066449273,
1224
+ "loss": 0.0014,
1225
+ "step": 497380
1226
  },
1227
  {
1228
+ "epoch": 43.39,
1229
+ "learning_rate": 0.00014687328658413414,
1230
+ "loss": 0.0012,
1231
+ "step": 501206
1232
  },
1233
  {
1234
+ "epoch": 43.72,
1235
+ "learning_rate": 0.00013951269250377552,
1236
+ "loss": 0.0007,
1237
+ "step": 505032
1238
  },
1239
  {
1240
+ "epoch": 44.0,
1241
+ "eval_loss": 0.06715985387563705,
1242
+ "eval_max_distance": 29,
1243
+ "eval_mean_distance": 0,
1244
+ "eval_runtime": 6.224,
1245
+ "eval_samples_per_second": 80.334,
1246
+ "eval_steps_per_second": 4.017,
1247
+ "step": 508244
1248
  },
1249
  {
1250
+ "epoch": 44.05,
1251
+ "learning_rate": 0.00013215209842341695,
1252
+ "loss": 0.0012,
1253
+ "step": 508858
1254
  },
1255
  {
1256
+ "epoch": 44.38,
1257
+ "learning_rate": 0.00012479150434305833,
1258
+ "loss": 0.0011,
1259
+ "step": 512684
1260
  },
1261
  {
1262
+ "epoch": 44.72,
1263
+ "learning_rate": 0.00011743091026269972,
1264
+ "loss": 0.0007,
1265
+ "step": 516510
1266
  },
1267
  {
1268
+ "epoch": 45.0,
1269
+ "eval_loss": 0.06622961163520813,
1270
+ "eval_max_distance": 19,
1271
  "eval_mean_distance": 0,
1272
+ "eval_runtime": 6.2443,
1273
+ "eval_samples_per_second": 80.072,
1274
+ "eval_steps_per_second": 4.004,
1275
+ "step": 519795
1276
  },
1277
  {
1278
+ "epoch": 45.05,
1279
+ "learning_rate": 0.00011007031618234111,
1280
+ "loss": 0.0012,
1281
+ "step": 520336
1282
  },
1283
  {
1284
+ "epoch": 45.38,
1285
+ "learning_rate": 0.0001027097221019825,
1286
+ "loss": 0.001,
1287
+ "step": 524162
1288
  },
1289
  {
1290
+ "epoch": 45.71,
1291
+ "learning_rate": 9.534912802162391e-05,
1292
+ "loss": 0.0006,
1293
+ "step": 527988
1294
  },
1295
  {
1296
+ "epoch": 46.0,
1297
+ "eval_loss": 0.06662679463624954,
1298
+ "eval_max_distance": 24,
1299
+ "eval_mean_distance": 0,
1300
+ "eval_runtime": 6.231,
1301
+ "eval_samples_per_second": 80.244,
1302
+ "eval_steps_per_second": 4.012,
1303
+ "step": 531346
1304
  },
1305
  {
1306
+ "epoch": 46.04,
1307
+ "learning_rate": 8.79885339412653e-05,
1308
+ "loss": 0.001,
1309
+ "step": 531814
1310
  },
1311
  {
1312
+ "epoch": 46.37,
1313
+ "learning_rate": 8.062793986090671e-05,
1314
+ "loss": 0.0009,
1315
+ "step": 535640
1316
  },
1317
  {
1318
+ "epoch": 46.7,
1319
+ "learning_rate": 7.32673457805481e-05,
1320
+ "loss": 0.0006,
1321
+ "step": 539466
1322
  },
1323
  {
1324
+ "epoch": 47.0,
1325
+ "eval_loss": 0.06897546350955963,
1326
+ "eval_max_distance": 22,
1327
+ "eval_mean_distance": 0,
1328
+ "eval_runtime": 6.1748,
1329
+ "eval_samples_per_second": 80.974,
1330
+ "eval_steps_per_second": 4.049,
1331
+ "step": 542897
1332
  },
1333
  {
1334
+ "epoch": 47.03,
1335
+ "learning_rate": 6.590675170018951e-05,
1336
+ "loss": 0.0008,
1337
+ "step": 543292
1338
  },
1339
  {
1340
+ "epoch": 47.37,
1341
+ "learning_rate": 5.8546157619830895e-05,
1342
+ "loss": 0.0009,
1343
+ "step": 547118
 
 
 
 
1344
  },
1345
  {
1346
+ "epoch": 47.7,
1347
+ "learning_rate": 5.118556353947229e-05,
1348
+ "loss": 0.0005,
1349
+ "step": 550944
1350
  },
1351
  {
1352
+ "epoch": 48.0,
1353
+ "eval_loss": 0.06699105352163315,
1354
+ "eval_max_distance": 15,
1355
+ "eval_mean_distance": 0,
1356
+ "eval_runtime": 6.2182,
1357
+ "eval_samples_per_second": 80.408,
1358
+ "eval_steps_per_second": 4.02,
1359
+ "step": 554448
1360
  },
1361
  {
1362
+ "epoch": 48.03,
1363
+ "learning_rate": 4.3824969459113694e-05,
1364
+ "loss": 0.0007,
1365
+ "step": 554770
1366
  },
1367
  {
1368
+ "epoch": 48.36,
1369
+ "learning_rate": 3.6464375378755086e-05,
1370
+ "loss": 0.0008,
1371
+ "step": 558596
1372
  },
1373
  {
1374
+ "epoch": 48.69,
1375
+ "learning_rate": 2.9103781298396486e-05,
1376
+ "loss": 0.0005,
1377
+ "step": 562422
1378
  },
1379
  {
1380
+ "epoch": 49.0,
1381
+ "eval_loss": 0.06619732826948166,
1382
+ "eval_max_distance": 18,
1383
+ "eval_mean_distance": 0,
1384
+ "eval_runtime": 6.1996,
1385
+ "eval_samples_per_second": 80.65,
1386
+ "eval_steps_per_second": 4.033,
1387
+ "step": 565999
1388
  },
1389
  {
1390
+ "epoch": 49.02,
1391
+ "learning_rate": 2.174318721803788e-05,
1392
+ "loss": 0.0006,
1393
+ "step": 566248
1394
  },
1395
  {
1396
+ "epoch": 49.35,
1397
+ "learning_rate": 1.4382593137679277e-05,
1398
+ "loss": 0.0008,
1399
+ "step": 570074
1400
  },
1401
  {
1402
+ "epoch": 49.68,
1403
+ "learning_rate": 7.021999057320674e-06,
1404
+ "loss": 0.0004,
1405
+ "step": 573900
1406
  },
1407
  {
1408
+ "epoch": 50.0,
1409
+ "eval_loss": 0.06643164157867432,
1410
+ "eval_max_distance": 18,
1411
  "eval_mean_distance": 0,
1412
+ "eval_runtime": 6.1338,
1413
+ "eval_samples_per_second": 81.515,
1414
+ "eval_steps_per_second": 4.076,
1415
+ "step": 577550
1416
+ },
1417
+ {
1418
+ "epoch": 50.0,
1419
+ "step": 577550,
1420
+ "total_flos": 1.5364586760247296e+17,
1421
+ "train_loss": 0.0020910642477260833,
1422
+ "train_runtime": 25597.0641,
1423
+ "train_samples_per_second": 451.228,
1424
+ "train_steps_per_second": 22.563
1425
  }
1426
  ],
1427
+ "logging_steps": 3826,
1428
+ "max_steps": 577550,
1429
+ "num_train_epochs": 50,
1430
+ "save_steps": 7651,
1431
+ "total_flos": 1.5364586760247296e+17,
1432
  "trial_name": null,
1433
  "trial_params": null
1434
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8d22aeb992bd9320e28438aa9818e67e4f3f8675fde0b30b307f79b248f69c20
3
  size 4091
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a64071366e9d286f33b1e765c4a984a49be987c34b012df2ba201216f721839e
3
  size 4091