chavinlo commited on
Commit
b07cccb
1 Parent(s): eca4c40

Upload checkpoint-200/trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. checkpoint-200/trainer_state.json +1216 -0
checkpoint-200/trainer_state.json ADDED
@@ -0,0 +1,1216 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.4921562596124269,
5
+ "global_step": 200,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.0,
12
+ "learning_rate": 5.405405405405406e-07,
13
+ "loss": 1.5039,
14
+ "step": 1
15
+ },
16
+ {
17
+ "epoch": 0.0,
18
+ "learning_rate": 1.0810810810810812e-06,
19
+ "loss": 1.5785,
20
+ "step": 2
21
+ },
22
+ {
23
+ "epoch": 0.01,
24
+ "learning_rate": 1.6216216216216219e-06,
25
+ "loss": 1.2981,
26
+ "step": 3
27
+ },
28
+ {
29
+ "epoch": 0.01,
30
+ "learning_rate": 2.1621621621621623e-06,
31
+ "loss": 1.1631,
32
+ "step": 4
33
+ },
34
+ {
35
+ "epoch": 0.01,
36
+ "learning_rate": 2.702702702702703e-06,
37
+ "loss": 1.2493,
38
+ "step": 5
39
+ },
40
+ {
41
+ "epoch": 0.01,
42
+ "learning_rate": 3.2432432432432437e-06,
43
+ "loss": 1.2413,
44
+ "step": 6
45
+ },
46
+ {
47
+ "epoch": 0.02,
48
+ "learning_rate": 3.7837837837837844e-06,
49
+ "loss": 1.2368,
50
+ "step": 7
51
+ },
52
+ {
53
+ "epoch": 0.02,
54
+ "learning_rate": 4.324324324324325e-06,
55
+ "loss": 1.2152,
56
+ "step": 8
57
+ },
58
+ {
59
+ "epoch": 0.02,
60
+ "learning_rate": 4.864864864864866e-06,
61
+ "loss": 1.1487,
62
+ "step": 9
63
+ },
64
+ {
65
+ "epoch": 0.02,
66
+ "learning_rate": 5.405405405405406e-06,
67
+ "loss": 1.116,
68
+ "step": 10
69
+ },
70
+ {
71
+ "epoch": 0.03,
72
+ "learning_rate": 5.945945945945947e-06,
73
+ "loss": 1.179,
74
+ "step": 11
75
+ },
76
+ {
77
+ "epoch": 0.03,
78
+ "learning_rate": 6.486486486486487e-06,
79
+ "loss": 1.2134,
80
+ "step": 12
81
+ },
82
+ {
83
+ "epoch": 0.03,
84
+ "learning_rate": 7.027027027027028e-06,
85
+ "loss": 1.1641,
86
+ "step": 13
87
+ },
88
+ {
89
+ "epoch": 0.03,
90
+ "learning_rate": 7.567567567567569e-06,
91
+ "loss": 1.0623,
92
+ "step": 14
93
+ },
94
+ {
95
+ "epoch": 0.04,
96
+ "learning_rate": 8.108108108108109e-06,
97
+ "loss": 1.0922,
98
+ "step": 15
99
+ },
100
+ {
101
+ "epoch": 0.04,
102
+ "learning_rate": 8.64864864864865e-06,
103
+ "loss": 1.0979,
104
+ "step": 16
105
+ },
106
+ {
107
+ "epoch": 0.04,
108
+ "learning_rate": 9.189189189189191e-06,
109
+ "loss": 1.1572,
110
+ "step": 17
111
+ },
112
+ {
113
+ "epoch": 0.04,
114
+ "learning_rate": 9.729729729729732e-06,
115
+ "loss": 1.0788,
116
+ "step": 18
117
+ },
118
+ {
119
+ "epoch": 0.05,
120
+ "learning_rate": 1.027027027027027e-05,
121
+ "loss": 1.1467,
122
+ "step": 19
123
+ },
124
+ {
125
+ "epoch": 0.05,
126
+ "learning_rate": 1.0810810810810812e-05,
127
+ "loss": 1.0852,
128
+ "step": 20
129
+ },
130
+ {
131
+ "epoch": 0.05,
132
+ "learning_rate": 1.1351351351351352e-05,
133
+ "loss": 1.1176,
134
+ "step": 21
135
+ },
136
+ {
137
+ "epoch": 0.05,
138
+ "learning_rate": 1.1891891891891894e-05,
139
+ "loss": 1.1049,
140
+ "step": 22
141
+ },
142
+ {
143
+ "epoch": 0.06,
144
+ "learning_rate": 1.2432432432432433e-05,
145
+ "loss": 1.0972,
146
+ "step": 23
147
+ },
148
+ {
149
+ "epoch": 0.06,
150
+ "learning_rate": 1.2972972972972975e-05,
151
+ "loss": 1.1404,
152
+ "step": 24
153
+ },
154
+ {
155
+ "epoch": 0.06,
156
+ "learning_rate": 1.3513513513513515e-05,
157
+ "loss": 1.1416,
158
+ "step": 25
159
+ },
160
+ {
161
+ "epoch": 0.06,
162
+ "learning_rate": 1.4054054054054055e-05,
163
+ "loss": 1.1458,
164
+ "step": 26
165
+ },
166
+ {
167
+ "epoch": 0.07,
168
+ "learning_rate": 1.4594594594594596e-05,
169
+ "loss": 1.1497,
170
+ "step": 27
171
+ },
172
+ {
173
+ "epoch": 0.07,
174
+ "learning_rate": 1.5135135135135138e-05,
175
+ "loss": 1.14,
176
+ "step": 28
177
+ },
178
+ {
179
+ "epoch": 0.07,
180
+ "learning_rate": 1.5675675675675676e-05,
181
+ "loss": 1.0808,
182
+ "step": 29
183
+ },
184
+ {
185
+ "epoch": 0.07,
186
+ "learning_rate": 1.6216216216216218e-05,
187
+ "loss": 1.0996,
188
+ "step": 30
189
+ },
190
+ {
191
+ "epoch": 0.08,
192
+ "learning_rate": 1.6756756756756757e-05,
193
+ "loss": 1.1064,
194
+ "step": 31
195
+ },
196
+ {
197
+ "epoch": 0.08,
198
+ "learning_rate": 1.72972972972973e-05,
199
+ "loss": 1.1823,
200
+ "step": 32
201
+ },
202
+ {
203
+ "epoch": 0.08,
204
+ "learning_rate": 1.783783783783784e-05,
205
+ "loss": 1.0483,
206
+ "step": 33
207
+ },
208
+ {
209
+ "epoch": 0.08,
210
+ "learning_rate": 1.8378378378378383e-05,
211
+ "loss": 1.1019,
212
+ "step": 34
213
+ },
214
+ {
215
+ "epoch": 0.09,
216
+ "learning_rate": 1.891891891891892e-05,
217
+ "loss": 1.0325,
218
+ "step": 35
219
+ },
220
+ {
221
+ "epoch": 0.09,
222
+ "learning_rate": 1.9459459459459463e-05,
223
+ "loss": 1.1006,
224
+ "step": 36
225
+ },
226
+ {
227
+ "epoch": 0.09,
228
+ "learning_rate": 2e-05,
229
+ "loss": 1.0715,
230
+ "step": 37
231
+ },
232
+ {
233
+ "epoch": 0.09,
234
+ "learning_rate": 1.999996461903301e-05,
235
+ "loss": 1.1542,
236
+ "step": 38
237
+ },
238
+ {
239
+ "epoch": 0.1,
240
+ "learning_rate": 1.9999858476382388e-05,
241
+ "loss": 1.1402,
242
+ "step": 39
243
+ },
244
+ {
245
+ "epoch": 0.1,
246
+ "learning_rate": 1.9999681572799226e-05,
247
+ "loss": 1.0163,
248
+ "step": 40
249
+ },
250
+ {
251
+ "epoch": 0.1,
252
+ "learning_rate": 1.9999433909535333e-05,
253
+ "loss": 1.0851,
254
+ "step": 41
255
+ },
256
+ {
257
+ "epoch": 0.1,
258
+ "learning_rate": 1.9999115488343213e-05,
259
+ "loss": 1.1119,
260
+ "step": 42
261
+ },
262
+ {
263
+ "epoch": 0.11,
264
+ "learning_rate": 1.999872631147608e-05,
265
+ "loss": 1.0726,
266
+ "step": 43
267
+ },
268
+ {
269
+ "epoch": 0.11,
270
+ "learning_rate": 1.999826638168783e-05,
271
+ "loss": 1.0927,
272
+ "step": 44
273
+ },
274
+ {
275
+ "epoch": 0.11,
276
+ "learning_rate": 1.9997735702233006e-05,
277
+ "loss": 1.0697,
278
+ "step": 45
279
+ },
280
+ {
281
+ "epoch": 0.11,
282
+ "learning_rate": 1.99971342768668e-05,
283
+ "loss": 1.0363,
284
+ "step": 46
285
+ },
286
+ {
287
+ "epoch": 0.12,
288
+ "learning_rate": 1.999646210984502e-05,
289
+ "loss": 1.1251,
290
+ "step": 47
291
+ },
292
+ {
293
+ "epoch": 0.12,
294
+ "learning_rate": 1.999571920592405e-05,
295
+ "loss": 1.127,
296
+ "step": 48
297
+ },
298
+ {
299
+ "epoch": 0.12,
300
+ "learning_rate": 1.9994905570360817e-05,
301
+ "loss": 1.0994,
302
+ "step": 49
303
+ },
304
+ {
305
+ "epoch": 0.12,
306
+ "learning_rate": 1.999402120891276e-05,
307
+ "loss": 1.0688,
308
+ "step": 50
309
+ },
310
+ {
311
+ "epoch": 0.13,
312
+ "learning_rate": 1.99930661278378e-05,
313
+ "loss": 1.0544,
314
+ "step": 51
315
+ },
316
+ {
317
+ "epoch": 0.13,
318
+ "learning_rate": 1.9992040333894273e-05,
319
+ "loss": 1.0908,
320
+ "step": 52
321
+ },
322
+ {
323
+ "epoch": 0.13,
324
+ "learning_rate": 1.9990943834340893e-05,
325
+ "loss": 1.1394,
326
+ "step": 53
327
+ },
328
+ {
329
+ "epoch": 0.13,
330
+ "learning_rate": 1.9989776636936705e-05,
331
+ "loss": 1.1065,
332
+ "step": 54
333
+ },
334
+ {
335
+ "epoch": 0.14,
336
+ "learning_rate": 1.9988538749941024e-05,
337
+ "loss": 1.0752,
338
+ "step": 55
339
+ },
340
+ {
341
+ "epoch": 0.14,
342
+ "learning_rate": 1.9987230182113374e-05,
343
+ "loss": 1.076,
344
+ "step": 56
345
+ },
346
+ {
347
+ "epoch": 0.14,
348
+ "learning_rate": 1.998585094271344e-05,
349
+ "loss": 1.0504,
350
+ "step": 57
351
+ },
352
+ {
353
+ "epoch": 0.14,
354
+ "learning_rate": 1.998440104150098e-05,
355
+ "loss": 1.0934,
356
+ "step": 58
357
+ },
358
+ {
359
+ "epoch": 0.15,
360
+ "learning_rate": 1.998288048873578e-05,
361
+ "loss": 1.1355,
362
+ "step": 59
363
+ },
364
+ {
365
+ "epoch": 0.15,
366
+ "learning_rate": 1.9981289295177566e-05,
367
+ "loss": 1.1693,
368
+ "step": 60
369
+ },
370
+ {
371
+ "epoch": 0.15,
372
+ "learning_rate": 1.9979627472085927e-05,
373
+ "loss": 1.0514,
374
+ "step": 61
375
+ },
376
+ {
377
+ "epoch": 0.15,
378
+ "learning_rate": 1.997789503122025e-05,
379
+ "loss": 1.119,
380
+ "step": 62
381
+ },
382
+ {
383
+ "epoch": 0.16,
384
+ "learning_rate": 1.9976091984839616e-05,
385
+ "loss": 1.0635,
386
+ "step": 63
387
+ },
388
+ {
389
+ "epoch": 0.16,
390
+ "learning_rate": 1.9974218345702733e-05,
391
+ "loss": 1.1111,
392
+ "step": 64
393
+ },
394
+ {
395
+ "epoch": 0.16,
396
+ "learning_rate": 1.9972274127067838e-05,
397
+ "loss": 1.1732,
398
+ "step": 65
399
+ },
400
+ {
401
+ "epoch": 0.16,
402
+ "learning_rate": 1.997025934269259e-05,
403
+ "loss": 1.149,
404
+ "step": 66
405
+ },
406
+ {
407
+ "epoch": 0.16,
408
+ "learning_rate": 1.9968174006833996e-05,
409
+ "loss": 1.1011,
410
+ "step": 67
411
+ },
412
+ {
413
+ "epoch": 0.17,
414
+ "learning_rate": 1.9966018134248296e-05,
415
+ "loss": 1.1367,
416
+ "step": 68
417
+ },
418
+ {
419
+ "epoch": 0.17,
420
+ "learning_rate": 1.9963791740190863e-05,
421
+ "loss": 1.1354,
422
+ "step": 69
423
+ },
424
+ {
425
+ "epoch": 0.17,
426
+ "learning_rate": 1.996149484041609e-05,
427
+ "loss": 1.0343,
428
+ "step": 70
429
+ },
430
+ {
431
+ "epoch": 0.17,
432
+ "learning_rate": 1.9959127451177287e-05,
433
+ "loss": 1.1997,
434
+ "step": 71
435
+ },
436
+ {
437
+ "epoch": 0.18,
438
+ "learning_rate": 1.9956689589226555e-05,
439
+ "loss": 1.1131,
440
+ "step": 72
441
+ },
442
+ {
443
+ "epoch": 0.18,
444
+ "learning_rate": 1.9954181271814673e-05,
445
+ "loss": 1.0335,
446
+ "step": 73
447
+ },
448
+ {
449
+ "epoch": 0.18,
450
+ "learning_rate": 1.9951602516690988e-05,
451
+ "loss": 1.0786,
452
+ "step": 74
453
+ },
454
+ {
455
+ "epoch": 0.18,
456
+ "learning_rate": 1.9948953342103268e-05,
457
+ "loss": 1.0926,
458
+ "step": 75
459
+ },
460
+ {
461
+ "epoch": 0.19,
462
+ "learning_rate": 1.994623376679758e-05,
463
+ "loss": 1.1353,
464
+ "step": 76
465
+ },
466
+ {
467
+ "epoch": 0.19,
468
+ "learning_rate": 1.9943443810018174e-05,
469
+ "loss": 1.0605,
470
+ "step": 77
471
+ },
472
+ {
473
+ "epoch": 0.19,
474
+ "learning_rate": 1.9940583491507314e-05,
475
+ "loss": 1.1255,
476
+ "step": 78
477
+ },
478
+ {
479
+ "epoch": 0.19,
480
+ "learning_rate": 1.993765283150517e-05,
481
+ "loss": 1.0541,
482
+ "step": 79
483
+ },
484
+ {
485
+ "epoch": 0.2,
486
+ "learning_rate": 1.9934651850749663e-05,
487
+ "loss": 1.1197,
488
+ "step": 80
489
+ },
490
+ {
491
+ "epoch": 0.2,
492
+ "learning_rate": 1.9931580570476306e-05,
493
+ "loss": 1.1297,
494
+ "step": 81
495
+ },
496
+ {
497
+ "epoch": 0.2,
498
+ "learning_rate": 1.9928439012418076e-05,
499
+ "loss": 1.1324,
500
+ "step": 82
501
+ },
502
+ {
503
+ "epoch": 0.2,
504
+ "learning_rate": 1.9925227198805247e-05,
505
+ "loss": 1.0846,
506
+ "step": 83
507
+ },
508
+ {
509
+ "epoch": 0.21,
510
+ "learning_rate": 1.9921945152365235e-05,
511
+ "loss": 1.1831,
512
+ "step": 84
513
+ },
514
+ {
515
+ "epoch": 0.21,
516
+ "learning_rate": 1.9918592896322432e-05,
517
+ "loss": 1.0921,
518
+ "step": 85
519
+ },
520
+ {
521
+ "epoch": 0.21,
522
+ "learning_rate": 1.9915170454398045e-05,
523
+ "loss": 1.1037,
524
+ "step": 86
525
+ },
526
+ {
527
+ "epoch": 0.21,
528
+ "learning_rate": 1.9911677850809943e-05,
529
+ "loss": 1.125,
530
+ "step": 87
531
+ },
532
+ {
533
+ "epoch": 0.22,
534
+ "learning_rate": 1.9908115110272463e-05,
535
+ "loss": 1.1279,
536
+ "step": 88
537
+ },
538
+ {
539
+ "epoch": 0.22,
540
+ "learning_rate": 1.9904482257996244e-05,
541
+ "loss": 1.1427,
542
+ "step": 89
543
+ },
544
+ {
545
+ "epoch": 0.22,
546
+ "learning_rate": 1.990077931968805e-05,
547
+ "loss": 1.1475,
548
+ "step": 90
549
+ },
550
+ {
551
+ "epoch": 0.22,
552
+ "learning_rate": 1.9897006321550592e-05,
553
+ "loss": 1.136,
554
+ "step": 91
555
+ },
556
+ {
557
+ "epoch": 0.23,
558
+ "learning_rate": 1.9893163290282335e-05,
559
+ "loss": 1.0469,
560
+ "step": 92
561
+ },
562
+ {
563
+ "epoch": 0.23,
564
+ "learning_rate": 1.9889250253077306e-05,
565
+ "loss": 1.1028,
566
+ "step": 93
567
+ },
568
+ {
569
+ "epoch": 0.23,
570
+ "learning_rate": 1.9885267237624923e-05,
571
+ "loss": 1.1775,
572
+ "step": 94
573
+ },
574
+ {
575
+ "epoch": 0.23,
576
+ "learning_rate": 1.988121427210976e-05,
577
+ "loss": 1.1744,
578
+ "step": 95
579
+ },
580
+ {
581
+ "epoch": 0.24,
582
+ "learning_rate": 1.98770913852114e-05,
583
+ "loss": 1.1049,
584
+ "step": 96
585
+ },
586
+ {
587
+ "epoch": 0.24,
588
+ "learning_rate": 1.9872898606104175e-05,
589
+ "loss": 1.1085,
590
+ "step": 97
591
+ },
592
+ {
593
+ "epoch": 0.24,
594
+ "learning_rate": 1.9868635964457007e-05,
595
+ "loss": 1.1153,
596
+ "step": 98
597
+ },
598
+ {
599
+ "epoch": 0.24,
600
+ "learning_rate": 1.986430349043317e-05,
601
+ "loss": 1.1637,
602
+ "step": 99
603
+ },
604
+ {
605
+ "epoch": 0.25,
606
+ "learning_rate": 1.9859901214690094e-05,
607
+ "loss": 1.0971,
608
+ "step": 100
609
+ },
610
+ {
611
+ "epoch": 0.25,
612
+ "learning_rate": 1.9855429168379127e-05,
613
+ "loss": 1.1537,
614
+ "step": 101
615
+ },
616
+ {
617
+ "epoch": 0.25,
618
+ "learning_rate": 1.985088738314534e-05,
619
+ "loss": 1.0882,
620
+ "step": 102
621
+ },
622
+ {
623
+ "epoch": 0.25,
624
+ "learning_rate": 1.9846275891127275e-05,
625
+ "loss": 1.128,
626
+ "step": 103
627
+ },
628
+ {
629
+ "epoch": 0.26,
630
+ "learning_rate": 1.9841594724956746e-05,
631
+ "loss": 1.14,
632
+ "step": 104
633
+ },
634
+ {
635
+ "epoch": 0.26,
636
+ "learning_rate": 1.9836843917758593e-05,
637
+ "loss": 1.0273,
638
+ "step": 105
639
+ },
640
+ {
641
+ "epoch": 0.26,
642
+ "learning_rate": 1.983202350315044e-05,
643
+ "loss": 1.0891,
644
+ "step": 106
645
+ },
646
+ {
647
+ "epoch": 0.26,
648
+ "learning_rate": 1.982713351524248e-05,
649
+ "loss": 1.106,
650
+ "step": 107
651
+ },
652
+ {
653
+ "epoch": 0.27,
654
+ "learning_rate": 1.982217398863721e-05,
655
+ "loss": 1.0599,
656
+ "step": 108
657
+ },
658
+ {
659
+ "epoch": 0.27,
660
+ "learning_rate": 1.98171449584292e-05,
661
+ "loss": 1.0396,
662
+ "step": 109
663
+ },
664
+ {
665
+ "epoch": 0.27,
666
+ "learning_rate": 1.9812046460204837e-05,
667
+ "loss": 1.0767,
668
+ "step": 110
669
+ },
670
+ {
671
+ "epoch": 0.27,
672
+ "learning_rate": 1.9806878530042083e-05,
673
+ "loss": 1.1568,
674
+ "step": 111
675
+ },
676
+ {
677
+ "epoch": 0.28,
678
+ "learning_rate": 1.9801641204510216e-05,
679
+ "loss": 1.1534,
680
+ "step": 112
681
+ },
682
+ {
683
+ "epoch": 0.28,
684
+ "learning_rate": 1.9796334520669555e-05,
685
+ "loss": 1.0332,
686
+ "step": 113
687
+ },
688
+ {
689
+ "epoch": 0.28,
690
+ "learning_rate": 1.9790958516071228e-05,
691
+ "loss": 1.1018,
692
+ "step": 114
693
+ },
694
+ {
695
+ "epoch": 0.28,
696
+ "learning_rate": 1.978551322875688e-05,
697
+ "loss": 1.0634,
698
+ "step": 115
699
+ },
700
+ {
701
+ "epoch": 0.29,
702
+ "learning_rate": 1.977999869725842e-05,
703
+ "loss": 1.0766,
704
+ "step": 116
705
+ },
706
+ {
707
+ "epoch": 0.29,
708
+ "learning_rate": 1.977441496059774e-05,
709
+ "loss": 1.0456,
710
+ "step": 117
711
+ },
712
+ {
713
+ "epoch": 0.29,
714
+ "learning_rate": 1.9768762058286433e-05,
715
+ "loss": 1.0047,
716
+ "step": 118
717
+ },
718
+ {
719
+ "epoch": 0.29,
720
+ "learning_rate": 1.976304003032554e-05,
721
+ "loss": 1.0798,
722
+ "step": 119
723
+ },
724
+ {
725
+ "epoch": 0.3,
726
+ "learning_rate": 1.9757248917205228e-05,
727
+ "loss": 1.1055,
728
+ "step": 120
729
+ },
730
+ {
731
+ "epoch": 0.3,
732
+ "learning_rate": 1.975138875990454e-05,
733
+ "loss": 1.1921,
734
+ "step": 121
735
+ },
736
+ {
737
+ "epoch": 0.3,
738
+ "learning_rate": 1.974545959989108e-05,
739
+ "loss": 1.0465,
740
+ "step": 122
741
+ },
742
+ {
743
+ "epoch": 0.3,
744
+ "learning_rate": 1.9739461479120727e-05,
745
+ "loss": 1.129,
746
+ "step": 123
747
+ },
748
+ {
749
+ "epoch": 0.31,
750
+ "learning_rate": 1.973339444003735e-05,
751
+ "loss": 1.0773,
752
+ "step": 124
753
+ },
754
+ {
755
+ "epoch": 0.31,
756
+ "learning_rate": 1.9727258525572487e-05,
757
+ "loss": 1.0955,
758
+ "step": 125
759
+ },
760
+ {
761
+ "epoch": 0.31,
762
+ "learning_rate": 1.9721053779145057e-05,
763
+ "loss": 1.0941,
764
+ "step": 126
765
+ },
766
+ {
767
+ "epoch": 0.31,
768
+ "learning_rate": 1.9714780244661044e-05,
769
+ "loss": 1.1672,
770
+ "step": 127
771
+ },
772
+ {
773
+ "epoch": 0.31,
774
+ "learning_rate": 1.9708437966513196e-05,
775
+ "loss": 1.1014,
776
+ "step": 128
777
+ },
778
+ {
779
+ "epoch": 0.32,
780
+ "learning_rate": 1.9702026989580694e-05,
781
+ "loss": 1.1272,
782
+ "step": 129
783
+ },
784
+ {
785
+ "epoch": 0.32,
786
+ "learning_rate": 1.969554735922885e-05,
787
+ "loss": 1.0398,
788
+ "step": 130
789
+ },
790
+ {
791
+ "epoch": 0.32,
792
+ "learning_rate": 1.968899912130879e-05,
793
+ "loss": 1.106,
794
+ "step": 131
795
+ },
796
+ {
797
+ "epoch": 0.32,
798
+ "learning_rate": 1.9682382322157103e-05,
799
+ "loss": 1.1033,
800
+ "step": 132
801
+ },
802
+ {
803
+ "epoch": 0.33,
804
+ "learning_rate": 1.9675697008595545e-05,
805
+ "loss": 1.0154,
806
+ "step": 133
807
+ },
808
+ {
809
+ "epoch": 0.33,
810
+ "learning_rate": 1.9668943227930686e-05,
811
+ "loss": 1.1219,
812
+ "step": 134
813
+ },
814
+ {
815
+ "epoch": 0.33,
816
+ "learning_rate": 1.966212102795358e-05,
817
+ "loss": 1.0824,
818
+ "step": 135
819
+ },
820
+ {
821
+ "epoch": 0.33,
822
+ "learning_rate": 1.965523045693944e-05,
823
+ "loss": 1.0581,
824
+ "step": 136
825
+ },
826
+ {
827
+ "epoch": 0.34,
828
+ "learning_rate": 1.964827156364728e-05,
829
+ "loss": 1.1327,
830
+ "step": 137
831
+ },
832
+ {
833
+ "epoch": 0.34,
834
+ "learning_rate": 1.964124439731957e-05,
835
+ "loss": 1.1144,
836
+ "step": 138
837
+ },
838
+ {
839
+ "epoch": 0.34,
840
+ "learning_rate": 1.9634149007681894e-05,
841
+ "loss": 1.1436,
842
+ "step": 139
843
+ },
844
+ {
845
+ "epoch": 0.34,
846
+ "learning_rate": 1.962698544494261e-05,
847
+ "loss": 1.0457,
848
+ "step": 140
849
+ },
850
+ {
851
+ "epoch": 0.35,
852
+ "learning_rate": 1.9619753759792466e-05,
853
+ "loss": 1.1053,
854
+ "step": 141
855
+ },
856
+ {
857
+ "epoch": 0.35,
858
+ "learning_rate": 1.961245400340427e-05,
859
+ "loss": 1.0881,
860
+ "step": 142
861
+ },
862
+ {
863
+ "epoch": 0.35,
864
+ "learning_rate": 1.9605086227432512e-05,
865
+ "loss": 1.1058,
866
+ "step": 143
867
+ },
868
+ {
869
+ "epoch": 0.35,
870
+ "learning_rate": 1.9597650484012997e-05,
871
+ "loss": 1.1418,
872
+ "step": 144
873
+ },
874
+ {
875
+ "epoch": 0.36,
876
+ "learning_rate": 1.9590146825762476e-05,
877
+ "loss": 1.0623,
878
+ "step": 145
879
+ },
880
+ {
881
+ "epoch": 0.36,
882
+ "learning_rate": 1.9582575305778297e-05,
883
+ "loss": 1.0987,
884
+ "step": 146
885
+ },
886
+ {
887
+ "epoch": 0.36,
888
+ "learning_rate": 1.9574935977637994e-05,
889
+ "loss": 1.0764,
890
+ "step": 147
891
+ },
892
+ {
893
+ "epoch": 0.36,
894
+ "learning_rate": 1.9567228895398936e-05,
895
+ "loss": 1.153,
896
+ "step": 148
897
+ },
898
+ {
899
+ "epoch": 0.37,
900
+ "learning_rate": 1.955945411359792e-05,
901
+ "loss": 1.1233,
902
+ "step": 149
903
+ },
904
+ {
905
+ "epoch": 0.37,
906
+ "learning_rate": 1.9551611687250808e-05,
907
+ "loss": 1.0827,
908
+ "step": 150
909
+ },
910
+ {
911
+ "epoch": 0.37,
912
+ "learning_rate": 1.9543701671852127e-05,
913
+ "loss": 1.0487,
914
+ "step": 151
915
+ },
916
+ {
917
+ "epoch": 0.37,
918
+ "learning_rate": 1.9535724123374674e-05,
919
+ "loss": 1.0795,
920
+ "step": 152
921
+ },
922
+ {
923
+ "epoch": 0.38,
924
+ "learning_rate": 1.952767909826913e-05,
925
+ "loss": 1.15,
926
+ "step": 153
927
+ },
928
+ {
929
+ "epoch": 0.38,
930
+ "learning_rate": 1.951956665346364e-05,
931
+ "loss": 1.0645,
932
+ "step": 154
933
+ },
934
+ {
935
+ "epoch": 0.38,
936
+ "learning_rate": 1.951138684636344e-05,
937
+ "loss": 1.0764,
938
+ "step": 155
939
+ },
940
+ {
941
+ "epoch": 0.38,
942
+ "learning_rate": 1.9503139734850426e-05,
943
+ "loss": 1.1185,
944
+ "step": 156
945
+ },
946
+ {
947
+ "epoch": 0.39,
948
+ "learning_rate": 1.9494825377282746e-05,
949
+ "loss": 1.0879,
950
+ "step": 157
951
+ },
952
+ {
953
+ "epoch": 0.39,
954
+ "learning_rate": 1.9486443832494414e-05,
955
+ "loss": 1.0558,
956
+ "step": 158
957
+ },
958
+ {
959
+ "epoch": 0.39,
960
+ "learning_rate": 1.9477995159794854e-05,
961
+ "loss": 1.1042,
962
+ "step": 159
963
+ },
964
+ {
965
+ "epoch": 0.39,
966
+ "learning_rate": 1.9469479418968506e-05,
967
+ "loss": 1.1029,
968
+ "step": 160
969
+ },
970
+ {
971
+ "epoch": 0.4,
972
+ "learning_rate": 1.9460896670274408e-05,
973
+ "loss": 0.9836,
974
+ "step": 161
975
+ },
976
+ {
977
+ "epoch": 0.4,
978
+ "learning_rate": 1.9452246974445743e-05,
979
+ "loss": 1.1325,
980
+ "step": 162
981
+ },
982
+ {
983
+ "epoch": 0.4,
984
+ "learning_rate": 1.9443530392689434e-05,
985
+ "loss": 1.1018,
986
+ "step": 163
987
+ },
988
+ {
989
+ "epoch": 0.4,
990
+ "learning_rate": 1.94347469866857e-05,
991
+ "loss": 1.1525,
992
+ "step": 164
993
+ },
994
+ {
995
+ "epoch": 0.41,
996
+ "learning_rate": 1.9425896818587615e-05,
997
+ "loss": 1.0719,
998
+ "step": 165
999
+ },
1000
+ {
1001
+ "epoch": 0.41,
1002
+ "learning_rate": 1.941697995102069e-05,
1003
+ "loss": 1.1595,
1004
+ "step": 166
1005
+ },
1006
+ {
1007
+ "epoch": 0.41,
1008
+ "learning_rate": 1.9407996447082394e-05,
1009
+ "loss": 1.1698,
1010
+ "step": 167
1011
+ },
1012
+ {
1013
+ "epoch": 0.41,
1014
+ "learning_rate": 1.939894637034174e-05,
1015
+ "loss": 1.131,
1016
+ "step": 168
1017
+ },
1018
+ {
1019
+ "epoch": 0.42,
1020
+ "learning_rate": 1.9389829784838833e-05,
1021
+ "loss": 1.0604,
1022
+ "step": 169
1023
+ },
1024
+ {
1025
+ "epoch": 0.42,
1026
+ "learning_rate": 1.938064675508438e-05,
1027
+ "loss": 1.0422,
1028
+ "step": 170
1029
+ },
1030
+ {
1031
+ "epoch": 0.42,
1032
+ "learning_rate": 1.9371397346059286e-05,
1033
+ "loss": 1.1098,
1034
+ "step": 171
1035
+ },
1036
+ {
1037
+ "epoch": 0.42,
1038
+ "learning_rate": 1.936208162321415e-05,
1039
+ "loss": 1.0682,
1040
+ "step": 172
1041
+ },
1042
+ {
1043
+ "epoch": 0.43,
1044
+ "learning_rate": 1.9352699652468835e-05,
1045
+ "loss": 1.0763,
1046
+ "step": 173
1047
+ },
1048
+ {
1049
+ "epoch": 0.43,
1050
+ "learning_rate": 1.9343251500211977e-05,
1051
+ "loss": 1.1038,
1052
+ "step": 174
1053
+ },
1054
+ {
1055
+ "epoch": 0.43,
1056
+ "learning_rate": 1.933373723330053e-05,
1057
+ "loss": 1.0613,
1058
+ "step": 175
1059
+ },
1060
+ {
1061
+ "epoch": 0.43,
1062
+ "learning_rate": 1.9324156919059286e-05,
1063
+ "loss": 1.1157,
1064
+ "step": 176
1065
+ },
1066
+ {
1067
+ "epoch": 0.44,
1068
+ "learning_rate": 1.93145106252804e-05,
1069
+ "loss": 1.1264,
1070
+ "step": 177
1071
+ },
1072
+ {
1073
+ "epoch": 0.44,
1074
+ "learning_rate": 1.9304798420222918e-05,
1075
+ "loss": 1.0787,
1076
+ "step": 178
1077
+ },
1078
+ {
1079
+ "epoch": 0.44,
1080
+ "learning_rate": 1.9295020372612276e-05,
1081
+ "loss": 1.1298,
1082
+ "step": 179
1083
+ },
1084
+ {
1085
+ "epoch": 0.44,
1086
+ "learning_rate": 1.9285176551639826e-05,
1087
+ "loss": 1.1175,
1088
+ "step": 180
1089
+ },
1090
+ {
1091
+ "epoch": 0.45,
1092
+ "learning_rate": 1.9275267026962358e-05,
1093
+ "loss": 1.1686,
1094
+ "step": 181
1095
+ },
1096
+ {
1097
+ "epoch": 0.45,
1098
+ "learning_rate": 1.9265291868701584e-05,
1099
+ "loss": 1.0939,
1100
+ "step": 182
1101
+ },
1102
+ {
1103
+ "epoch": 0.45,
1104
+ "learning_rate": 1.9255251147443646e-05,
1105
+ "loss": 1.1674,
1106
+ "step": 183
1107
+ },
1108
+ {
1109
+ "epoch": 0.45,
1110
+ "learning_rate": 1.924514493423864e-05,
1111
+ "loss": 1.0747,
1112
+ "step": 184
1113
+ },
1114
+ {
1115
+ "epoch": 0.46,
1116
+ "learning_rate": 1.9234973300600074e-05,
1117
+ "loss": 1.1085,
1118
+ "step": 185
1119
+ },
1120
+ {
1121
+ "epoch": 0.46,
1122
+ "learning_rate": 1.92247363185044e-05,
1123
+ "loss": 1.1157,
1124
+ "step": 186
1125
+ },
1126
+ {
1127
+ "epoch": 0.46,
1128
+ "learning_rate": 1.9214434060390484e-05,
1129
+ "loss": 1.1827,
1130
+ "step": 187
1131
+ },
1132
+ {
1133
+ "epoch": 0.46,
1134
+ "learning_rate": 1.9204066599159094e-05,
1135
+ "loss": 1.1126,
1136
+ "step": 188
1137
+ },
1138
+ {
1139
+ "epoch": 0.47,
1140
+ "learning_rate": 1.9193634008172396e-05,
1141
+ "loss": 1.1158,
1142
+ "step": 189
1143
+ },
1144
+ {
1145
+ "epoch": 0.47,
1146
+ "learning_rate": 1.9183136361253417e-05,
1147
+ "loss": 1.0609,
1148
+ "step": 190
1149
+ },
1150
+ {
1151
+ "epoch": 0.47,
1152
+ "learning_rate": 1.917257373268554e-05,
1153
+ "loss": 1.0617,
1154
+ "step": 191
1155
+ },
1156
+ {
1157
+ "epoch": 0.47,
1158
+ "learning_rate": 1.916194619721196e-05,
1159
+ "loss": 1.0724,
1160
+ "step": 192
1161
+ },
1162
+ {
1163
+ "epoch": 0.47,
1164
+ "learning_rate": 1.915125383003518e-05,
1165
+ "loss": 1.0784,
1166
+ "step": 193
1167
+ },
1168
+ {
1169
+ "epoch": 0.48,
1170
+ "learning_rate": 1.914049670681646e-05,
1171
+ "loss": 1.1141,
1172
+ "step": 194
1173
+ },
1174
+ {
1175
+ "epoch": 0.48,
1176
+ "learning_rate": 1.912967490367528e-05,
1177
+ "loss": 1.0305,
1178
+ "step": 195
1179
+ },
1180
+ {
1181
+ "epoch": 0.48,
1182
+ "learning_rate": 1.9118788497188815e-05,
1183
+ "loss": 1.026,
1184
+ "step": 196
1185
+ },
1186
+ {
1187
+ "epoch": 0.48,
1188
+ "learning_rate": 1.9107837564391376e-05,
1189
+ "loss": 1.0663,
1190
+ "step": 197
1191
+ },
1192
+ {
1193
+ "epoch": 0.49,
1194
+ "learning_rate": 1.9096822182773887e-05,
1195
+ "loss": 1.0368,
1196
+ "step": 198
1197
+ },
1198
+ {
1199
+ "epoch": 0.49,
1200
+ "learning_rate": 1.9085742430283322e-05,
1201
+ "loss": 1.1052,
1202
+ "step": 199
1203
+ },
1204
+ {
1205
+ "epoch": 0.49,
1206
+ "learning_rate": 1.907459838532215e-05,
1207
+ "loss": 1.0834,
1208
+ "step": 200
1209
+ }
1210
+ ],
1211
+ "max_steps": 1218,
1212
+ "num_train_epochs": 3,
1213
+ "total_flos": 4.842042281898803e+16,
1214
+ "trial_name": null,
1215
+ "trial_params": null
1216
+ }