ClarenceTKX commited on
Commit
3900a6f
·
1 Parent(s): f66639a

Upload 6 files

Browse files
Files changed (3) hide show
  1. rng_state.pth +3 -0
  2. scheduler.pt +3 -0
  3. trainer_state.json +919 -0
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3a43ed107614bafcd0a0e4467c9e193c84619e8024d61b542f61417217b9177
3
+ size 14308
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e7dbaf01f8e731d2a9a0777e4e9d5d8a65040da4670a22f6e2f536c27d5baba
3
+ size 1064
trainer_state.json ADDED
@@ -0,0 +1,919 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.4926270156655391,
5
+ "eval_steps": 99999,
6
+ "global_step": 15000,
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": 1.9800000000000004e-05,
14
+ "loss": 3.0349,
15
+ "step": 100
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 3.94e-05,
20
+ "loss": 3.2422,
21
+ "step": 200
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 5.92e-05,
26
+ "loss": 3.2539,
27
+ "step": 300
28
+ },
29
+ {
30
+ "epoch": 0.01,
31
+ "learning_rate": 7.920000000000001e-05,
32
+ "loss": 3.111,
33
+ "step": 400
34
+ },
35
+ {
36
+ "epoch": 0.02,
37
+ "learning_rate": 9.92e-05,
38
+ "loss": 3.0833,
39
+ "step": 500
40
+ },
41
+ {
42
+ "epoch": 0.02,
43
+ "learning_rate": 9.967945507362518e-05,
44
+ "loss": 3.0845,
45
+ "step": 600
46
+ },
47
+ {
48
+ "epoch": 0.02,
49
+ "learning_rate": 9.934555410865138e-05,
50
+ "loss": 3.0598,
51
+ "step": 700
52
+ },
53
+ {
54
+ "epoch": 0.03,
55
+ "learning_rate": 9.901165314367759e-05,
56
+ "loss": 3.0556,
57
+ "step": 800
58
+ },
59
+ {
60
+ "epoch": 0.03,
61
+ "learning_rate": 9.86777521787038e-05,
62
+ "loss": 3.0666,
63
+ "step": 900
64
+ },
65
+ {
66
+ "epoch": 0.03,
67
+ "learning_rate": 9.834385121373001e-05,
68
+ "loss": 3.0681,
69
+ "step": 1000
70
+ },
71
+ {
72
+ "epoch": 0.04,
73
+ "learning_rate": 9.800995024875622e-05,
74
+ "loss": 3.0349,
75
+ "step": 1100
76
+ },
77
+ {
78
+ "epoch": 0.04,
79
+ "learning_rate": 9.767604928378244e-05,
80
+ "loss": 3.0354,
81
+ "step": 1200
82
+ },
83
+ {
84
+ "epoch": 0.04,
85
+ "learning_rate": 9.734214831880864e-05,
86
+ "loss": 3.1831,
87
+ "step": 1300
88
+ },
89
+ {
90
+ "epoch": 0.05,
91
+ "learning_rate": 9.700824735383486e-05,
92
+ "loss": 3.0632,
93
+ "step": 1400
94
+ },
95
+ {
96
+ "epoch": 0.05,
97
+ "learning_rate": 9.667434638886107e-05,
98
+ "loss": 3.1388,
99
+ "step": 1500
100
+ },
101
+ {
102
+ "epoch": 0.05,
103
+ "learning_rate": 9.634044542388728e-05,
104
+ "loss": 3.0572,
105
+ "step": 1600
106
+ },
107
+ {
108
+ "epoch": 0.06,
109
+ "learning_rate": 9.600654445891349e-05,
110
+ "loss": 3.0525,
111
+ "step": 1700
112
+ },
113
+ {
114
+ "epoch": 0.06,
115
+ "learning_rate": 9.567264349393971e-05,
116
+ "loss": 3.4111,
117
+ "step": 1800
118
+ },
119
+ {
120
+ "epoch": 0.06,
121
+ "learning_rate": 9.533874252896591e-05,
122
+ "loss": 3.1258,
123
+ "step": 1900
124
+ },
125
+ {
126
+ "epoch": 0.07,
127
+ "learning_rate": 9.500484156399212e-05,
128
+ "loss": 3.2196,
129
+ "step": 2000
130
+ },
131
+ {
132
+ "epoch": 0.07,
133
+ "learning_rate": 9.467094059901834e-05,
134
+ "loss": 3.0257,
135
+ "step": 2100
136
+ },
137
+ {
138
+ "epoch": 0.07,
139
+ "learning_rate": 9.433703963404454e-05,
140
+ "loss": 3.0797,
141
+ "step": 2200
142
+ },
143
+ {
144
+ "epoch": 0.08,
145
+ "learning_rate": 9.400313866907076e-05,
146
+ "loss": 3.4158,
147
+ "step": 2300
148
+ },
149
+ {
150
+ "epoch": 0.08,
151
+ "learning_rate": 9.366923770409697e-05,
152
+ "loss": 3.0948,
153
+ "step": 2400
154
+ },
155
+ {
156
+ "epoch": 0.08,
157
+ "learning_rate": 9.333533673912319e-05,
158
+ "loss": 3.0128,
159
+ "step": 2500
160
+ },
161
+ {
162
+ "epoch": 0.09,
163
+ "learning_rate": 9.300143577414939e-05,
164
+ "loss": 3.0169,
165
+ "step": 2600
166
+ },
167
+ {
168
+ "epoch": 0.09,
169
+ "learning_rate": 9.266753480917561e-05,
170
+ "loss": 3.0737,
171
+ "step": 2700
172
+ },
173
+ {
174
+ "epoch": 0.09,
175
+ "learning_rate": 9.233363384420181e-05,
176
+ "loss": 3.0862,
177
+ "step": 2800
178
+ },
179
+ {
180
+ "epoch": 0.1,
181
+ "learning_rate": 9.199973287922803e-05,
182
+ "loss": 3.0661,
183
+ "step": 2900
184
+ },
185
+ {
186
+ "epoch": 0.1,
187
+ "learning_rate": 9.166583191425424e-05,
188
+ "loss": 3.0505,
189
+ "step": 3000
190
+ },
191
+ {
192
+ "epoch": 0.1,
193
+ "learning_rate": 9.133193094928046e-05,
194
+ "loss": 3.0437,
195
+ "step": 3100
196
+ },
197
+ {
198
+ "epoch": 0.11,
199
+ "learning_rate": 9.099802998430665e-05,
200
+ "loss": 3.0424,
201
+ "step": 3200
202
+ },
203
+ {
204
+ "epoch": 0.11,
205
+ "learning_rate": 9.066412901933287e-05,
206
+ "loss": 3.0934,
207
+ "step": 3300
208
+ },
209
+ {
210
+ "epoch": 0.11,
211
+ "learning_rate": 9.033356706400882e-05,
212
+ "loss": 3.2011,
213
+ "step": 3400
214
+ },
215
+ {
216
+ "epoch": 0.11,
217
+ "learning_rate": 8.999966609903502e-05,
218
+ "loss": 3.1089,
219
+ "step": 3500
220
+ },
221
+ {
222
+ "epoch": 0.12,
223
+ "learning_rate": 8.966576513406124e-05,
224
+ "loss": 3.0698,
225
+ "step": 3600
226
+ },
227
+ {
228
+ "epoch": 0.12,
229
+ "learning_rate": 8.933186416908745e-05,
230
+ "loss": 3.0202,
231
+ "step": 3700
232
+ },
233
+ {
234
+ "epoch": 0.12,
235
+ "learning_rate": 8.899796320411367e-05,
236
+ "loss": 3.0918,
237
+ "step": 3800
238
+ },
239
+ {
240
+ "epoch": 0.13,
241
+ "learning_rate": 8.866406223913988e-05,
242
+ "loss": 3.1013,
243
+ "step": 3900
244
+ },
245
+ {
246
+ "epoch": 0.13,
247
+ "learning_rate": 8.833016127416609e-05,
248
+ "loss": 3.3244,
249
+ "step": 4000
250
+ },
251
+ {
252
+ "epoch": 0.13,
253
+ "learning_rate": 8.799626030919231e-05,
254
+ "loss": 3.0985,
255
+ "step": 4100
256
+ },
257
+ {
258
+ "epoch": 0.14,
259
+ "learning_rate": 8.766235934421851e-05,
260
+ "loss": 3.0286,
261
+ "step": 4200
262
+ },
263
+ {
264
+ "epoch": 0.14,
265
+ "learning_rate": 8.732845837924472e-05,
266
+ "loss": 3.0809,
267
+ "step": 4300
268
+ },
269
+ {
270
+ "epoch": 0.14,
271
+ "learning_rate": 8.699455741427092e-05,
272
+ "loss": 3.0599,
273
+ "step": 4400
274
+ },
275
+ {
276
+ "epoch": 0.15,
277
+ "learning_rate": 8.666065644929714e-05,
278
+ "loss": 3.0703,
279
+ "step": 4500
280
+ },
281
+ {
282
+ "epoch": 0.15,
283
+ "learning_rate": 8.632675548432335e-05,
284
+ "loss": 3.0816,
285
+ "step": 4600
286
+ },
287
+ {
288
+ "epoch": 0.15,
289
+ "learning_rate": 8.599285451934957e-05,
290
+ "loss": 3.0659,
291
+ "step": 4700
292
+ },
293
+ {
294
+ "epoch": 0.16,
295
+ "learning_rate": 8.565895355437577e-05,
296
+ "loss": 3.0197,
297
+ "step": 4800
298
+ },
299
+ {
300
+ "epoch": 0.16,
301
+ "learning_rate": 8.532505258940199e-05,
302
+ "loss": 3.0335,
303
+ "step": 4900
304
+ },
305
+ {
306
+ "epoch": 0.16,
307
+ "learning_rate": 8.49911516244282e-05,
308
+ "loss": 3.1091,
309
+ "step": 5000
310
+ },
311
+ {
312
+ "epoch": 0.17,
313
+ "learning_rate": 8.465725065945441e-05,
314
+ "loss": 3.0402,
315
+ "step": 5100
316
+ },
317
+ {
318
+ "epoch": 0.17,
319
+ "learning_rate": 8.432334969448062e-05,
320
+ "loss": 3.0689,
321
+ "step": 5200
322
+ },
323
+ {
324
+ "epoch": 0.17,
325
+ "learning_rate": 8.398944872950684e-05,
326
+ "loss": 3.0209,
327
+ "step": 5300
328
+ },
329
+ {
330
+ "epoch": 0.18,
331
+ "learning_rate": 8.365554776453304e-05,
332
+ "loss": 3.0484,
333
+ "step": 5400
334
+ },
335
+ {
336
+ "epoch": 0.18,
337
+ "learning_rate": 8.332164679955925e-05,
338
+ "loss": 3.0419,
339
+ "step": 5500
340
+ },
341
+ {
342
+ "epoch": 0.18,
343
+ "learning_rate": 8.29910848442352e-05,
344
+ "loss": 3.1417,
345
+ "step": 5600
346
+ },
347
+ {
348
+ "epoch": 0.19,
349
+ "learning_rate": 8.265718387926142e-05,
350
+ "loss": 3.03,
351
+ "step": 5700
352
+ },
353
+ {
354
+ "epoch": 0.19,
355
+ "learning_rate": 8.232328291428762e-05,
356
+ "loss": 3.0755,
357
+ "step": 5800
358
+ },
359
+ {
360
+ "epoch": 0.19,
361
+ "learning_rate": 8.198938194931384e-05,
362
+ "loss": 3.0759,
363
+ "step": 5900
364
+ },
365
+ {
366
+ "epoch": 0.2,
367
+ "learning_rate": 8.165548098434005e-05,
368
+ "loss": 3.0822,
369
+ "step": 6000
370
+ },
371
+ {
372
+ "epoch": 0.2,
373
+ "learning_rate": 8.132158001936627e-05,
374
+ "loss": 3.3225,
375
+ "step": 6100
376
+ },
377
+ {
378
+ "epoch": 0.2,
379
+ "learning_rate": 8.098767905439247e-05,
380
+ "loss": 4.2645,
381
+ "step": 6200
382
+ },
383
+ {
384
+ "epoch": 0.21,
385
+ "learning_rate": 8.065377808941869e-05,
386
+ "loss": 3.3215,
387
+ "step": 6300
388
+ },
389
+ {
390
+ "epoch": 0.21,
391
+ "learning_rate": 8.03198771244449e-05,
392
+ "loss": 3.3495,
393
+ "step": 6400
394
+ },
395
+ {
396
+ "epoch": 0.21,
397
+ "learning_rate": 7.998597615947111e-05,
398
+ "loss": 3.9503,
399
+ "step": 6500
400
+ },
401
+ {
402
+ "epoch": 0.22,
403
+ "learning_rate": 7.96520751944973e-05,
404
+ "loss": 3.2799,
405
+ "step": 6600
406
+ },
407
+ {
408
+ "epoch": 0.22,
409
+ "learning_rate": 7.931817422952352e-05,
410
+ "loss": 3.3251,
411
+ "step": 6700
412
+ },
413
+ {
414
+ "epoch": 0.22,
415
+ "learning_rate": 7.898427326454973e-05,
416
+ "loss": 3.3824,
417
+ "step": 6800
418
+ },
419
+ {
420
+ "epoch": 0.23,
421
+ "learning_rate": 7.865037229957595e-05,
422
+ "loss": 3.2103,
423
+ "step": 6900
424
+ },
425
+ {
426
+ "epoch": 0.23,
427
+ "learning_rate": 7.831647133460215e-05,
428
+ "loss": 3.4477,
429
+ "step": 7000
430
+ },
431
+ {
432
+ "epoch": 0.23,
433
+ "learning_rate": 7.798257036962837e-05,
434
+ "loss": 3.3625,
435
+ "step": 7100
436
+ },
437
+ {
438
+ "epoch": 0.24,
439
+ "learning_rate": 7.764866940465458e-05,
440
+ "loss": 3.4537,
441
+ "step": 7200
442
+ },
443
+ {
444
+ "epoch": 0.24,
445
+ "learning_rate": 7.73147684396808e-05,
446
+ "loss": 3.5719,
447
+ "step": 7300
448
+ },
449
+ {
450
+ "epoch": 0.24,
451
+ "learning_rate": 7.698086747470701e-05,
452
+ "loss": 3.1677,
453
+ "step": 7400
454
+ },
455
+ {
456
+ "epoch": 0.25,
457
+ "learning_rate": 7.664696650973322e-05,
458
+ "loss": 3.3993,
459
+ "step": 7500
460
+ },
461
+ {
462
+ "epoch": 0.25,
463
+ "learning_rate": 7.631306554475944e-05,
464
+ "loss": 3.1136,
465
+ "step": 7600
466
+ },
467
+ {
468
+ "epoch": 0.25,
469
+ "learning_rate": 7.597916457978564e-05,
470
+ "loss": 3.1078,
471
+ "step": 7700
472
+ },
473
+ {
474
+ "epoch": 0.26,
475
+ "learning_rate": 7.564526361481185e-05,
476
+ "loss": 3.0449,
477
+ "step": 7800
478
+ },
479
+ {
480
+ "epoch": 0.26,
481
+ "learning_rate": 7.53147016594878e-05,
482
+ "loss": 3.1574,
483
+ "step": 7900
484
+ },
485
+ {
486
+ "epoch": 0.26,
487
+ "learning_rate": 7.4980800694514e-05,
488
+ "loss": 3.2764,
489
+ "step": 8000
490
+ },
491
+ {
492
+ "epoch": 0.27,
493
+ "learning_rate": 7.464689972954022e-05,
494
+ "loss": 3.0553,
495
+ "step": 8100
496
+ },
497
+ {
498
+ "epoch": 0.27,
499
+ "learning_rate": 7.431299876456643e-05,
500
+ "loss": 3.1408,
501
+ "step": 8200
502
+ },
503
+ {
504
+ "epoch": 0.27,
505
+ "learning_rate": 7.397909779959265e-05,
506
+ "loss": 3.1765,
507
+ "step": 8300
508
+ },
509
+ {
510
+ "epoch": 0.28,
511
+ "learning_rate": 7.364519683461885e-05,
512
+ "loss": 3.0974,
513
+ "step": 8400
514
+ },
515
+ {
516
+ "epoch": 0.28,
517
+ "learning_rate": 7.331129586964507e-05,
518
+ "loss": 3.0998,
519
+ "step": 8500
520
+ },
521
+ {
522
+ "epoch": 0.28,
523
+ "learning_rate": 7.297739490467128e-05,
524
+ "loss": 3.1488,
525
+ "step": 8600
526
+ },
527
+ {
528
+ "epoch": 0.29,
529
+ "learning_rate": 7.26434939396975e-05,
530
+ "loss": 3.1722,
531
+ "step": 8700
532
+ },
533
+ {
534
+ "epoch": 0.29,
535
+ "learning_rate": 7.23095929747237e-05,
536
+ "loss": 3.1605,
537
+ "step": 8800
538
+ },
539
+ {
540
+ "epoch": 0.29,
541
+ "learning_rate": 7.19756920097499e-05,
542
+ "loss": 3.0762,
543
+ "step": 8900
544
+ },
545
+ {
546
+ "epoch": 0.3,
547
+ "learning_rate": 7.164179104477612e-05,
548
+ "loss": 3.1019,
549
+ "step": 9000
550
+ },
551
+ {
552
+ "epoch": 0.3,
553
+ "learning_rate": 7.130789007980233e-05,
554
+ "loss": 3.0633,
555
+ "step": 9100
556
+ },
557
+ {
558
+ "epoch": 0.3,
559
+ "learning_rate": 7.097398911482855e-05,
560
+ "loss": 3.1291,
561
+ "step": 9200
562
+ },
563
+ {
564
+ "epoch": 0.31,
565
+ "learning_rate": 7.064008814985475e-05,
566
+ "loss": 3.1701,
567
+ "step": 9300
568
+ },
569
+ {
570
+ "epoch": 0.31,
571
+ "learning_rate": 7.030618718488097e-05,
572
+ "loss": 3.1169,
573
+ "step": 9400
574
+ },
575
+ {
576
+ "epoch": 0.31,
577
+ "learning_rate": 6.997228621990718e-05,
578
+ "loss": 3.3379,
579
+ "step": 9500
580
+ },
581
+ {
582
+ "epoch": 0.32,
583
+ "learning_rate": 6.96383852549334e-05,
584
+ "loss": 3.0393,
585
+ "step": 9600
586
+ },
587
+ {
588
+ "epoch": 0.32,
589
+ "learning_rate": 6.93044842899596e-05,
590
+ "loss": 3.1069,
591
+ "step": 9700
592
+ },
593
+ {
594
+ "epoch": 0.32,
595
+ "learning_rate": 6.897058332498582e-05,
596
+ "loss": 3.1976,
597
+ "step": 9800
598
+ },
599
+ {
600
+ "epoch": 0.33,
601
+ "learning_rate": 6.863668236001203e-05,
602
+ "loss": 3.0431,
603
+ "step": 9900
604
+ },
605
+ {
606
+ "epoch": 0.33,
607
+ "learning_rate": 6.830278139503824e-05,
608
+ "loss": 3.063,
609
+ "step": 10000
610
+ },
611
+ {
612
+ "epoch": 0.33,
613
+ "learning_rate": 6.796888043006444e-05,
614
+ "loss": 3.0741,
615
+ "step": 10100
616
+ },
617
+ {
618
+ "epoch": 0.33,
619
+ "learning_rate": 6.763497946509065e-05,
620
+ "loss": 3.1159,
621
+ "step": 10200
622
+ },
623
+ {
624
+ "epoch": 0.34,
625
+ "learning_rate": 6.730107850011686e-05,
626
+ "loss": 3.1405,
627
+ "step": 10300
628
+ },
629
+ {
630
+ "epoch": 0.34,
631
+ "learning_rate": 6.697051654479281e-05,
632
+ "loss": 3.45,
633
+ "step": 10400
634
+ },
635
+ {
636
+ "epoch": 0.34,
637
+ "learning_rate": 6.663661557981903e-05,
638
+ "loss": 3.0671,
639
+ "step": 10500
640
+ },
641
+ {
642
+ "epoch": 0.35,
643
+ "learning_rate": 6.630271461484523e-05,
644
+ "loss": 3.1104,
645
+ "step": 10600
646
+ },
647
+ {
648
+ "epoch": 0.35,
649
+ "learning_rate": 6.596881364987145e-05,
650
+ "loss": 3.262,
651
+ "step": 10700
652
+ },
653
+ {
654
+ "epoch": 0.35,
655
+ "learning_rate": 6.563491268489766e-05,
656
+ "loss": 3.3126,
657
+ "step": 10800
658
+ },
659
+ {
660
+ "epoch": 0.36,
661
+ "learning_rate": 6.530101171992388e-05,
662
+ "loss": 3.136,
663
+ "step": 10900
664
+ },
665
+ {
666
+ "epoch": 0.36,
667
+ "learning_rate": 6.49671107549501e-05,
668
+ "loss": 3.0665,
669
+ "step": 11000
670
+ },
671
+ {
672
+ "epoch": 0.36,
673
+ "learning_rate": 6.46332097899763e-05,
674
+ "loss": 3.1046,
675
+ "step": 11100
676
+ },
677
+ {
678
+ "epoch": 0.37,
679
+ "learning_rate": 6.42993088250025e-05,
680
+ "loss": 3.072,
681
+ "step": 11200
682
+ },
683
+ {
684
+ "epoch": 0.37,
685
+ "learning_rate": 6.396540786002871e-05,
686
+ "loss": 3.1027,
687
+ "step": 11300
688
+ },
689
+ {
690
+ "epoch": 0.37,
691
+ "learning_rate": 6.363150689505493e-05,
692
+ "loss": 3.1536,
693
+ "step": 11400
694
+ },
695
+ {
696
+ "epoch": 0.38,
697
+ "learning_rate": 6.329760593008114e-05,
698
+ "loss": 3.0932,
699
+ "step": 11500
700
+ },
701
+ {
702
+ "epoch": 0.38,
703
+ "learning_rate": 6.296370496510735e-05,
704
+ "loss": 3.0798,
705
+ "step": 11600
706
+ },
707
+ {
708
+ "epoch": 0.38,
709
+ "learning_rate": 6.262980400013356e-05,
710
+ "loss": 3.1229,
711
+ "step": 11700
712
+ },
713
+ {
714
+ "epoch": 0.39,
715
+ "learning_rate": 6.229590303515978e-05,
716
+ "loss": 3.1848,
717
+ "step": 11800
718
+ },
719
+ {
720
+ "epoch": 0.39,
721
+ "learning_rate": 6.196200207018598e-05,
722
+ "loss": 3.2213,
723
+ "step": 11900
724
+ },
725
+ {
726
+ "epoch": 0.39,
727
+ "learning_rate": 6.16281011052122e-05,
728
+ "loss": 3.3552,
729
+ "step": 12000
730
+ },
731
+ {
732
+ "epoch": 0.4,
733
+ "learning_rate": 6.129420014023841e-05,
734
+ "loss": 3.2678,
735
+ "step": 12100
736
+ },
737
+ {
738
+ "epoch": 0.4,
739
+ "learning_rate": 6.0960299175264626e-05,
740
+ "loss": 3.2389,
741
+ "step": 12200
742
+ },
743
+ {
744
+ "epoch": 0.4,
745
+ "learning_rate": 6.062639821029084e-05,
746
+ "loss": 3.3173,
747
+ "step": 12300
748
+ },
749
+ {
750
+ "epoch": 0.41,
751
+ "learning_rate": 6.0292497245317036e-05,
752
+ "loss": 3.0254,
753
+ "step": 12400
754
+ },
755
+ {
756
+ "epoch": 0.41,
757
+ "learning_rate": 5.995859628034325e-05,
758
+ "loss": 3.0548,
759
+ "step": 12500
760
+ },
761
+ {
762
+ "epoch": 0.41,
763
+ "learning_rate": 5.962469531536946e-05,
764
+ "loss": 3.0759,
765
+ "step": 12600
766
+ },
767
+ {
768
+ "epoch": 0.42,
769
+ "learning_rate": 5.929079435039567e-05,
770
+ "loss": 3.0964,
771
+ "step": 12700
772
+ },
773
+ {
774
+ "epoch": 0.42,
775
+ "learning_rate": 5.8956893385421884e-05,
776
+ "loss": 3.2731,
777
+ "step": 12800
778
+ },
779
+ {
780
+ "epoch": 0.42,
781
+ "learning_rate": 5.8622992420448096e-05,
782
+ "loss": 3.0974,
783
+ "step": 12900
784
+ },
785
+ {
786
+ "epoch": 0.43,
787
+ "learning_rate": 5.828909145547431e-05,
788
+ "loss": 3.2184,
789
+ "step": 13000
790
+ },
791
+ {
792
+ "epoch": 0.43,
793
+ "learning_rate": 5.795519049050052e-05,
794
+ "loss": 3.1414,
795
+ "step": 13100
796
+ },
797
+ {
798
+ "epoch": 0.43,
799
+ "learning_rate": 5.762128952552673e-05,
800
+ "loss": 3.3148,
801
+ "step": 13200
802
+ },
803
+ {
804
+ "epoch": 0.44,
805
+ "learning_rate": 5.7287388560552944e-05,
806
+ "loss": 3.2656,
807
+ "step": 13300
808
+ },
809
+ {
810
+ "epoch": 0.44,
811
+ "learning_rate": 5.6953487595579156e-05,
812
+ "loss": 3.0808,
813
+ "step": 13400
814
+ },
815
+ {
816
+ "epoch": 0.44,
817
+ "learning_rate": 5.661958663060537e-05,
818
+ "loss": 3.1319,
819
+ "step": 13500
820
+ },
821
+ {
822
+ "epoch": 0.45,
823
+ "learning_rate": 5.628568566563157e-05,
824
+ "loss": 3.0652,
825
+ "step": 13600
826
+ },
827
+ {
828
+ "epoch": 0.45,
829
+ "learning_rate": 5.5951784700657785e-05,
830
+ "loss": 3.2544,
831
+ "step": 13700
832
+ },
833
+ {
834
+ "epoch": 0.45,
835
+ "learning_rate": 5.5617883735684e-05,
836
+ "loss": 3.1063,
837
+ "step": 13800
838
+ },
839
+ {
840
+ "epoch": 0.46,
841
+ "learning_rate": 5.528398277071021e-05,
842
+ "loss": 3.146,
843
+ "step": 13900
844
+ },
845
+ {
846
+ "epoch": 0.46,
847
+ "learning_rate": 5.495008180573642e-05,
848
+ "loss": 3.0885,
849
+ "step": 14000
850
+ },
851
+ {
852
+ "epoch": 0.46,
853
+ "learning_rate": 5.461618084076263e-05,
854
+ "loss": 3.0842,
855
+ "step": 14100
856
+ },
857
+ {
858
+ "epoch": 0.47,
859
+ "learning_rate": 5.4282279875788845e-05,
860
+ "loss": 3.2433,
861
+ "step": 14200
862
+ },
863
+ {
864
+ "epoch": 0.47,
865
+ "learning_rate": 5.3948378910815057e-05,
866
+ "loss": 3.2303,
867
+ "step": 14300
868
+ },
869
+ {
870
+ "epoch": 0.47,
871
+ "learning_rate": 5.361447794584127e-05,
872
+ "loss": 3.0701,
873
+ "step": 14400
874
+ },
875
+ {
876
+ "epoch": 0.48,
877
+ "learning_rate": 5.328391599051722e-05,
878
+ "loss": 3.303,
879
+ "step": 14500
880
+ },
881
+ {
882
+ "epoch": 0.48,
883
+ "learning_rate": 5.295001502554343e-05,
884
+ "loss": 3.1973,
885
+ "step": 14600
886
+ },
887
+ {
888
+ "epoch": 0.48,
889
+ "learning_rate": 5.261611406056963e-05,
890
+ "loss": 3.1066,
891
+ "step": 14700
892
+ },
893
+ {
894
+ "epoch": 0.49,
895
+ "learning_rate": 5.228221309559584e-05,
896
+ "loss": 3.2033,
897
+ "step": 14800
898
+ },
899
+ {
900
+ "epoch": 0.49,
901
+ "learning_rate": 5.195165114027179e-05,
902
+ "loss": 3.2673,
903
+ "step": 14900
904
+ },
905
+ {
906
+ "epoch": 0.49,
907
+ "learning_rate": 5.161775017529801e-05,
908
+ "loss": 3.0243,
909
+ "step": 15000
910
+ }
911
+ ],
912
+ "logging_steps": 100,
913
+ "max_steps": 30449,
914
+ "num_train_epochs": 1,
915
+ "save_steps": 100,
916
+ "total_flos": 1.105248502226903e+18,
917
+ "trial_name": null,
918
+ "trial_params": null
919
+ }