Polly1231 commited on
Commit
e855c1d
·
verified ·
1 Parent(s): 89257fc

Upload trainer_state.json

Browse files
Files changed (1) hide show
  1. trainer_state.json +780 -0
trainer_state.json ADDED
@@ -0,0 +1,780 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 125,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 5e-05,
14
+ "loss": 1.8138,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.02,
19
+ "learning_rate": 0.0001,
20
+ "loss": 1.7936,
21
+ "step": 2
22
+ },
23
+ {
24
+ "epoch": 0.02,
25
+ "learning_rate": 0.00015000000000000001,
26
+ "loss": 1.7012,
27
+ "step": 3
28
+ },
29
+ {
30
+ "epoch": 0.03,
31
+ "learning_rate": 0.0002,
32
+ "loss": 1.3853,
33
+ "step": 4
34
+ },
35
+ {
36
+ "epoch": 0.04,
37
+ "learning_rate": 0.00019996629653035126,
38
+ "loss": 1.4042,
39
+ "step": 5
40
+ },
41
+ {
42
+ "epoch": 0.05,
43
+ "learning_rate": 0.00019986520883988232,
44
+ "loss": 1.4301,
45
+ "step": 6
46
+ },
47
+ {
48
+ "epoch": 0.06,
49
+ "learning_rate": 0.00019969680506871137,
50
+ "loss": 1.3754,
51
+ "step": 7
52
+ },
53
+ {
54
+ "epoch": 0.06,
55
+ "learning_rate": 0.00019946119873266613,
56
+ "loss": 1.3265,
57
+ "step": 8
58
+ },
59
+ {
60
+ "epoch": 0.07,
61
+ "learning_rate": 0.00019915854864676664,
62
+ "loss": 1.2392,
63
+ "step": 9
64
+ },
65
+ {
66
+ "epoch": 0.08,
67
+ "learning_rate": 0.00019878905881817252,
68
+ "loss": 1.344,
69
+ "step": 10
70
+ },
71
+ {
72
+ "epoch": 0.09,
73
+ "learning_rate": 0.00019835297830866826,
74
+ "loss": 1.2774,
75
+ "step": 11
76
+ },
77
+ {
78
+ "epoch": 0.1,
79
+ "learning_rate": 0.00019785060106677818,
80
+ "loss": 1.2003,
81
+ "step": 12
82
+ },
83
+ {
84
+ "epoch": 0.1,
85
+ "learning_rate": 0.00019728226572962473,
86
+ "loss": 1.1964,
87
+ "step": 13
88
+ },
89
+ {
90
+ "epoch": 0.11,
91
+ "learning_rate": 0.0001966483553946637,
92
+ "loss": 1.2569,
93
+ "step": 14
94
+ },
95
+ {
96
+ "epoch": 0.12,
97
+ "learning_rate": 0.00019594929736144976,
98
+ "loss": 1.1401,
99
+ "step": 15
100
+ },
101
+ {
102
+ "epoch": 0.13,
103
+ "learning_rate": 0.00019518556284360696,
104
+ "loss": 1.1552,
105
+ "step": 16
106
+ },
107
+ {
108
+ "epoch": 0.14,
109
+ "learning_rate": 0.0001943576666511982,
110
+ "loss": 1.2524,
111
+ "step": 17
112
+ },
113
+ {
114
+ "epoch": 0.14,
115
+ "learning_rate": 0.0001934661668437073,
116
+ "loss": 1.1609,
117
+ "step": 18
118
+ },
119
+ {
120
+ "epoch": 0.15,
121
+ "learning_rate": 0.0001925116643538684,
122
+ "loss": 1.2637,
123
+ "step": 19
124
+ },
125
+ {
126
+ "epoch": 0.16,
127
+ "learning_rate": 0.00019149480258259533,
128
+ "loss": 1.2292,
129
+ "step": 20
130
+ },
131
+ {
132
+ "epoch": 0.17,
133
+ "learning_rate": 0.00019041626696528503,
134
+ "loss": 1.1428,
135
+ "step": 21
136
+ },
137
+ {
138
+ "epoch": 0.18,
139
+ "learning_rate": 0.0001892767845097864,
140
+ "loss": 1.304,
141
+ "step": 22
142
+ },
143
+ {
144
+ "epoch": 0.18,
145
+ "learning_rate": 0.00018807712330634642,
146
+ "loss": 1.3038,
147
+ "step": 23
148
+ },
149
+ {
150
+ "epoch": 0.19,
151
+ "learning_rate": 0.0001868180920098644,
152
+ "loss": 1.1857,
153
+ "step": 24
154
+ },
155
+ {
156
+ "epoch": 0.2,
157
+ "learning_rate": 0.00018550053929480202,
158
+ "loss": 1.2526,
159
+ "step": 25
160
+ },
161
+ {
162
+ "epoch": 0.21,
163
+ "learning_rate": 0.00018412535328311814,
164
+ "loss": 1.1447,
165
+ "step": 26
166
+ },
167
+ {
168
+ "epoch": 0.22,
169
+ "learning_rate": 0.0001826934609456129,
170
+ "loss": 1.1373,
171
+ "step": 27
172
+ },
173
+ {
174
+ "epoch": 0.22,
175
+ "learning_rate": 0.00018120582747708502,
176
+ "loss": 1.1123,
177
+ "step": 28
178
+ },
179
+ {
180
+ "epoch": 0.23,
181
+ "learning_rate": 0.0001796634556457236,
182
+ "loss": 1.1393,
183
+ "step": 29
184
+ },
185
+ {
186
+ "epoch": 0.24,
187
+ "learning_rate": 0.0001780673851171728,
188
+ "loss": 1.059,
189
+ "step": 30
190
+ },
191
+ {
192
+ "epoch": 0.25,
193
+ "learning_rate": 0.00017641869175372493,
194
+ "loss": 1.1767,
195
+ "step": 31
196
+ },
197
+ {
198
+ "epoch": 0.26,
199
+ "learning_rate": 0.00017471848688911464,
200
+ "loss": 1.2007,
201
+ "step": 32
202
+ },
203
+ {
204
+ "epoch": 0.26,
205
+ "learning_rate": 0.000172967916579403,
206
+ "loss": 1.1403,
207
+ "step": 33
208
+ },
209
+ {
210
+ "epoch": 0.27,
211
+ "learning_rate": 0.00017116816083045602,
212
+ "loss": 1.1573,
213
+ "step": 34
214
+ },
215
+ {
216
+ "epoch": 0.28,
217
+ "learning_rate": 0.0001693204328025389,
218
+ "loss": 1.1436,
219
+ "step": 35
220
+ },
221
+ {
222
+ "epoch": 0.29,
223
+ "learning_rate": 0.00016742597799256182,
224
+ "loss": 1.1916,
225
+ "step": 36
226
+ },
227
+ {
228
+ "epoch": 0.3,
229
+ "learning_rate": 0.00016548607339452853,
230
+ "loss": 1.1979,
231
+ "step": 37
232
+ },
233
+ {
234
+ "epoch": 0.3,
235
+ "learning_rate": 0.00016350202663875386,
236
+ "loss": 1.108,
237
+ "step": 38
238
+ },
239
+ {
240
+ "epoch": 0.31,
241
+ "learning_rate": 0.0001614751751104301,
242
+ "loss": 1.1139,
243
+ "step": 39
244
+ },
245
+ {
246
+ "epoch": 0.32,
247
+ "learning_rate": 0.00015940688504813662,
248
+ "loss": 1.1229,
249
+ "step": 40
250
+ },
251
+ {
252
+ "epoch": 0.33,
253
+ "learning_rate": 0.00015729855062290022,
254
+ "loss": 1.1243,
255
+ "step": 41
256
+ },
257
+ {
258
+ "epoch": 0.34,
259
+ "learning_rate": 0.00015515159299842707,
260
+ "loss": 1.1036,
261
+ "step": 42
262
+ },
263
+ {
264
+ "epoch": 0.34,
265
+ "learning_rate": 0.00015296745937313987,
266
+ "loss": 1.1246,
267
+ "step": 43
268
+ },
269
+ {
270
+ "epoch": 0.35,
271
+ "learning_rate": 0.00015074762200466556,
272
+ "loss": 1.1044,
273
+ "step": 44
274
+ },
275
+ {
276
+ "epoch": 0.36,
277
+ "learning_rate": 0.00014849357721743168,
278
+ "loss": 1.1097,
279
+ "step": 45
280
+ },
281
+ {
282
+ "epoch": 0.37,
283
+ "learning_rate": 0.00014620684439403962,
284
+ "loss": 1.0692,
285
+ "step": 46
286
+ },
287
+ {
288
+ "epoch": 0.38,
289
+ "learning_rate": 0.0001438889649510956,
290
+ "loss": 1.2537,
291
+ "step": 47
292
+ },
293
+ {
294
+ "epoch": 0.38,
295
+ "learning_rate": 0.00014154150130018866,
296
+ "loss": 1.186,
297
+ "step": 48
298
+ },
299
+ {
300
+ "epoch": 0.39,
301
+ "learning_rate": 0.00013916603579471705,
302
+ "loss": 1.1326,
303
+ "step": 49
304
+ },
305
+ {
306
+ "epoch": 0.4,
307
+ "learning_rate": 0.000136764169663272,
308
+ "loss": 1.1722,
309
+ "step": 50
310
+ },
311
+ {
312
+ "epoch": 0.41,
313
+ "learning_rate": 0.00013433752193029886,
314
+ "loss": 1.1804,
315
+ "step": 51
316
+ },
317
+ {
318
+ "epoch": 0.42,
319
+ "learning_rate": 0.00013188772832476188,
320
+ "loss": 1.2119,
321
+ "step": 52
322
+ },
323
+ {
324
+ "epoch": 0.42,
325
+ "learning_rate": 0.00012941644017754964,
326
+ "loss": 1.1719,
327
+ "step": 53
328
+ },
329
+ {
330
+ "epoch": 0.43,
331
+ "learning_rate": 0.00012692532330836346,
332
+ "loss": 1.1734,
333
+ "step": 54
334
+ },
335
+ {
336
+ "epoch": 0.44,
337
+ "learning_rate": 0.00012441605690283915,
338
+ "loss": 1.0837,
339
+ "step": 55
340
+ },
341
+ {
342
+ "epoch": 0.45,
343
+ "learning_rate": 0.0001218903323806595,
344
+ "loss": 1.0661,
345
+ "step": 56
346
+ },
347
+ {
348
+ "epoch": 0.46,
349
+ "learning_rate": 0.00011934985225541998,
350
+ "loss": 1.1172,
351
+ "step": 57
352
+ },
353
+ {
354
+ "epoch": 0.46,
355
+ "learning_rate": 0.00011679632898701649,
356
+ "loss": 1.06,
357
+ "step": 58
358
+ },
359
+ {
360
+ "epoch": 0.47,
361
+ "learning_rate": 0.00011423148382732853,
362
+ "loss": 1.0949,
363
+ "step": 59
364
+ },
365
+ {
366
+ "epoch": 0.48,
367
+ "learning_rate": 0.00011165704565997593,
368
+ "loss": 1.1851,
369
+ "step": 60
370
+ },
371
+ {
372
+ "epoch": 0.49,
373
+ "learning_rate": 0.00010907474983493144,
374
+ "loss": 1.1116,
375
+ "step": 61
376
+ },
377
+ {
378
+ "epoch": 0.5,
379
+ "learning_rate": 0.0001064863369987743,
380
+ "loss": 1.1421,
381
+ "step": 62
382
+ },
383
+ {
384
+ "epoch": 0.5,
385
+ "learning_rate": 0.00010389355192137377,
386
+ "loss": 1.0834,
387
+ "step": 63
388
+ },
389
+ {
390
+ "epoch": 0.51,
391
+ "learning_rate": 0.0001012981423197931,
392
+ "loss": 1.0071,
393
+ "step": 64
394
+ },
395
+ {
396
+ "epoch": 0.52,
397
+ "learning_rate": 9.870185768020693e-05,
398
+ "loss": 1.1504,
399
+ "step": 65
400
+ },
401
+ {
402
+ "epoch": 0.53,
403
+ "learning_rate": 9.610644807862625e-05,
404
+ "loss": 1.2267,
405
+ "step": 66
406
+ },
407
+ {
408
+ "epoch": 0.54,
409
+ "learning_rate": 9.35136630012257e-05,
410
+ "loss": 1.0855,
411
+ "step": 67
412
+ },
413
+ {
414
+ "epoch": 0.54,
415
+ "learning_rate": 9.092525016506858e-05,
416
+ "loss": 1.1245,
417
+ "step": 68
418
+ },
419
+ {
420
+ "epoch": 0.55,
421
+ "learning_rate": 8.83429543400241e-05,
422
+ "loss": 1.0357,
423
+ "step": 69
424
+ },
425
+ {
426
+ "epoch": 0.56,
427
+ "learning_rate": 8.57685161726715e-05,
428
+ "loss": 1.1321,
429
+ "step": 70
430
+ },
431
+ {
432
+ "epoch": 0.57,
433
+ "learning_rate": 8.320367101298351e-05,
434
+ "loss": 1.1019,
435
+ "step": 71
436
+ },
437
+ {
438
+ "epoch": 0.58,
439
+ "learning_rate": 8.065014774458003e-05,
440
+ "loss": 1.0587,
441
+ "step": 72
442
+ },
443
+ {
444
+ "epoch": 0.58,
445
+ "learning_rate": 7.810966761934053e-05,
446
+ "loss": 1.0647,
447
+ "step": 73
448
+ },
449
+ {
450
+ "epoch": 0.59,
451
+ "learning_rate": 7.558394309716088e-05,
452
+ "loss": 1.2036,
453
+ "step": 74
454
+ },
455
+ {
456
+ "epoch": 0.6,
457
+ "learning_rate": 7.307467669163655e-05,
458
+ "loss": 1.1131,
459
+ "step": 75
460
+ },
461
+ {
462
+ "epoch": 0.61,
463
+ "learning_rate": 7.058355982245037e-05,
464
+ "loss": 1.1824,
465
+ "step": 76
466
+ },
467
+ {
468
+ "epoch": 0.62,
469
+ "learning_rate": 6.811227167523815e-05,
470
+ "loss": 1.1728,
471
+ "step": 77
472
+ },
473
+ {
474
+ "epoch": 0.62,
475
+ "learning_rate": 6.566247806970119e-05,
476
+ "loss": 1.0121,
477
+ "step": 78
478
+ },
479
+ {
480
+ "epoch": 0.63,
481
+ "learning_rate": 6.323583033672799e-05,
482
+ "loss": 1.0867,
483
+ "step": 79
484
+ },
485
+ {
486
+ "epoch": 0.64,
487
+ "learning_rate": 6.083396420528298e-05,
488
+ "loss": 1.1176,
489
+ "step": 80
490
+ },
491
+ {
492
+ "epoch": 0.65,
493
+ "learning_rate": 5.845849869981137e-05,
494
+ "loss": 1.0947,
495
+ "step": 81
496
+ },
497
+ {
498
+ "epoch": 0.66,
499
+ "learning_rate": 5.611103504890444e-05,
500
+ "loss": 0.9342,
501
+ "step": 82
502
+ },
503
+ {
504
+ "epoch": 0.66,
505
+ "learning_rate": 5.379315560596038e-05,
506
+ "loss": 1.2286,
507
+ "step": 83
508
+ },
509
+ {
510
+ "epoch": 0.67,
511
+ "learning_rate": 5.1506422782568345e-05,
512
+ "loss": 0.9824,
513
+ "step": 84
514
+ },
515
+ {
516
+ "epoch": 0.68,
517
+ "learning_rate": 4.9252377995334444e-05,
518
+ "loss": 1.131,
519
+ "step": 85
520
+ },
521
+ {
522
+ "epoch": 0.69,
523
+ "learning_rate": 4.703254062686017e-05,
524
+ "loss": 1.0423,
525
+ "step": 86
526
+ },
527
+ {
528
+ "epoch": 0.7,
529
+ "learning_rate": 4.484840700157295e-05,
530
+ "loss": 1.0741,
531
+ "step": 87
532
+ },
533
+ {
534
+ "epoch": 0.7,
535
+ "learning_rate": 4.270144937709981e-05,
536
+ "loss": 1.1291,
537
+ "step": 88
538
+ },
539
+ {
540
+ "epoch": 0.71,
541
+ "learning_rate": 4.059311495186338e-05,
542
+ "loss": 1.1319,
543
+ "step": 89
544
+ },
545
+ {
546
+ "epoch": 0.72,
547
+ "learning_rate": 3.852482488956992e-05,
548
+ "loss": 1.1128,
549
+ "step": 90
550
+ },
551
+ {
552
+ "epoch": 0.73,
553
+ "learning_rate": 3.649797336124615e-05,
554
+ "loss": 1.0266,
555
+ "step": 91
556
+ },
557
+ {
558
+ "epoch": 0.74,
559
+ "learning_rate": 3.45139266054715e-05,
560
+ "loss": 1.1304,
561
+ "step": 92
562
+ },
563
+ {
564
+ "epoch": 0.74,
565
+ "learning_rate": 3.257402200743821e-05,
566
+ "loss": 1.1331,
567
+ "step": 93
568
+ },
569
+ {
570
+ "epoch": 0.75,
571
+ "learning_rate": 3.0679567197461134e-05,
572
+ "loss": 1.0754,
573
+ "step": 94
574
+ },
575
+ {
576
+ "epoch": 0.76,
577
+ "learning_rate": 2.8831839169543996e-05,
578
+ "loss": 1.014,
579
+ "step": 95
580
+ },
581
+ {
582
+ "epoch": 0.77,
583
+ "learning_rate": 2.7032083420597e-05,
584
+ "loss": 1.0225,
585
+ "step": 96
586
+ },
587
+ {
588
+ "epoch": 0.78,
589
+ "learning_rate": 2.528151311088537e-05,
590
+ "loss": 1.1003,
591
+ "step": 97
592
+ },
593
+ {
594
+ "epoch": 0.78,
595
+ "learning_rate": 2.3581308246275103e-05,
596
+ "loss": 0.9896,
597
+ "step": 98
598
+ },
599
+ {
600
+ "epoch": 0.79,
601
+ "learning_rate": 2.1932614882827197e-05,
602
+ "loss": 1.1506,
603
+ "step": 99
604
+ },
605
+ {
606
+ "epoch": 0.8,
607
+ "learning_rate": 2.03365443542764e-05,
608
+ "loss": 1.1285,
609
+ "step": 100
610
+ },
611
+ {
612
+ "epoch": 0.81,
613
+ "learning_rate": 1.879417252291502e-05,
614
+ "loss": 1.1775,
615
+ "step": 101
616
+ },
617
+ {
618
+ "epoch": 0.82,
619
+ "learning_rate": 1.730653905438714e-05,
620
+ "loss": 1.1682,
621
+ "step": 102
622
+ },
623
+ {
624
+ "epoch": 0.82,
625
+ "learning_rate": 1.587464671688187e-05,
626
+ "loss": 1.0003,
627
+ "step": 103
628
+ },
629
+ {
630
+ "epoch": 0.83,
631
+ "learning_rate": 1.4499460705197998e-05,
632
+ "loss": 1.186,
633
+ "step": 104
634
+ },
635
+ {
636
+ "epoch": 0.84,
637
+ "learning_rate": 1.3181907990135622e-05,
638
+ "loss": 1.1757,
639
+ "step": 105
640
+ },
641
+ {
642
+ "epoch": 0.85,
643
+ "learning_rate": 1.1922876693653585e-05,
644
+ "loss": 1.0854,
645
+ "step": 106
646
+ },
647
+ {
648
+ "epoch": 0.86,
649
+ "learning_rate": 1.0723215490213634e-05,
650
+ "loss": 1.0466,
651
+ "step": 107
652
+ },
653
+ {
654
+ "epoch": 0.86,
655
+ "learning_rate": 9.583733034714981e-06,
656
+ "loss": 1.0728,
657
+ "step": 108
658
+ },
659
+ {
660
+ "epoch": 0.87,
661
+ "learning_rate": 8.505197417404687e-06,
662
+ "loss": 1.2036,
663
+ "step": 109
664
+ },
665
+ {
666
+ "epoch": 0.88,
667
+ "learning_rate": 7.488335646131628e-06,
668
+ "loss": 1.0486,
669
+ "step": 110
670
+ },
671
+ {
672
+ "epoch": 0.89,
673
+ "learning_rate": 6.533833156292679e-06,
674
+ "loss": 1.0899,
675
+ "step": 111
676
+ },
677
+ {
678
+ "epoch": 0.9,
679
+ "learning_rate": 5.6423333488018095e-06,
680
+ "loss": 1.0798,
681
+ "step": 112
682
+ },
683
+ {
684
+ "epoch": 0.9,
685
+ "learning_rate": 4.8144371563930476e-06,
686
+ "loss": 1.0657,
687
+ "step": 113
688
+ },
689
+ {
690
+ "epoch": 0.91,
691
+ "learning_rate": 4.050702638550275e-06,
692
+ "loss": 1.0662,
693
+ "step": 114
694
+ },
695
+ {
696
+ "epoch": 0.92,
697
+ "learning_rate": 3.3516446053363015e-06,
698
+ "loss": 1.0746,
699
+ "step": 115
700
+ },
701
+ {
702
+ "epoch": 0.93,
703
+ "learning_rate": 2.717734270375272e-06,
704
+ "loss": 1.1993,
705
+ "step": 116
706
+ },
707
+ {
708
+ "epoch": 0.94,
709
+ "learning_rate": 2.1493989332218468e-06,
710
+ "loss": 1.0301,
711
+ "step": 117
712
+ },
713
+ {
714
+ "epoch": 0.94,
715
+ "learning_rate": 1.6470216913317626e-06,
716
+ "loss": 1.0373,
717
+ "step": 118
718
+ },
719
+ {
720
+ "epoch": 0.95,
721
+ "learning_rate": 1.2109411818274852e-06,
722
+ "loss": 0.9994,
723
+ "step": 119
724
+ },
725
+ {
726
+ "epoch": 0.96,
727
+ "learning_rate": 8.41451353233369e-07,
728
+ "loss": 1.044,
729
+ "step": 120
730
+ },
731
+ {
732
+ "epoch": 0.97,
733
+ "learning_rate": 5.388012673338661e-07,
734
+ "loss": 0.9979,
735
+ "step": 121
736
+ },
737
+ {
738
+ "epoch": 0.98,
739
+ "learning_rate": 3.0319493128866396e-07,
740
+ "loss": 1.1068,
741
+ "step": 122
742
+ },
743
+ {
744
+ "epoch": 0.98,
745
+ "learning_rate": 1.3479116011769767e-07,
746
+ "loss": 1.0095,
747
+ "step": 123
748
+ },
749
+ {
750
+ "epoch": 0.99,
751
+ "learning_rate": 3.370346964876036e-08,
752
+ "loss": 1.0024,
753
+ "step": 124
754
+ },
755
+ {
756
+ "epoch": 1.0,
757
+ "learning_rate": 0.0,
758
+ "loss": 1.0189,
759
+ "step": 125
760
+ },
761
+ {
762
+ "epoch": 1.0,
763
+ "step": 125,
764
+ "total_flos": 1901956792320.0,
765
+ "train_loss": 1.1506947622299195,
766
+ "train_runtime": 1117.0219,
767
+ "train_samples_per_second": 1.79,
768
+ "train_steps_per_second": 0.112
769
+ }
770
+ ],
771
+ "logging_steps": 1.0,
772
+ "max_steps": 125,
773
+ "num_input_tokens_seen": 0,
774
+ "num_train_epochs": 1,
775
+ "save_steps": 50000,
776
+ "total_flos": 1901956792320.0,
777
+ "train_batch_size": 16,
778
+ "trial_name": null,
779
+ "trial_params": null
780
+ }