pszemraj commited on
Commit
5bef7c9
·
verified ·
1 Parent(s): 4a9b9a9

Delete checkpoint-5310

Browse files
checkpoint-5310/config.json DELETED
@@ -1,28 +0,0 @@
1
- {
2
- "_name_or_path": "BEE-spoke-data/smol_llama-220M-GQA",
3
- "architectures": [
4
- "LlamaForCausalLM"
5
- ],
6
- "attention_bias": false,
7
- "attention_dropout": 0.0,
8
- "bos_token_id": 1,
9
- "eos_token_id": 2,
10
- "hidden_act": "silu",
11
- "hidden_size": 1024,
12
- "initializer_range": 0.02,
13
- "intermediate_size": 4096,
14
- "max_position_embeddings": 2048,
15
- "model_type": "llama",
16
- "num_attention_heads": 32,
17
- "num_hidden_layers": 10,
18
- "num_key_value_heads": 8,
19
- "pretraining_tp": 1,
20
- "rms_norm_eps": 1e-06,
21
- "rope_scaling": null,
22
- "rope_theta": 10000.0,
23
- "tie_word_embeddings": false,
24
- "torch_dtype": "bfloat16",
25
- "transformers_version": "4.36.2",
26
- "use_cache": false,
27
- "vocab_size": 32128
28
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
checkpoint-5310/generation_config.json DELETED
@@ -1,7 +0,0 @@
1
- {
2
- "_from_model_config": true,
3
- "bos_token_id": 1,
4
- "eos_token_id": 2,
5
- "transformers_version": "4.36.2",
6
- "use_cache": false
7
- }
 
 
 
 
 
 
 
 
checkpoint-5310/model.safetensors DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:a79237ebc65b4599ece03b942f87bf91c054ead0ba4b67dbc177e702db5b92b7
3
- size 435736840
 
 
 
 
checkpoint-5310/trainer_state.json DELETED
@@ -1,3295 +0,0 @@
1
- {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 1.784623217922607,
5
- "eval_steps": 491,
6
- "global_step": 5310,
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
- "eval_loss": 2.4941999912261963,
14
- "eval_runtime": 2.652,
15
- "eval_samples_per_second": 915.917,
16
- "eval_steps_per_second": 114.631,
17
- "step": 1
18
- },
19
- {
20
- "epoch": 0.0,
21
- "learning_rate": 8.771929824561403e-06,
22
- "loss": 1.6371,
23
- "step": 10
24
- },
25
- {
26
- "epoch": 0.01,
27
- "learning_rate": 1.7543859649122806e-05,
28
- "loss": 1.5433,
29
- "step": 20
30
- },
31
- {
32
- "epoch": 0.01,
33
- "learning_rate": 2.6315789473684212e-05,
34
- "loss": 1.4686,
35
- "step": 30
36
- },
37
- {
38
- "epoch": 0.01,
39
- "learning_rate": 3.508771929824561e-05,
40
- "loss": 1.4179,
41
- "step": 40
42
- },
43
- {
44
- "epoch": 0.02,
45
- "learning_rate": 4.3859649122807014e-05,
46
- "loss": 1.4009,
47
- "step": 50
48
- },
49
- {
50
- "epoch": 0.02,
51
- "learning_rate": 5.2631578947368424e-05,
52
- "loss": 1.3749,
53
- "step": 60
54
- },
55
- {
56
- "epoch": 0.02,
57
- "learning_rate": 6.140350877192983e-05,
58
- "loss": 1.385,
59
- "step": 70
60
- },
61
- {
62
- "epoch": 0.03,
63
- "learning_rate": 7.017543859649122e-05,
64
- "loss": 1.3324,
65
- "step": 80
66
- },
67
- {
68
- "epoch": 0.03,
69
- "learning_rate": 7.894736842105263e-05,
70
- "loss": 1.3405,
71
- "step": 90
72
- },
73
- {
74
- "epoch": 0.03,
75
- "learning_rate": 8.771929824561403e-05,
76
- "loss": 1.343,
77
- "step": 100
78
- },
79
- {
80
- "epoch": 0.04,
81
- "learning_rate": 9.649122807017544e-05,
82
- "loss": 1.292,
83
- "step": 110
84
- },
85
- {
86
- "epoch": 0.04,
87
- "learning_rate": 0.00010526315789473685,
88
- "loss": 1.2815,
89
- "step": 120
90
- },
91
- {
92
- "epoch": 0.04,
93
- "learning_rate": 0.00011403508771929824,
94
- "loss": 1.3075,
95
- "step": 130
96
- },
97
- {
98
- "epoch": 0.05,
99
- "learning_rate": 0.00012280701754385965,
100
- "loss": 1.3112,
101
- "step": 140
102
- },
103
- {
104
- "epoch": 0.05,
105
- "learning_rate": 0.00013157894736842108,
106
- "loss": 1.2992,
107
- "step": 150
108
- },
109
- {
110
- "epoch": 0.05,
111
- "learning_rate": 0.00014035087719298245,
112
- "loss": 1.2765,
113
- "step": 160
114
- },
115
- {
116
- "epoch": 0.06,
117
- "learning_rate": 0.00014912280701754387,
118
- "loss": 1.2929,
119
- "step": 170
120
- },
121
- {
122
- "epoch": 0.06,
123
- "learning_rate": 0.00015789473684210527,
124
- "loss": 1.3165,
125
- "step": 180
126
- },
127
- {
128
- "epoch": 0.06,
129
- "learning_rate": 0.0001666666666666667,
130
- "loss": 1.2967,
131
- "step": 190
132
- },
133
- {
134
- "epoch": 0.07,
135
- "learning_rate": 0.00017543859649122806,
136
- "loss": 1.2985,
137
- "step": 200
138
- },
139
- {
140
- "epoch": 0.07,
141
- "learning_rate": 0.00018421052631578948,
142
- "loss": 1.2911,
143
- "step": 210
144
- },
145
- {
146
- "epoch": 0.07,
147
- "learning_rate": 0.00019298245614035088,
148
- "loss": 1.2855,
149
- "step": 220
150
- },
151
- {
152
- "epoch": 0.08,
153
- "learning_rate": 0.00019999993847052535,
154
- "loss": 1.3145,
155
- "step": 230
156
- },
157
- {
158
- "epoch": 0.08,
159
- "learning_rate": 0.00019999778494686225,
160
- "loss": 1.3182,
161
- "step": 240
162
- },
163
- {
164
- "epoch": 0.08,
165
- "learning_rate": 0.00019999255502518282,
166
- "loss": 1.3009,
167
- "step": 250
168
- },
169
- {
170
- "epoch": 0.09,
171
- "learning_rate": 0.0001999842488663838,
172
- "loss": 1.2914,
173
- "step": 260
174
- },
175
- {
176
- "epoch": 0.09,
177
- "learning_rate": 0.00019997286672600138,
178
- "loss": 1.3277,
179
- "step": 270
180
- },
181
- {
182
- "epoch": 0.1,
183
- "learning_rate": 0.0001999584089542032,
184
- "loss": 1.3203,
185
- "step": 280
186
- },
187
- {
188
- "epoch": 0.1,
189
- "learning_rate": 0.00019994087599577783,
190
- "loss": 1.2884,
191
- "step": 290
192
- },
193
- {
194
- "epoch": 0.1,
195
- "learning_rate": 0.00019992026839012067,
196
- "loss": 1.2917,
197
- "step": 300
198
- },
199
- {
200
- "epoch": 0.11,
201
- "learning_rate": 0.00019989658677121785,
202
- "loss": 1.2794,
203
- "step": 310
204
- },
205
- {
206
- "epoch": 0.11,
207
- "learning_rate": 0.00019986983186762628,
208
- "loss": 1.2987,
209
- "step": 320
210
- },
211
- {
212
- "epoch": 0.11,
213
- "learning_rate": 0.00019984000450245156,
214
- "loss": 1.2775,
215
- "step": 330
216
- },
217
- {
218
- "epoch": 0.12,
219
- "learning_rate": 0.00019980710559332248,
220
- "loss": 1.2878,
221
- "step": 340
222
- },
223
- {
224
- "epoch": 0.12,
225
- "learning_rate": 0.00019977113615236283,
226
- "loss": 1.2705,
227
- "step": 350
228
- },
229
- {
230
- "epoch": 0.12,
231
- "learning_rate": 0.00019973209728616032,
232
- "loss": 1.2717,
233
- "step": 360
234
- },
235
- {
236
- "epoch": 0.13,
237
- "learning_rate": 0.0001996899901957324,
238
- "loss": 1.2748,
239
- "step": 370
240
- },
241
- {
242
- "epoch": 0.13,
243
- "learning_rate": 0.00019964481617648953,
244
- "loss": 1.2804,
245
- "step": 380
246
- },
247
- {
248
- "epoch": 0.13,
249
- "learning_rate": 0.00019959657661819508,
250
- "loss": 1.2697,
251
- "step": 390
252
- },
253
- {
254
- "epoch": 0.14,
255
- "learning_rate": 0.00019954527300492276,
256
- "loss": 1.2716,
257
- "step": 400
258
- },
259
- {
260
- "epoch": 0.14,
261
- "learning_rate": 0.00019949090691501084,
262
- "loss": 1.27,
263
- "step": 410
264
- },
265
- {
266
- "epoch": 0.14,
267
- "learning_rate": 0.00019943348002101371,
268
- "loss": 1.2659,
269
- "step": 420
270
- },
271
- {
272
- "epoch": 0.15,
273
- "learning_rate": 0.00019937299408965035,
274
- "loss": 1.2765,
275
- "step": 430
276
- },
277
- {
278
- "epoch": 0.15,
279
- "learning_rate": 0.0001993094509817499,
280
- "loss": 1.2886,
281
- "step": 440
282
- },
283
- {
284
- "epoch": 0.15,
285
- "learning_rate": 0.00019924285265219467,
286
- "loss": 1.2459,
287
- "step": 450
288
- },
289
- {
290
- "epoch": 0.16,
291
- "learning_rate": 0.00019917320114985965,
292
- "loss": 1.2433,
293
- "step": 460
294
- },
295
- {
296
- "epoch": 0.16,
297
- "learning_rate": 0.00019910049861754977,
298
- "loss": 1.2468,
299
- "step": 470
300
- },
301
- {
302
- "epoch": 0.16,
303
- "learning_rate": 0.00019902474729193386,
304
- "loss": 1.239,
305
- "step": 480
306
- },
307
- {
308
- "epoch": 0.17,
309
- "learning_rate": 0.00019894594950347575,
310
- "loss": 1.2391,
311
- "step": 490
312
- },
313
- {
314
- "epoch": 0.17,
315
- "eval_loss": 1.329033374786377,
316
- "eval_runtime": 2.5491,
317
- "eval_samples_per_second": 952.867,
318
- "eval_steps_per_second": 119.255,
319
- "step": 491
320
- },
321
- {
322
- "epoch": 0.17,
323
- "learning_rate": 0.00019886410767636282,
324
- "loss": 1.2567,
325
- "step": 500
326
- },
327
- {
328
- "epoch": 0.17,
329
- "learning_rate": 0.00019877922432843116,
330
- "loss": 1.2706,
331
- "step": 510
332
- },
333
- {
334
- "epoch": 0.18,
335
- "learning_rate": 0.0001986913020710882,
336
- "loss": 1.2458,
337
- "step": 520
338
- },
339
- {
340
- "epoch": 0.18,
341
- "learning_rate": 0.0001986003436092325,
342
- "loss": 1.2401,
343
- "step": 530
344
- },
345
- {
346
- "epoch": 0.18,
347
- "learning_rate": 0.00019850635174117033,
348
- "loss": 1.2642,
349
- "step": 540
350
- },
351
- {
352
- "epoch": 0.19,
353
- "learning_rate": 0.00019840932935852968,
354
- "loss": 1.2225,
355
- "step": 550
356
- },
357
- {
358
- "epoch": 0.19,
359
- "learning_rate": 0.0001983092794461714,
360
- "loss": 1.2084,
361
- "step": 560
362
- },
363
- {
364
- "epoch": 0.19,
365
- "learning_rate": 0.00019820620508209712,
366
- "loss": 1.2257,
367
- "step": 570
368
- },
369
- {
370
- "epoch": 0.2,
371
- "learning_rate": 0.00019810010943735479,
372
- "loss": 1.2629,
373
- "step": 580
374
- },
375
- {
376
- "epoch": 0.2,
377
- "learning_rate": 0.00019799099577594097,
378
- "loss": 1.2333,
379
- "step": 590
380
- },
381
- {
382
- "epoch": 0.2,
383
- "learning_rate": 0.00019787886745470065,
384
- "loss": 1.2591,
385
- "step": 600
386
- },
387
- {
388
- "epoch": 0.21,
389
- "learning_rate": 0.00019776372792322357,
390
- "loss": 1.2474,
391
- "step": 610
392
- },
393
- {
394
- "epoch": 0.21,
395
- "learning_rate": 0.00019764558072373856,
396
- "loss": 1.2386,
397
- "step": 620
398
- },
399
- {
400
- "epoch": 0.21,
401
- "learning_rate": 0.00019752442949100415,
402
- "loss": 1.2251,
403
- "step": 630
404
- },
405
- {
406
- "epoch": 0.22,
407
- "learning_rate": 0.00019740027795219707,
408
- "loss": 1.2031,
409
- "step": 640
410
- },
411
- {
412
- "epoch": 0.22,
413
- "learning_rate": 0.00019727312992679733,
414
- "loss": 1.2242,
415
- "step": 650
416
- },
417
- {
418
- "epoch": 0.22,
419
- "learning_rate": 0.00019714298932647098,
420
- "loss": 1.2364,
421
- "step": 660
422
- },
423
- {
424
- "epoch": 0.23,
425
- "learning_rate": 0.00019700986015494957,
426
- "loss": 1.2115,
427
- "step": 670
428
- },
429
- {
430
- "epoch": 0.23,
431
- "learning_rate": 0.00019687374650790692,
432
- "loss": 1.1958,
433
- "step": 680
434
- },
435
- {
436
- "epoch": 0.23,
437
- "learning_rate": 0.00019673465257283342,
438
- "loss": 1.2332,
439
- "step": 690
440
- },
441
- {
442
- "epoch": 0.24,
443
- "learning_rate": 0.00019659258262890683,
444
- "loss": 1.2024,
445
- "step": 700
446
- },
447
- {
448
- "epoch": 0.24,
449
- "learning_rate": 0.00019644754104686097,
450
- "loss": 1.2209,
451
- "step": 710
452
- },
453
- {
454
- "epoch": 0.24,
455
- "learning_rate": 0.00019629953228885098,
456
- "loss": 1.2114,
457
- "step": 720
458
- },
459
- {
460
- "epoch": 0.25,
461
- "learning_rate": 0.00019614856090831626,
462
- "loss": 1.1819,
463
- "step": 730
464
- },
465
- {
466
- "epoch": 0.25,
467
- "learning_rate": 0.0001959946315498402,
468
- "loss": 1.1833,
469
- "step": 740
470
- },
471
- {
472
- "epoch": 0.25,
473
- "learning_rate": 0.00019583774894900747,
474
- "loss": 1.2144,
475
- "step": 750
476
- },
477
- {
478
- "epoch": 0.26,
479
- "learning_rate": 0.00019567791793225818,
480
- "loss": 1.2038,
481
- "step": 760
482
- },
483
- {
484
- "epoch": 0.26,
485
- "learning_rate": 0.00019551514341673948,
486
- "loss": 1.2166,
487
- "step": 770
488
- },
489
- {
490
- "epoch": 0.26,
491
- "learning_rate": 0.00019534943041015423,
492
- "loss": 1.1866,
493
- "step": 780
494
- },
495
- {
496
- "epoch": 0.27,
497
- "learning_rate": 0.00019518078401060707,
498
- "loss": 1.2352,
499
- "step": 790
500
- },
501
- {
502
- "epoch": 0.27,
503
- "learning_rate": 0.00019500920940644734,
504
- "loss": 1.1941,
505
- "step": 800
506
- },
507
- {
508
- "epoch": 0.27,
509
- "learning_rate": 0.00019483471187610972,
510
- "loss": 1.176,
511
- "step": 810
512
- },
513
- {
514
- "epoch": 0.28,
515
- "learning_rate": 0.00019465729678795168,
516
- "loss": 1.1938,
517
- "step": 820
518
- },
519
- {
520
- "epoch": 0.28,
521
- "learning_rate": 0.00019447696960008838,
522
- "loss": 1.2077,
523
- "step": 830
524
- },
525
- {
526
- "epoch": 0.29,
527
- "learning_rate": 0.00019429373586022473,
528
- "loss": 1.1884,
529
- "step": 840
530
- },
531
- {
532
- "epoch": 0.29,
533
- "learning_rate": 0.00019410760120548473,
534
- "loss": 1.2051,
535
- "step": 850
536
- },
537
- {
538
- "epoch": 0.29,
539
- "learning_rate": 0.00019391857136223802,
540
- "loss": 1.1529,
541
- "step": 860
542
- },
543
- {
544
- "epoch": 0.3,
545
- "learning_rate": 0.00019372665214592382,
546
- "loss": 1.1833,
547
- "step": 870
548
- },
549
- {
550
- "epoch": 0.3,
551
- "learning_rate": 0.00019353184946087183,
552
- "loss": 1.1731,
553
- "step": 880
554
- },
555
- {
556
- "epoch": 0.3,
557
- "learning_rate": 0.00019333416930012077,
558
- "loss": 1.1893,
559
- "step": 890
560
- },
561
- {
562
- "epoch": 0.31,
563
- "learning_rate": 0.00019313361774523385,
564
- "loss": 1.1804,
565
- "step": 900
566
- },
567
- {
568
- "epoch": 0.31,
569
- "learning_rate": 0.00019293020096611185,
570
- "loss": 1.1439,
571
- "step": 910
572
- },
573
- {
574
- "epoch": 0.31,
575
- "learning_rate": 0.0001927239252208031,
576
- "loss": 1.1609,
577
- "step": 920
578
- },
579
- {
580
- "epoch": 0.32,
581
- "learning_rate": 0.00019251479685531114,
582
- "loss": 1.1852,
583
- "step": 930
584
- },
585
- {
586
- "epoch": 0.32,
587
- "learning_rate": 0.00019230282230339935,
588
- "loss": 1.1543,
589
- "step": 940
590
- },
591
- {
592
- "epoch": 0.32,
593
- "learning_rate": 0.00019208800808639308,
594
- "loss": 1.1703,
595
- "step": 950
596
- },
597
- {
598
- "epoch": 0.33,
599
- "learning_rate": 0.00019187036081297907,
600
- "loss": 1.1608,
601
- "step": 960
602
- },
603
- {
604
- "epoch": 0.33,
605
- "learning_rate": 0.00019164988717900198,
606
- "loss": 1.1504,
607
- "step": 970
608
- },
609
- {
610
- "epoch": 0.33,
611
- "learning_rate": 0.0001914265939672586,
612
- "loss": 1.1867,
613
- "step": 980
614
- },
615
- {
616
- "epoch": 0.33,
617
- "eval_loss": 1.2649266719818115,
618
- "eval_runtime": 2.5506,
619
- "eval_samples_per_second": 952.326,
620
- "eval_steps_per_second": 119.188,
621
- "step": 982
622
- },
623
- {
624
- "epoch": 0.34,
625
- "learning_rate": 0.000191200488047289,
626
- "loss": 1.1861,
627
- "step": 990
628
- },
629
- {
630
- "epoch": 0.34,
631
- "learning_rate": 0.00019097157637516533,
632
- "loss": 1.2134,
633
- "step": 1000
634
- },
635
- {
636
- "epoch": 0.34,
637
- "learning_rate": 0.00019073986599327768,
638
- "loss": 1.1595,
639
- "step": 1010
640
- },
641
- {
642
- "epoch": 0.35,
643
- "learning_rate": 0.0001905053640301176,
644
- "loss": 1.2316,
645
- "step": 1020
646
- },
647
- {
648
- "epoch": 0.35,
649
- "learning_rate": 0.00019026807770005857,
650
- "loss": 1.1175,
651
- "step": 1030
652
- },
653
- {
654
- "epoch": 0.35,
655
- "learning_rate": 0.00019002801430313427,
656
- "loss": 1.1671,
657
- "step": 1040
658
- },
659
- {
660
- "epoch": 0.36,
661
- "learning_rate": 0.0001897851812248139,
662
- "loss": 1.1988,
663
- "step": 1050
664
- },
665
- {
666
- "epoch": 0.36,
667
- "learning_rate": 0.00018953958593577493,
668
- "loss": 1.1415,
669
- "step": 1060
670
- },
671
- {
672
- "epoch": 0.36,
673
- "learning_rate": 0.0001892912359916733,
674
- "loss": 1.1851,
675
- "step": 1070
676
- },
677
- {
678
- "epoch": 0.37,
679
- "learning_rate": 0.000189040139032911,
680
- "loss": 1.1746,
681
- "step": 1080
682
- },
683
- {
684
- "epoch": 0.37,
685
- "learning_rate": 0.00018878630278440102,
686
- "loss": 1.201,
687
- "step": 1090
688
- },
689
- {
690
- "epoch": 0.37,
691
- "learning_rate": 0.00018852973505532962,
692
- "loss": 1.1623,
693
- "step": 1100
694
- },
695
- {
696
- "epoch": 0.38,
697
- "learning_rate": 0.00018827044373891623,
698
- "loss": 1.153,
699
- "step": 1110
700
- },
701
- {
702
- "epoch": 0.38,
703
- "learning_rate": 0.0001880084368121704,
704
- "loss": 1.17,
705
- "step": 1120
706
- },
707
- {
708
- "epoch": 0.38,
709
- "learning_rate": 0.0001877437223356466,
710
- "loss": 1.1549,
711
- "step": 1130
712
- },
713
- {
714
- "epoch": 0.39,
715
- "learning_rate": 0.00018747630845319612,
716
- "loss": 1.1541,
717
- "step": 1140
718
- },
719
- {
720
- "epoch": 0.39,
721
- "learning_rate": 0.00018720620339171656,
722
- "loss": 1.1692,
723
- "step": 1150
724
- },
725
- {
726
- "epoch": 0.39,
727
- "learning_rate": 0.00018693341546089878,
728
- "loss": 1.1707,
729
- "step": 1160
730
- },
731
- {
732
- "epoch": 0.4,
733
- "learning_rate": 0.0001866579530529711,
734
- "loss": 1.1899,
735
- "step": 1170
736
- },
737
- {
738
- "epoch": 0.4,
739
- "learning_rate": 0.0001863798246424413,
740
- "loss": 1.165,
741
- "step": 1180
742
- },
743
- {
744
- "epoch": 0.4,
745
- "learning_rate": 0.00018609903878583585,
746
- "loss": 1.1669,
747
- "step": 1190
748
- },
749
- {
750
- "epoch": 0.41,
751
- "learning_rate": 0.00018581560412143661,
752
- "loss": 1.1309,
753
- "step": 1200
754
- },
755
- {
756
- "epoch": 0.41,
757
- "learning_rate": 0.0001855295293690151,
758
- "loss": 1.1622,
759
- "step": 1210
760
- },
761
- {
762
- "epoch": 0.41,
763
- "learning_rate": 0.00018524082332956428,
764
- "loss": 1.1471,
765
- "step": 1220
766
- },
767
- {
768
- "epoch": 0.42,
769
- "learning_rate": 0.00018494949488502774,
770
- "loss": 1.1352,
771
- "step": 1230
772
- },
773
- {
774
- "epoch": 0.42,
775
- "learning_rate": 0.00018465555299802656,
776
- "loss": 1.1646,
777
- "step": 1240
778
- },
779
- {
780
- "epoch": 0.42,
781
- "learning_rate": 0.00018435900671158334,
782
- "loss": 1.1649,
783
- "step": 1250
784
- },
785
- {
786
- "epoch": 0.43,
787
- "learning_rate": 0.00018405986514884434,
788
- "loss": 1.1693,
789
- "step": 1260
790
- },
791
- {
792
- "epoch": 0.43,
793
- "learning_rate": 0.0001837581375127984,
794
- "loss": 1.1341,
795
- "step": 1270
796
- },
797
- {
798
- "epoch": 0.43,
799
- "learning_rate": 0.00018345383308599424,
800
- "loss": 1.1467,
801
- "step": 1280
802
- },
803
- {
804
- "epoch": 0.44,
805
- "learning_rate": 0.00018314696123025454,
806
- "loss": 1.1563,
807
- "step": 1290
808
- },
809
- {
810
- "epoch": 0.44,
811
- "learning_rate": 0.00018283753138638808,
812
- "loss": 1.1624,
813
- "step": 1300
814
- },
815
- {
816
- "epoch": 0.44,
817
- "learning_rate": 0.00018252555307389935,
818
- "loss": 1.1402,
819
- "step": 1310
820
- },
821
- {
822
- "epoch": 0.45,
823
- "learning_rate": 0.00018221103589069552,
824
- "loss": 1.122,
825
- "step": 1320
826
- },
827
- {
828
- "epoch": 0.45,
829
- "learning_rate": 0.00018189398951279134,
830
- "loss": 1.1523,
831
- "step": 1330
832
- },
833
- {
834
- "epoch": 0.45,
835
- "learning_rate": 0.0001815744236940113,
836
- "loss": 1.1174,
837
- "step": 1340
838
- },
839
- {
840
- "epoch": 0.46,
841
- "learning_rate": 0.00018125234826568972,
842
- "loss": 1.1418,
843
- "step": 1350
844
- },
845
- {
846
- "epoch": 0.46,
847
- "learning_rate": 0.00018092777313636813,
848
- "loss": 1.1447,
849
- "step": 1360
850
- },
851
- {
852
- "epoch": 0.47,
853
- "learning_rate": 0.0001806007082914906,
854
- "loss": 1.1358,
855
- "step": 1370
856
- },
857
- {
858
- "epoch": 0.47,
859
- "learning_rate": 0.00018027116379309638,
860
- "loss": 1.1245,
861
- "step": 1380
862
- },
863
- {
864
- "epoch": 0.47,
865
- "learning_rate": 0.00017993914977951046,
866
- "loss": 1.1407,
867
- "step": 1390
868
- },
869
- {
870
- "epoch": 0.48,
871
- "learning_rate": 0.0001796046764650316,
872
- "loss": 1.1243,
873
- "step": 1400
874
- },
875
- {
876
- "epoch": 0.48,
877
- "learning_rate": 0.00017926775413961815,
878
- "loss": 1.129,
879
- "step": 1410
880
- },
881
- {
882
- "epoch": 0.48,
883
- "learning_rate": 0.00017892839316857145,
884
- "loss": 1.1337,
885
- "step": 1420
886
- },
887
- {
888
- "epoch": 0.49,
889
- "learning_rate": 0.00017858660399221697,
890
- "loss": 1.1148,
891
- "step": 1430
892
- },
893
- {
894
- "epoch": 0.49,
895
- "learning_rate": 0.00017824239712558302,
896
- "loss": 1.0814,
897
- "step": 1440
898
- },
899
- {
900
- "epoch": 0.49,
901
- "learning_rate": 0.0001778957831580774,
902
- "loss": 1.1489,
903
- "step": 1450
904
- },
905
- {
906
- "epoch": 0.5,
907
- "learning_rate": 0.0001775467727531615,
908
- "loss": 1.1341,
909
- "step": 1460
910
- },
911
- {
912
- "epoch": 0.5,
913
- "learning_rate": 0.00017719537664802234,
914
- "loss": 1.1031,
915
- "step": 1470
916
- },
917
- {
918
- "epoch": 0.5,
919
- "eval_loss": 1.2352267503738403,
920
- "eval_runtime": 2.5573,
921
- "eval_samples_per_second": 949.821,
922
- "eval_steps_per_second": 118.874,
923
- "step": 1473
924
- },
925
- {
926
- "epoch": 0.5,
927
- "learning_rate": 0.00017684160565324222,
928
- "loss": 1.1272,
929
- "step": 1480
930
- },
931
- {
932
- "epoch": 0.51,
933
- "learning_rate": 0.00017648547065246607,
934
- "loss": 1.1159,
935
- "step": 1490
936
- },
937
- {
938
- "epoch": 0.51,
939
- "learning_rate": 0.00017612698260206666,
940
- "loss": 1.1203,
941
- "step": 1500
942
- },
943
- {
944
- "epoch": 0.51,
945
- "learning_rate": 0.0001757661525308076,
946
- "loss": 1.0934,
947
- "step": 1510
948
- },
949
- {
950
- "epoch": 0.52,
951
- "learning_rate": 0.00017540299153950386,
952
- "loss": 1.0958,
953
- "step": 1520
954
- },
955
- {
956
- "epoch": 0.52,
957
- "learning_rate": 0.00017503751080068052,
958
- "loss": 1.1486,
959
- "step": 1530
960
- },
961
- {
962
- "epoch": 0.52,
963
- "learning_rate": 0.0001746697215582288,
964
- "loss": 1.1456,
965
- "step": 1540
966
- },
967
- {
968
- "epoch": 0.53,
969
- "learning_rate": 0.00017429963512706026,
970
- "loss": 1.1408,
971
- "step": 1550
972
- },
973
- {
974
- "epoch": 0.53,
975
- "learning_rate": 0.0001739272628927588,
976
- "loss": 1.0965,
977
- "step": 1560
978
- },
979
- {
980
- "epoch": 0.53,
981
- "learning_rate": 0.00017355261631123015,
982
- "loss": 1.1339,
983
- "step": 1570
984
- },
985
- {
986
- "epoch": 0.54,
987
- "learning_rate": 0.00017317570690834967,
988
- "loss": 1.1235,
989
- "step": 1580
990
- },
991
- {
992
- "epoch": 0.54,
993
- "learning_rate": 0.00017279654627960756,
994
- "loss": 1.1254,
995
- "step": 1590
996
- },
997
- {
998
- "epoch": 0.54,
999
- "learning_rate": 0.00017241514608975223,
1000
- "loss": 1.1341,
1001
- "step": 1600
1002
- },
1003
- {
1004
- "epoch": 0.55,
1005
- "learning_rate": 0.0001720315180724315,
1006
- "loss": 1.1417,
1007
- "step": 1610
1008
- },
1009
- {
1010
- "epoch": 0.55,
1011
- "learning_rate": 0.00017164567402983152,
1012
- "loss": 1.1376,
1013
- "step": 1620
1014
- },
1015
- {
1016
- "epoch": 0.55,
1017
- "learning_rate": 0.0001712576258323137,
1018
- "loss": 1.1065,
1019
- "step": 1630
1020
- },
1021
- {
1022
- "epoch": 0.56,
1023
- "learning_rate": 0.00017086738541804947,
1024
- "loss": 1.1081,
1025
- "step": 1640
1026
- },
1027
- {
1028
- "epoch": 0.56,
1029
- "learning_rate": 0.00017047496479265318,
1030
- "loss": 1.1324,
1031
- "step": 1650
1032
- },
1033
- {
1034
- "epoch": 0.56,
1035
- "learning_rate": 0.0001700803760288126,
1036
- "loss": 1.1182,
1037
- "step": 1660
1038
- },
1039
- {
1040
- "epoch": 0.57,
1041
- "learning_rate": 0.00016968363126591748,
1042
- "loss": 1.1284,
1043
- "step": 1670
1044
- },
1045
- {
1046
- "epoch": 0.57,
1047
- "learning_rate": 0.0001692847427096862,
1048
- "loss": 1.1011,
1049
- "step": 1680
1050
- },
1051
- {
1052
- "epoch": 0.57,
1053
- "learning_rate": 0.00016888372263179022,
1054
- "loss": 1.1196,
1055
- "step": 1690
1056
- },
1057
- {
1058
- "epoch": 0.58,
1059
- "learning_rate": 0.00016848058336947657,
1060
- "loss": 1.1012,
1061
- "step": 1700
1062
- },
1063
- {
1064
- "epoch": 0.58,
1065
- "learning_rate": 0.00016807533732518821,
1066
- "loss": 1.1076,
1067
- "step": 1710
1068
- },
1069
- {
1070
- "epoch": 0.58,
1071
- "learning_rate": 0.00016766799696618255,
1072
- "loss": 1.1175,
1073
- "step": 1720
1074
- },
1075
- {
1076
- "epoch": 0.59,
1077
- "learning_rate": 0.00016725857482414796,
1078
- "loss": 1.0902,
1079
- "step": 1730
1080
- },
1081
- {
1082
- "epoch": 0.59,
1083
- "learning_rate": 0.00016684708349481804,
1084
- "loss": 1.1322,
1085
- "step": 1740
1086
- },
1087
- {
1088
- "epoch": 0.59,
1089
- "learning_rate": 0.0001664335356375843,
1090
- "loss": 1.1016,
1091
- "step": 1750
1092
- },
1093
- {
1094
- "epoch": 0.6,
1095
- "learning_rate": 0.00016601794397510665,
1096
- "loss": 1.1073,
1097
- "step": 1760
1098
- },
1099
- {
1100
- "epoch": 0.6,
1101
- "learning_rate": 0.00016560032129292197,
1102
- "loss": 1.1127,
1103
- "step": 1770
1104
- },
1105
- {
1106
- "epoch": 0.6,
1107
- "learning_rate": 0.00016518068043905074,
1108
- "loss": 1.0561,
1109
- "step": 1780
1110
- },
1111
- {
1112
- "epoch": 0.61,
1113
- "learning_rate": 0.0001647590343236018,
1114
- "loss": 1.0951,
1115
- "step": 1790
1116
- },
1117
- {
1118
- "epoch": 0.61,
1119
- "learning_rate": 0.00016433539591837525,
1120
- "loss": 1.1138,
1121
- "step": 1800
1122
- },
1123
- {
1124
- "epoch": 0.61,
1125
- "learning_rate": 0.00016390977825646327,
1126
- "loss": 1.0807,
1127
- "step": 1810
1128
- },
1129
- {
1130
- "epoch": 0.62,
1131
- "learning_rate": 0.00016348219443184923,
1132
- "loss": 1.1041,
1133
- "step": 1820
1134
- },
1135
- {
1136
- "epoch": 0.62,
1137
- "learning_rate": 0.00016305265759900477,
1138
- "loss": 1.0948,
1139
- "step": 1830
1140
- },
1141
- {
1142
- "epoch": 0.62,
1143
- "learning_rate": 0.0001626211809724852,
1144
- "loss": 1.1167,
1145
- "step": 1840
1146
- },
1147
- {
1148
- "epoch": 0.63,
1149
- "learning_rate": 0.000162187777826523,
1150
- "loss": 1.1128,
1151
- "step": 1850
1152
- },
1153
- {
1154
- "epoch": 0.63,
1155
- "learning_rate": 0.0001617524614946192,
1156
- "loss": 1.124,
1157
- "step": 1860
1158
- },
1159
- {
1160
- "epoch": 0.63,
1161
- "learning_rate": 0.00016131524536913352,
1162
- "loss": 1.0904,
1163
- "step": 1870
1164
- },
1165
- {
1166
- "epoch": 0.64,
1167
- "learning_rate": 0.00016087614290087208,
1168
- "loss": 1.0797,
1169
- "step": 1880
1170
- },
1171
- {
1172
- "epoch": 0.64,
1173
- "learning_rate": 0.00016043516759867374,
1174
- "loss": 1.1135,
1175
- "step": 1890
1176
- },
1177
- {
1178
- "epoch": 0.64,
1179
- "learning_rate": 0.00015999233302899443,
1180
- "loss": 1.0823,
1181
- "step": 1900
1182
- },
1183
- {
1184
- "epoch": 0.65,
1185
- "learning_rate": 0.0001595476528154899,
1186
- "loss": 1.1011,
1187
- "step": 1910
1188
- },
1189
- {
1190
- "epoch": 0.65,
1191
- "learning_rate": 0.0001591011406385964,
1192
- "loss": 1.0804,
1193
- "step": 1920
1194
- },
1195
- {
1196
- "epoch": 0.66,
1197
- "learning_rate": 0.00015865281023510997,
1198
- "loss": 1.0908,
1199
- "step": 1930
1200
- },
1201
- {
1202
- "epoch": 0.66,
1203
- "learning_rate": 0.00015820267539776376,
1204
- "loss": 1.1049,
1205
- "step": 1940
1206
- },
1207
- {
1208
- "epoch": 0.66,
1209
- "learning_rate": 0.00015775074997480374,
1210
- "loss": 1.0597,
1211
- "step": 1950
1212
- },
1213
- {
1214
- "epoch": 0.67,
1215
- "learning_rate": 0.0001572970478695626,
1216
- "loss": 1.0856,
1217
- "step": 1960
1218
- },
1219
- {
1220
- "epoch": 0.67,
1221
- "eval_loss": 1.2079788446426392,
1222
- "eval_runtime": 2.5537,
1223
- "eval_samples_per_second": 951.168,
1224
- "eval_steps_per_second": 119.043,
1225
- "step": 1964
1226
- },
1227
- {
1228
- "epoch": 0.67,
1229
- "learning_rate": 0.00015684158304003215,
1230
- "loss": 1.092,
1231
- "step": 1970
1232
- },
1233
- {
1234
- "epoch": 0.67,
1235
- "learning_rate": 0.0001563843694984336,
1236
- "loss": 1.1,
1237
- "step": 1980
1238
- },
1239
- {
1240
- "epoch": 0.68,
1241
- "learning_rate": 0.000155925421310787,
1242
- "loss": 1.0925,
1243
- "step": 1990
1244
- },
1245
- {
1246
- "epoch": 0.68,
1247
- "learning_rate": 0.00015546475259647794,
1248
- "loss": 1.0717,
1249
- "step": 2000
1250
- },
1251
- {
1252
- "epoch": 0.68,
1253
- "learning_rate": 0.00015500237752782362,
1254
- "loss": 1.0865,
1255
- "step": 2010
1256
- },
1257
- {
1258
- "epoch": 0.69,
1259
- "learning_rate": 0.0001545383103296365,
1260
- "loss": 1.0816,
1261
- "step": 2020
1262
- },
1263
- {
1264
- "epoch": 0.69,
1265
- "learning_rate": 0.0001540725652787869,
1266
- "loss": 1.0913,
1267
- "step": 2030
1268
- },
1269
- {
1270
- "epoch": 0.69,
1271
- "learning_rate": 0.00015360515670376375,
1272
- "loss": 1.0932,
1273
- "step": 2040
1274
- },
1275
- {
1276
- "epoch": 0.7,
1277
- "learning_rate": 0.00015313609898423366,
1278
- "loss": 1.0835,
1279
- "step": 2050
1280
- },
1281
- {
1282
- "epoch": 0.7,
1283
- "learning_rate": 0.0001526654065505987,
1284
- "loss": 1.1036,
1285
- "step": 2060
1286
- },
1287
- {
1288
- "epoch": 0.7,
1289
- "learning_rate": 0.00015219309388355226,
1290
- "loss": 1.0698,
1291
- "step": 2070
1292
- },
1293
- {
1294
- "epoch": 0.71,
1295
- "learning_rate": 0.00015171917551363374,
1296
- "loss": 1.0814,
1297
- "step": 2080
1298
- },
1299
- {
1300
- "epoch": 0.71,
1301
- "learning_rate": 0.00015124366602078146,
1302
- "loss": 1.0823,
1303
- "step": 2090
1304
- },
1305
- {
1306
- "epoch": 0.71,
1307
- "learning_rate": 0.000150766580033884,
1308
- "loss": 1.0964,
1309
- "step": 2100
1310
- },
1311
- {
1312
- "epoch": 0.72,
1313
- "learning_rate": 0.00015028793223033037,
1314
- "loss": 1.0968,
1315
- "step": 2110
1316
- },
1317
- {
1318
- "epoch": 0.72,
1319
- "learning_rate": 0.0001498077373355582,
1320
- "loss": 1.0848,
1321
- "step": 2120
1322
- },
1323
- {
1324
- "epoch": 0.72,
1325
- "learning_rate": 0.00014932601012260105,
1326
- "loss": 1.0647,
1327
- "step": 2130
1328
- },
1329
- {
1330
- "epoch": 0.73,
1331
- "learning_rate": 0.0001488427654116335,
1332
- "loss": 1.0795,
1333
- "step": 2140
1334
- },
1335
- {
1336
- "epoch": 0.73,
1337
- "learning_rate": 0.00014835801806951566,
1338
- "loss": 1.0949,
1339
- "step": 2150
1340
- },
1341
- {
1342
- "epoch": 0.73,
1343
- "learning_rate": 0.00014787178300933542,
1344
- "loss": 1.0942,
1345
- "step": 2160
1346
- },
1347
- {
1348
- "epoch": 0.74,
1349
- "learning_rate": 0.0001473840751899499,
1350
- "loss": 1.0799,
1351
- "step": 2170
1352
- },
1353
- {
1354
- "epoch": 0.74,
1355
- "learning_rate": 0.00014689490961552513,
1356
- "loss": 1.0883,
1357
- "step": 2180
1358
- },
1359
- {
1360
- "epoch": 0.74,
1361
- "learning_rate": 0.00014640430133507445,
1362
- "loss": 1.0913,
1363
- "step": 2190
1364
- },
1365
- {
1366
- "epoch": 0.75,
1367
- "learning_rate": 0.00014591226544199564,
1368
- "loss": 1.0884,
1369
- "step": 2200
1370
- },
1371
- {
1372
- "epoch": 0.75,
1373
- "learning_rate": 0.00014541881707360642,
1374
- "loss": 1.0917,
1375
- "step": 2210
1376
- },
1377
- {
1378
- "epoch": 0.75,
1379
- "learning_rate": 0.00014492397141067887,
1380
- "loss": 1.0883,
1381
- "step": 2220
1382
- },
1383
- {
1384
- "epoch": 0.76,
1385
- "learning_rate": 0.0001444277436769724,
1386
- "loss": 1.1236,
1387
- "step": 2230
1388
- },
1389
- {
1390
- "epoch": 0.76,
1391
- "learning_rate": 0.0001439301491387653,
1392
- "loss": 1.0542,
1393
- "step": 2240
1394
- },
1395
- {
1396
- "epoch": 0.76,
1397
- "learning_rate": 0.00014343120310438513,
1398
- "loss": 1.0335,
1399
- "step": 2250
1400
- },
1401
- {
1402
- "epoch": 0.77,
1403
- "learning_rate": 0.00014293092092373786,
1404
- "loss": 1.0561,
1405
- "step": 2260
1406
- },
1407
- {
1408
- "epoch": 0.77,
1409
- "learning_rate": 0.00014242931798783546,
1410
- "loss": 1.0721,
1411
- "step": 2270
1412
- },
1413
- {
1414
- "epoch": 0.77,
1415
- "learning_rate": 0.0001419264097283225,
1416
- "loss": 1.0415,
1417
- "step": 2280
1418
- },
1419
- {
1420
- "epoch": 0.78,
1421
- "learning_rate": 0.00014142221161700142,
1422
- "loss": 1.0871,
1423
- "step": 2290
1424
- },
1425
- {
1426
- "epoch": 0.78,
1427
- "learning_rate": 0.00014091673916535654,
1428
- "loss": 1.044,
1429
- "step": 2300
1430
- },
1431
- {
1432
- "epoch": 0.78,
1433
- "learning_rate": 0.00014041000792407675,
1434
- "loss": 1.0579,
1435
- "step": 2310
1436
- },
1437
- {
1438
- "epoch": 0.79,
1439
- "learning_rate": 0.00013990203348257721,
1440
- "loss": 1.052,
1441
- "step": 2320
1442
- },
1443
- {
1444
- "epoch": 0.79,
1445
- "learning_rate": 0.0001393928314685198,
1446
- "loss": 1.0641,
1447
- "step": 2330
1448
- },
1449
- {
1450
- "epoch": 0.79,
1451
- "learning_rate": 0.00013888241754733208,
1452
- "loss": 1.0662,
1453
- "step": 2340
1454
- },
1455
- {
1456
- "epoch": 0.8,
1457
- "learning_rate": 0.0001383708074217257,
1458
- "loss": 1.0359,
1459
- "step": 2350
1460
- },
1461
- {
1462
- "epoch": 0.8,
1463
- "learning_rate": 0.00013785801683121294,
1464
- "loss": 1.0764,
1465
- "step": 2360
1466
- },
1467
- {
1468
- "epoch": 0.8,
1469
- "learning_rate": 0.00013734406155162292,
1470
- "loss": 1.044,
1471
- "step": 2370
1472
- },
1473
- {
1474
- "epoch": 0.81,
1475
- "learning_rate": 0.00013682895739461587,
1476
- "loss": 1.049,
1477
- "step": 2380
1478
- },
1479
- {
1480
- "epoch": 0.81,
1481
- "learning_rate": 0.00013631272020719686,
1482
- "loss": 1.0523,
1483
- "step": 2390
1484
- },
1485
- {
1486
- "epoch": 0.81,
1487
- "learning_rate": 0.00013579536587122828,
1488
- "loss": 1.0728,
1489
- "step": 2400
1490
- },
1491
- {
1492
- "epoch": 0.82,
1493
- "learning_rate": 0.0001352769103029412,
1494
- "loss": 1.0209,
1495
- "step": 2410
1496
- },
1497
- {
1498
- "epoch": 0.82,
1499
- "learning_rate": 0.00013475736945244574,
1500
- "loss": 1.0455,
1501
- "step": 2420
1502
- },
1503
- {
1504
- "epoch": 0.82,
1505
- "learning_rate": 0.0001342367593032404,
1506
- "loss": 1.0341,
1507
- "step": 2430
1508
- },
1509
- {
1510
- "epoch": 0.83,
1511
- "learning_rate": 0.00013371509587172022,
1512
- "loss": 1.0626,
1513
- "step": 2440
1514
- },
1515
- {
1516
- "epoch": 0.83,
1517
- "learning_rate": 0.00013319239520668416,
1518
- "loss": 1.0602,
1519
- "step": 2450
1520
- },
1521
- {
1522
- "epoch": 0.83,
1523
- "eval_loss": 1.1820294857025146,
1524
- "eval_runtime": 2.5668,
1525
- "eval_samples_per_second": 946.308,
1526
- "eval_steps_per_second": 118.435,
1527
- "step": 2455
1528
- },
1529
- {
1530
- "epoch": 0.84,
1531
- "learning_rate": 0.0001326686733888413,
1532
- "loss": 1.048,
1533
- "step": 2460
1534
- },
1535
- {
1536
- "epoch": 0.84,
1537
- "learning_rate": 0.00013214394653031616,
1538
- "loss": 1.0772,
1539
- "step": 2470
1540
- },
1541
- {
1542
- "epoch": 0.84,
1543
- "learning_rate": 0.00013161823077415297,
1544
- "loss": 1.0481,
1545
- "step": 2480
1546
- },
1547
- {
1548
- "epoch": 0.85,
1549
- "learning_rate": 0.0001310915422938191,
1550
- "loss": 1.0333,
1551
- "step": 2490
1552
- },
1553
- {
1554
- "epoch": 0.85,
1555
- "learning_rate": 0.00013056389729270738,
1556
- "loss": 1.0429,
1557
- "step": 2500
1558
- },
1559
- {
1560
- "epoch": 0.85,
1561
- "learning_rate": 0.0001300353120036378,
1562
- "loss": 1.012,
1563
- "step": 2510
1564
- },
1565
- {
1566
- "epoch": 0.86,
1567
- "learning_rate": 0.00012950580268835784,
1568
- "loss": 1.0759,
1569
- "step": 2520
1570
- },
1571
- {
1572
- "epoch": 0.86,
1573
- "learning_rate": 0.0001289753856370424,
1574
- "loss": 1.0514,
1575
- "step": 2530
1576
- },
1577
- {
1578
- "epoch": 0.86,
1579
- "learning_rate": 0.00012844407716779266,
1580
- "loss": 1.0419,
1581
- "step": 2540
1582
- },
1583
- {
1584
- "epoch": 0.87,
1585
- "learning_rate": 0.0001279118936261338,
1586
- "loss": 1.0373,
1587
- "step": 2550
1588
- },
1589
- {
1590
- "epoch": 0.87,
1591
- "learning_rate": 0.00012737885138451246,
1592
- "loss": 1.0306,
1593
- "step": 2560
1594
- },
1595
- {
1596
- "epoch": 0.87,
1597
- "learning_rate": 0.0001268449668417928,
1598
- "loss": 1.0389,
1599
- "step": 2570
1600
- },
1601
- {
1602
- "epoch": 0.88,
1603
- "learning_rate": 0.00012631025642275212,
1604
- "loss": 1.0332,
1605
- "step": 2580
1606
- },
1607
- {
1608
- "epoch": 0.88,
1609
- "learning_rate": 0.00012577473657757556,
1610
- "loss": 1.0231,
1611
- "step": 2590
1612
- },
1613
- {
1614
- "epoch": 0.88,
1615
- "learning_rate": 0.00012523842378134992,
1616
- "loss": 1.0648,
1617
- "step": 2600
1618
- },
1619
- {
1620
- "epoch": 0.89,
1621
- "learning_rate": 0.0001247013345335569,
1622
- "loss": 1.0809,
1623
- "step": 2610
1624
- },
1625
- {
1626
- "epoch": 0.89,
1627
- "learning_rate": 0.00012416348535756544,
1628
- "loss": 1.0716,
1629
- "step": 2620
1630
- },
1631
- {
1632
- "epoch": 0.89,
1633
- "learning_rate": 0.00012362489280012348,
1634
- "loss": 1.0139,
1635
- "step": 2630
1636
- },
1637
- {
1638
- "epoch": 0.9,
1639
- "learning_rate": 0.0001230855734308488,
1640
- "loss": 1.0637,
1641
- "step": 2640
1642
- },
1643
- {
1644
- "epoch": 0.9,
1645
- "learning_rate": 0.00012254554384171924,
1646
- "loss": 1.0393,
1647
- "step": 2650
1648
- },
1649
- {
1650
- "epoch": 0.9,
1651
- "learning_rate": 0.00012200482064656248,
1652
- "loss": 1.0292,
1653
- "step": 2660
1654
- },
1655
- {
1656
- "epoch": 0.91,
1657
- "learning_rate": 0.00012146342048054453,
1658
- "loss": 1.04,
1659
- "step": 2670
1660
- },
1661
- {
1662
- "epoch": 0.91,
1663
- "learning_rate": 0.0001209213599996584,
1664
- "loss": 1.0494,
1665
- "step": 2680
1666
- },
1667
- {
1668
- "epoch": 0.91,
1669
- "learning_rate": 0.00012037865588021127,
1670
- "loss": 1.021,
1671
- "step": 2690
1672
- },
1673
- {
1674
- "epoch": 0.92,
1675
- "learning_rate": 0.0001198353248183118,
1676
- "loss": 1.0079,
1677
- "step": 2700
1678
- },
1679
- {
1680
- "epoch": 0.92,
1681
- "learning_rate": 0.00011929138352935623,
1682
- "loss": 1.0282,
1683
- "step": 2710
1684
- },
1685
- {
1686
- "epoch": 0.92,
1687
- "learning_rate": 0.00011874684874751421,
1688
- "loss": 1.0061,
1689
- "step": 2720
1690
- },
1691
- {
1692
- "epoch": 0.93,
1693
- "learning_rate": 0.00011820173722521413,
1694
- "loss": 1.0071,
1695
- "step": 2730
1696
- },
1697
- {
1698
- "epoch": 0.93,
1699
- "learning_rate": 0.00011765606573262744,
1700
- "loss": 0.9862,
1701
- "step": 2740
1702
- },
1703
- {
1704
- "epoch": 0.93,
1705
- "learning_rate": 0.00011710985105715302,
1706
- "loss": 1.0342,
1707
- "step": 2750
1708
- },
1709
- {
1710
- "epoch": 0.94,
1711
- "learning_rate": 0.0001165631100029005,
1712
- "loss": 1.0448,
1713
- "step": 2760
1714
- },
1715
- {
1716
- "epoch": 0.94,
1717
- "learning_rate": 0.00011601585939017344,
1718
- "loss": 1.037,
1719
- "step": 2770
1720
- },
1721
- {
1722
- "epoch": 0.94,
1723
- "learning_rate": 0.00011546811605495169,
1724
- "loss": 1.0428,
1725
- "step": 2780
1726
- },
1727
- {
1728
- "epoch": 0.95,
1729
- "learning_rate": 0.00011491989684837366,
1730
- "loss": 1.0409,
1731
- "step": 2790
1732
- },
1733
- {
1734
- "epoch": 0.95,
1735
- "learning_rate": 0.00011437121863621775,
1736
- "loss": 1.0337,
1737
- "step": 2800
1738
- },
1739
- {
1740
- "epoch": 0.95,
1741
- "learning_rate": 0.00011382209829838347,
1742
- "loss": 1.0188,
1743
- "step": 2810
1744
- },
1745
- {
1746
- "epoch": 0.96,
1747
- "learning_rate": 0.00011327255272837221,
1748
- "loss": 1.0313,
1749
- "step": 2820
1750
- },
1751
- {
1752
- "epoch": 0.96,
1753
- "learning_rate": 0.00011272259883276745,
1754
- "loss": 1.0414,
1755
- "step": 2830
1756
- },
1757
- {
1758
- "epoch": 0.96,
1759
- "learning_rate": 0.00011217225353071472,
1760
- "loss": 1.0126,
1761
- "step": 2840
1762
- },
1763
- {
1764
- "epoch": 0.97,
1765
- "learning_rate": 0.00011162153375340101,
1766
- "loss": 1.0258,
1767
- "step": 2850
1768
- },
1769
- {
1770
- "epoch": 0.97,
1771
- "learning_rate": 0.00011107045644353387,
1772
- "loss": 1.0141,
1773
- "step": 2860
1774
- },
1775
- {
1776
- "epoch": 0.97,
1777
- "learning_rate": 0.00011051903855482034,
1778
- "loss": 1.022,
1779
- "step": 2870
1780
- },
1781
- {
1782
- "epoch": 0.98,
1783
- "learning_rate": 0.0001099672970514451,
1784
- "loss": 1.0469,
1785
- "step": 2880
1786
- },
1787
- {
1788
- "epoch": 0.98,
1789
- "learning_rate": 0.00010941524890754884,
1790
- "loss": 1.0254,
1791
- "step": 2890
1792
- },
1793
- {
1794
- "epoch": 0.98,
1795
- "learning_rate": 0.00010886291110670593,
1796
- "loss": 1.0386,
1797
- "step": 2900
1798
- },
1799
- {
1800
- "epoch": 0.99,
1801
- "learning_rate": 0.00010831030064140184,
1802
- "loss": 1.0029,
1803
- "step": 2910
1804
- },
1805
- {
1806
- "epoch": 0.99,
1807
- "learning_rate": 0.00010775743451251059,
1808
- "loss": 1.0349,
1809
- "step": 2920
1810
- },
1811
- {
1812
- "epoch": 0.99,
1813
- "learning_rate": 0.00010720432972877155,
1814
- "loss": 1.0296,
1815
- "step": 2930
1816
- },
1817
- {
1818
- "epoch": 1.0,
1819
- "learning_rate": 0.00010665100330626625,
1820
- "loss": 1.0105,
1821
- "step": 2940
1822
- },
1823
- {
1824
- "epoch": 1.0,
1825
- "eval_loss": 1.1566660404205322,
1826
- "eval_runtime": 2.5517,
1827
- "eval_samples_per_second": 951.906,
1828
- "eval_steps_per_second": 119.135,
1829
- "step": 2946
1830
- },
1831
- {
1832
- "epoch": 1.0,
1833
- "learning_rate": 0.00010609747226789476,
1834
- "loss": 1.0512,
1835
- "step": 2950
1836
- },
1837
- {
1838
- "epoch": 1.0,
1839
- "learning_rate": 0.0001055437536428523,
1840
- "loss": 0.9893,
1841
- "step": 2960
1842
- },
1843
- {
1844
- "epoch": 1.01,
1845
- "learning_rate": 0.00010498986446610488,
1846
- "loss": 0.9981,
1847
- "step": 2970
1848
- },
1849
- {
1850
- "epoch": 1.01,
1851
- "learning_rate": 0.00010443582177786564,
1852
- "loss": 1.0188,
1853
- "step": 2980
1854
- },
1855
- {
1856
- "epoch": 1.01,
1857
- "learning_rate": 0.00010388164262307046,
1858
- "loss": 0.9988,
1859
- "step": 2990
1860
- },
1861
- {
1862
- "epoch": 1.0,
1863
- "learning_rate": 0.00010332734405085342,
1864
- "loss": 1.0102,
1865
- "step": 3000
1866
- },
1867
- {
1868
- "epoch": 1.0,
1869
- "learning_rate": 0.00010277294311402259,
1870
- "loss": 0.9336,
1871
- "step": 3010
1872
- },
1873
- {
1874
- "epoch": 1.01,
1875
- "learning_rate": 0.00010221845686853517,
1876
- "loss": 0.8864,
1877
- "step": 3020
1878
- },
1879
- {
1880
- "epoch": 1.01,
1881
- "learning_rate": 0.00010166390237297287,
1882
- "loss": 0.8955,
1883
- "step": 3030
1884
- },
1885
- {
1886
- "epoch": 1.01,
1887
- "learning_rate": 0.00010110929668801711,
1888
- "loss": 0.9231,
1889
- "step": 3040
1890
- },
1891
- {
1892
- "epoch": 1.02,
1893
- "learning_rate": 0.00010055465687592411,
1894
- "loss": 0.9006,
1895
- "step": 3050
1896
- },
1897
- {
1898
- "epoch": 1.02,
1899
- "learning_rate": 0.0001,
1900
- "loss": 0.9126,
1901
- "step": 3060
1902
- },
1903
- {
1904
- "epoch": 1.02,
1905
- "learning_rate": 9.944534312407591e-05,
1906
- "loss": 0.8792,
1907
- "step": 3070
1908
- },
1909
- {
1910
- "epoch": 1.03,
1911
- "learning_rate": 9.889070331198292e-05,
1912
- "loss": 0.8911,
1913
- "step": 3080
1914
- },
1915
- {
1916
- "epoch": 1.03,
1917
- "learning_rate": 9.833609762702714e-05,
1918
- "loss": 0.8952,
1919
- "step": 3090
1920
- },
1921
- {
1922
- "epoch": 1.03,
1923
- "learning_rate": 9.778154313146484e-05,
1924
- "loss": 0.9196,
1925
- "step": 3100
1926
- },
1927
- {
1928
- "epoch": 1.04,
1929
- "learning_rate": 9.722705688597744e-05,
1930
- "loss": 0.9232,
1931
- "step": 3110
1932
- },
1933
- {
1934
- "epoch": 1.04,
1935
- "learning_rate": 9.667265594914662e-05,
1936
- "loss": 0.8986,
1937
- "step": 3120
1938
- },
1939
- {
1940
- "epoch": 1.04,
1941
- "learning_rate": 9.611835737692958e-05,
1942
- "loss": 0.8852,
1943
- "step": 3130
1944
- },
1945
- {
1946
- "epoch": 1.05,
1947
- "learning_rate": 9.556417822213435e-05,
1948
- "loss": 0.8771,
1949
- "step": 3140
1950
- },
1951
- {
1952
- "epoch": 1.05,
1953
- "learning_rate": 9.501013553389516e-05,
1954
- "loss": 0.8751,
1955
- "step": 3150
1956
- },
1957
- {
1958
- "epoch": 1.05,
1959
- "learning_rate": 9.445624635714773e-05,
1960
- "loss": 0.949,
1961
- "step": 3160
1962
- },
1963
- {
1964
- "epoch": 1.06,
1965
- "learning_rate": 9.390252773210524e-05,
1966
- "loss": 0.9212,
1967
- "step": 3170
1968
- },
1969
- {
1970
- "epoch": 1.06,
1971
- "learning_rate": 9.334899669373379e-05,
1972
- "loss": 0.9122,
1973
- "step": 3180
1974
- },
1975
- {
1976
- "epoch": 1.07,
1977
- "learning_rate": 9.279567027122845e-05,
1978
- "loss": 0.8839,
1979
- "step": 3190
1980
- },
1981
- {
1982
- "epoch": 1.07,
1983
- "learning_rate": 9.224256548748942e-05,
1984
- "loss": 0.894,
1985
- "step": 3200
1986
- },
1987
- {
1988
- "epoch": 1.07,
1989
- "learning_rate": 9.16896993585982e-05,
1990
- "loss": 0.8844,
1991
- "step": 3210
1992
- },
1993
- {
1994
- "epoch": 1.08,
1995
- "learning_rate": 9.113708889329413e-05,
1996
- "loss": 0.854,
1997
- "step": 3220
1998
- },
1999
- {
2000
- "epoch": 1.08,
2001
- "learning_rate": 9.058475109245116e-05,
2002
- "loss": 0.8669,
2003
- "step": 3230
2004
- },
2005
- {
2006
- "epoch": 1.08,
2007
- "learning_rate": 9.003270294855491e-05,
2008
- "loss": 0.9132,
2009
- "step": 3240
2010
- },
2011
- {
2012
- "epoch": 1.09,
2013
- "learning_rate": 8.94809614451797e-05,
2014
- "loss": 0.9012,
2015
- "step": 3250
2016
- },
2017
- {
2018
- "epoch": 1.09,
2019
- "learning_rate": 8.892954355646616e-05,
2020
- "loss": 0.884,
2021
- "step": 3260
2022
- },
2023
- {
2024
- "epoch": 1.09,
2025
- "learning_rate": 8.837846624659904e-05,
2026
- "loss": 0.8963,
2027
- "step": 3270
2028
- },
2029
- {
2030
- "epoch": 1.1,
2031
- "learning_rate": 8.78277464692853e-05,
2032
- "loss": 0.9208,
2033
- "step": 3280
2034
- },
2035
- {
2036
- "epoch": 1.1,
2037
- "learning_rate": 8.727740116723257e-05,
2038
- "loss": 0.8954,
2039
- "step": 3290
2040
- },
2041
- {
2042
- "epoch": 1.1,
2043
- "learning_rate": 8.672744727162781e-05,
2044
- "loss": 0.8794,
2045
- "step": 3300
2046
- },
2047
- {
2048
- "epoch": 1.11,
2049
- "learning_rate": 8.617790170161656e-05,
2050
- "loss": 0.898,
2051
- "step": 3310
2052
- },
2053
- {
2054
- "epoch": 1.11,
2055
- "learning_rate": 8.562878136378226e-05,
2056
- "loss": 0.9053,
2057
- "step": 3320
2058
- },
2059
- {
2060
- "epoch": 1.11,
2061
- "learning_rate": 8.508010315162634e-05,
2062
- "loss": 0.9039,
2063
- "step": 3330
2064
- },
2065
- {
2066
- "epoch": 1.12,
2067
- "learning_rate": 8.453188394504834e-05,
2068
- "loss": 0.8774,
2069
- "step": 3340
2070
- },
2071
- {
2072
- "epoch": 1.12,
2073
- "learning_rate": 8.398414060982659e-05,
2074
- "loss": 0.9022,
2075
- "step": 3350
2076
- },
2077
- {
2078
- "epoch": 1.12,
2079
- "learning_rate": 8.343688999709952e-05,
2080
- "loss": 0.8809,
2081
- "step": 3360
2082
- },
2083
- {
2084
- "epoch": 1.13,
2085
- "learning_rate": 8.289014894284702e-05,
2086
- "loss": 0.919,
2087
- "step": 3370
2088
- },
2089
- {
2090
- "epoch": 1.13,
2091
- "learning_rate": 8.234393426737258e-05,
2092
- "loss": 0.8623,
2093
- "step": 3380
2094
- },
2095
- {
2096
- "epoch": 1.13,
2097
- "learning_rate": 8.179826277478589e-05,
2098
- "loss": 0.923,
2099
- "step": 3390
2100
- },
2101
- {
2102
- "epoch": 1.14,
2103
- "learning_rate": 8.12531512524858e-05,
2104
- "loss": 0.8968,
2105
- "step": 3400
2106
- },
2107
- {
2108
- "epoch": 1.14,
2109
- "learning_rate": 8.070861647064383e-05,
2110
- "loss": 0.9099,
2111
- "step": 3410
2112
- },
2113
- {
2114
- "epoch": 1.14,
2115
- "learning_rate": 8.016467518168821e-05,
2116
- "loss": 0.9087,
2117
- "step": 3420
2118
- },
2119
- {
2120
- "epoch": 1.15,
2121
- "learning_rate": 7.962134411978874e-05,
2122
- "loss": 0.9047,
2123
- "step": 3430
2124
- },
2125
- {
2126
- "epoch": 1.15,
2127
- "eval_loss": 1.1462568044662476,
2128
- "eval_runtime": 2.5537,
2129
- "eval_samples_per_second": 951.181,
2130
- "eval_steps_per_second": 119.044,
2131
- "step": 3437
2132
- },
2133
- {
2134
- "epoch": 1.15,
2135
- "learning_rate": 7.907864000034163e-05,
2136
- "loss": 0.8912,
2137
- "step": 3440
2138
- },
2139
- {
2140
- "epoch": 1.15,
2141
- "learning_rate": 7.853657951945548e-05,
2142
- "loss": 0.8925,
2143
- "step": 3450
2144
- },
2145
- {
2146
- "epoch": 1.16,
2147
- "learning_rate": 7.799517935343757e-05,
2148
- "loss": 0.8751,
2149
- "step": 3460
2150
- },
2151
- {
2152
- "epoch": 1.16,
2153
- "learning_rate": 7.745445615828073e-05,
2154
- "loss": 0.9157,
2155
- "step": 3470
2156
- },
2157
- {
2158
- "epoch": 1.16,
2159
- "learning_rate": 7.69144265691512e-05,
2160
- "loss": 0.8971,
2161
- "step": 3480
2162
- },
2163
- {
2164
- "epoch": 1.17,
2165
- "learning_rate": 7.637510719987653e-05,
2166
- "loss": 0.8905,
2167
- "step": 3490
2168
- },
2169
- {
2170
- "epoch": 1.17,
2171
- "learning_rate": 7.583651464243458e-05,
2172
- "loss": 0.8786,
2173
- "step": 3500
2174
- },
2175
- {
2176
- "epoch": 1.17,
2177
- "learning_rate": 7.529866546644315e-05,
2178
- "loss": 0.8687,
2179
- "step": 3510
2180
- },
2181
- {
2182
- "epoch": 1.18,
2183
- "learning_rate": 7.476157621865009e-05,
2184
- "loss": 0.8842,
2185
- "step": 3520
2186
- },
2187
- {
2188
- "epoch": 1.18,
2189
- "learning_rate": 7.422526342242445e-05,
2190
- "loss": 0.9067,
2191
- "step": 3530
2192
- },
2193
- {
2194
- "epoch": 1.18,
2195
- "learning_rate": 7.368974357724789e-05,
2196
- "loss": 0.8755,
2197
- "step": 3540
2198
- },
2199
- {
2200
- "epoch": 1.19,
2201
- "learning_rate": 7.315503315820722e-05,
2202
- "loss": 0.8982,
2203
- "step": 3550
2204
- },
2205
- {
2206
- "epoch": 1.19,
2207
- "learning_rate": 7.262114861548759e-05,
2208
- "loss": 0.8743,
2209
- "step": 3560
2210
- },
2211
- {
2212
- "epoch": 1.19,
2213
- "learning_rate": 7.208810637386622e-05,
2214
- "loss": 0.8612,
2215
- "step": 3570
2216
- },
2217
- {
2218
- "epoch": 1.2,
2219
- "learning_rate": 7.155592283220738e-05,
2220
- "loss": 0.9104,
2221
- "step": 3580
2222
- },
2223
- {
2224
- "epoch": 1.2,
2225
- "learning_rate": 7.102461436295762e-05,
2226
- "loss": 0.8638,
2227
- "step": 3590
2228
- },
2229
- {
2230
- "epoch": 1.2,
2231
- "learning_rate": 7.049419731164221e-05,
2232
- "loss": 0.8811,
2233
- "step": 3600
2234
- },
2235
- {
2236
- "epoch": 1.21,
2237
- "learning_rate": 6.996468799636224e-05,
2238
- "loss": 0.8819,
2239
- "step": 3610
2240
- },
2241
- {
2242
- "epoch": 1.21,
2243
- "learning_rate": 6.94361027072926e-05,
2244
- "loss": 0.9108,
2245
- "step": 3620
2246
- },
2247
- {
2248
- "epoch": 1.21,
2249
- "learning_rate": 6.890845770618091e-05,
2250
- "loss": 0.8328,
2251
- "step": 3630
2252
- },
2253
- {
2254
- "epoch": 1.22,
2255
- "learning_rate": 6.838176922584705e-05,
2256
- "loss": 0.898,
2257
- "step": 3640
2258
- },
2259
- {
2260
- "epoch": 1.22,
2261
- "learning_rate": 6.785605346968386e-05,
2262
- "loss": 0.8752,
2263
- "step": 3650
2264
- },
2265
- {
2266
- "epoch": 1.22,
2267
- "learning_rate": 6.73313266111587e-05,
2268
- "loss": 0.8823,
2269
- "step": 3660
2270
- },
2271
- {
2272
- "epoch": 1.23,
2273
- "learning_rate": 6.680760479331585e-05,
2274
- "loss": 0.8907,
2275
- "step": 3670
2276
- },
2277
- {
2278
- "epoch": 1.23,
2279
- "learning_rate": 6.62849041282798e-05,
2280
- "loss": 0.9016,
2281
- "step": 3680
2282
- },
2283
- {
2284
- "epoch": 1.23,
2285
- "learning_rate": 6.576324069675963e-05,
2286
- "loss": 0.8745,
2287
- "step": 3690
2288
- },
2289
- {
2290
- "epoch": 1.24,
2291
- "learning_rate": 6.52426305475543e-05,
2292
- "loss": 0.8954,
2293
- "step": 3700
2294
- },
2295
- {
2296
- "epoch": 1.24,
2297
- "learning_rate": 6.472308969705882e-05,
2298
- "loss": 0.8873,
2299
- "step": 3710
2300
- },
2301
- {
2302
- "epoch": 1.24,
2303
- "learning_rate": 6.420463412877176e-05,
2304
- "loss": 0.8926,
2305
- "step": 3720
2306
- },
2307
- {
2308
- "epoch": 1.25,
2309
- "learning_rate": 6.368727979280317e-05,
2310
- "loss": 0.8842,
2311
- "step": 3730
2312
- },
2313
- {
2314
- "epoch": 1.25,
2315
- "learning_rate": 6.317104260538414e-05,
2316
- "loss": 0.8959,
2317
- "step": 3740
2318
- },
2319
- {
2320
- "epoch": 1.26,
2321
- "learning_rate": 6.265593844837709e-05,
2322
- "loss": 0.8738,
2323
- "step": 3750
2324
- },
2325
- {
2326
- "epoch": 1.26,
2327
- "learning_rate": 6.214198316878707e-05,
2328
- "loss": 0.8721,
2329
- "step": 3760
2330
- },
2331
- {
2332
- "epoch": 1.26,
2333
- "learning_rate": 6.162919257827435e-05,
2334
- "loss": 0.8922,
2335
- "step": 3770
2336
- },
2337
- {
2338
- "epoch": 1.27,
2339
- "learning_rate": 6.111758245266794e-05,
2340
- "loss": 0.8858,
2341
- "step": 3780
2342
- },
2343
- {
2344
- "epoch": 1.27,
2345
- "learning_rate": 6.060716853148024e-05,
2346
- "loss": 0.8973,
2347
- "step": 3790
2348
- },
2349
- {
2350
- "epoch": 1.27,
2351
- "learning_rate": 6.009796651742281e-05,
2352
- "loss": 0.896,
2353
- "step": 3800
2354
- },
2355
- {
2356
- "epoch": 1.28,
2357
- "learning_rate": 5.9589992075923265e-05,
2358
- "loss": 0.8763,
2359
- "step": 3810
2360
- },
2361
- {
2362
- "epoch": 1.28,
2363
- "learning_rate": 5.9083260834643486e-05,
2364
- "loss": 0.892,
2365
- "step": 3820
2366
- },
2367
- {
2368
- "epoch": 1.28,
2369
- "learning_rate": 5.8577788382998585e-05,
2370
- "loss": 0.8727,
2371
- "step": 3830
2372
- },
2373
- {
2374
- "epoch": 1.29,
2375
- "learning_rate": 5.8073590271677536e-05,
2376
- "loss": 0.8751,
2377
- "step": 3840
2378
- },
2379
- {
2380
- "epoch": 1.29,
2381
- "learning_rate": 5.757068201216456e-05,
2382
- "loss": 0.869,
2383
- "step": 3850
2384
- },
2385
- {
2386
- "epoch": 1.29,
2387
- "learning_rate": 5.706907907626215e-05,
2388
- "loss": 0.8727,
2389
- "step": 3860
2390
- },
2391
- {
2392
- "epoch": 1.3,
2393
- "learning_rate": 5.656879689561487e-05,
2394
- "loss": 0.8937,
2395
- "step": 3870
2396
- },
2397
- {
2398
- "epoch": 1.3,
2399
- "learning_rate": 5.606985086123473e-05,
2400
- "loss": 0.87,
2401
- "step": 3880
2402
- },
2403
- {
2404
- "epoch": 1.3,
2405
- "learning_rate": 5.5572256323027625e-05,
2406
- "loss": 0.8837,
2407
- "step": 3890
2408
- },
2409
- {
2410
- "epoch": 1.31,
2411
- "learning_rate": 5.507602858932113e-05,
2412
- "loss": 0.8804,
2413
- "step": 3900
2414
- },
2415
- {
2416
- "epoch": 1.31,
2417
- "learning_rate": 5.45811829263936e-05,
2418
- "loss": 0.9172,
2419
- "step": 3910
2420
- },
2421
- {
2422
- "epoch": 1.31,
2423
- "learning_rate": 5.40877345580044e-05,
2424
- "loss": 0.8849,
2425
- "step": 3920
2426
- },
2427
- {
2428
- "epoch": 1.32,
2429
- "eval_loss": 1.1395624876022339,
2430
- "eval_runtime": 2.5544,
2431
- "eval_samples_per_second": 950.915,
2432
- "eval_steps_per_second": 119.011,
2433
- "step": 3928
2434
- },
2435
- {
2436
- "epoch": 1.32,
2437
- "learning_rate": 5.3595698664925573e-05,
2438
- "loss": 0.8883,
2439
- "step": 3930
2440
- },
2441
- {
2442
- "epoch": 1.32,
2443
- "learning_rate": 5.310509038447492e-05,
2444
- "loss": 0.8841,
2445
- "step": 3940
2446
- },
2447
- {
2448
- "epoch": 1.32,
2449
- "learning_rate": 5.261592481005011e-05,
2450
- "loss": 0.8565,
2451
- "step": 3950
2452
- },
2453
- {
2454
- "epoch": 1.33,
2455
- "learning_rate": 5.212821699066459e-05,
2456
- "loss": 0.8722,
2457
- "step": 3960
2458
- },
2459
- {
2460
- "epoch": 1.33,
2461
- "learning_rate": 5.1641981930484374e-05,
2462
- "loss": 0.9014,
2463
- "step": 3970
2464
- },
2465
- {
2466
- "epoch": 1.33,
2467
- "learning_rate": 5.115723458836652e-05,
2468
- "loss": 0.8695,
2469
- "step": 3980
2470
- },
2471
- {
2472
- "epoch": 1.34,
2473
- "learning_rate": 5.067398987739901e-05,
2474
- "loss": 0.8876,
2475
- "step": 3990
2476
- },
2477
- {
2478
- "epoch": 1.34,
2479
- "learning_rate": 5.01922626644418e-05,
2480
- "loss": 0.8721,
2481
- "step": 4000
2482
- },
2483
- {
2484
- "epoch": 1.34,
2485
- "learning_rate": 4.9712067769669665e-05,
2486
- "loss": 0.867,
2487
- "step": 4010
2488
- },
2489
- {
2490
- "epoch": 1.35,
2491
- "learning_rate": 4.9233419966116036e-05,
2492
- "loss": 0.8673,
2493
- "step": 4020
2494
- },
2495
- {
2496
- "epoch": 1.35,
2497
- "learning_rate": 4.875633397921855e-05,
2498
- "loss": 0.8854,
2499
- "step": 4030
2500
- },
2501
- {
2502
- "epoch": 1.35,
2503
- "learning_rate": 4.828082448636626e-05,
2504
- "loss": 0.8818,
2505
- "step": 4040
2506
- },
2507
- {
2508
- "epoch": 1.36,
2509
- "learning_rate": 4.780690611644776e-05,
2510
- "loss": 0.8916,
2511
- "step": 4050
2512
- },
2513
- {
2514
- "epoch": 1.36,
2515
- "learning_rate": 4.73345934494013e-05,
2516
- "loss": 0.8962,
2517
- "step": 4060
2518
- },
2519
- {
2520
- "epoch": 1.36,
2521
- "learning_rate": 4.686390101576633e-05,
2522
- "loss": 0.9049,
2523
- "step": 4070
2524
- },
2525
- {
2526
- "epoch": 1.37,
2527
- "learning_rate": 4.639484329623627e-05,
2528
- "loss": 0.868,
2529
- "step": 4080
2530
- },
2531
- {
2532
- "epoch": 1.37,
2533
- "learning_rate": 4.592743472121309e-05,
2534
- "loss": 0.8673,
2535
- "step": 4090
2536
- },
2537
- {
2538
- "epoch": 1.37,
2539
- "learning_rate": 4.546168967036351e-05,
2540
- "loss": 0.8602,
2541
- "step": 4100
2542
- },
2543
- {
2544
- "epoch": 1.38,
2545
- "learning_rate": 4.499762247217639e-05,
2546
- "loss": 0.8695,
2547
- "step": 4110
2548
- },
2549
- {
2550
- "epoch": 1.38,
2551
- "learning_rate": 4.453524740352206e-05,
2552
- "loss": 0.8698,
2553
- "step": 4120
2554
- },
2555
- {
2556
- "epoch": 1.38,
2557
- "learning_rate": 4.407457868921303e-05,
2558
- "loss": 0.9047,
2559
- "step": 4130
2560
- },
2561
- {
2562
- "epoch": 1.39,
2563
- "learning_rate": 4.3615630501566384e-05,
2564
- "loss": 0.8802,
2565
- "step": 4140
2566
- },
2567
- {
2568
- "epoch": 1.39,
2569
- "learning_rate": 4.3158416959967884e-05,
2570
- "loss": 0.8553,
2571
- "step": 4150
2572
- },
2573
- {
2574
- "epoch": 1.39,
2575
- "learning_rate": 4.27029521304374e-05,
2576
- "loss": 0.8767,
2577
- "step": 4160
2578
- },
2579
- {
2580
- "epoch": 1.4,
2581
- "learning_rate": 4.2249250025196274e-05,
2582
- "loss": 0.8774,
2583
- "step": 4170
2584
- },
2585
- {
2586
- "epoch": 1.4,
2587
- "learning_rate": 4.179732460223626e-05,
2588
- "loss": 0.885,
2589
- "step": 4180
2590
- },
2591
- {
2592
- "epoch": 1.4,
2593
- "learning_rate": 4.1347189764890016e-05,
2594
- "loss": 0.8937,
2595
- "step": 4190
2596
- },
2597
- {
2598
- "epoch": 1.41,
2599
- "learning_rate": 4.08988593614036e-05,
2600
- "loss": 0.9031,
2601
- "step": 4200
2602
- },
2603
- {
2604
- "epoch": 1.41,
2605
- "learning_rate": 4.045234718451011e-05,
2606
- "loss": 0.8615,
2607
- "step": 4210
2608
- },
2609
- {
2610
- "epoch": 1.41,
2611
- "learning_rate": 4.000766697100558e-05,
2612
- "loss": 0.8887,
2613
- "step": 4220
2614
- },
2615
- {
2616
- "epoch": 1.42,
2617
- "learning_rate": 3.95648324013263e-05,
2618
- "loss": 0.892,
2619
- "step": 4230
2620
- },
2621
- {
2622
- "epoch": 1.42,
2623
- "learning_rate": 3.9123857099127936e-05,
2624
- "loss": 0.8536,
2625
- "step": 4240
2626
- },
2627
- {
2628
- "epoch": 1.42,
2629
- "learning_rate": 3.868475463086651e-05,
2630
- "loss": 0.8937,
2631
- "step": 4250
2632
- },
2633
- {
2634
- "epoch": 1.43,
2635
- "learning_rate": 3.824753850538082e-05,
2636
- "loss": 0.8575,
2637
- "step": 4260
2638
- },
2639
- {
2640
- "epoch": 1.43,
2641
- "learning_rate": 3.781222217347704e-05,
2642
- "loss": 0.8774,
2643
- "step": 4270
2644
- },
2645
- {
2646
- "epoch": 1.43,
2647
- "learning_rate": 3.737881902751479e-05,
2648
- "loss": 0.8906,
2649
- "step": 4280
2650
- },
2651
- {
2652
- "epoch": 1.44,
2653
- "learning_rate": 3.694734240099524e-05,
2654
- "loss": 0.8901,
2655
- "step": 4290
2656
- },
2657
- {
2658
- "epoch": 1.44,
2659
- "learning_rate": 3.651780556815079e-05,
2660
- "loss": 0.8603,
2661
- "step": 4300
2662
- },
2663
- {
2664
- "epoch": 1.45,
2665
- "learning_rate": 3.609022174353674e-05,
2666
- "loss": 0.8865,
2667
- "step": 4310
2668
- },
2669
- {
2670
- "epoch": 1.45,
2671
- "learning_rate": 3.5664604081624786e-05,
2672
- "loss": 0.864,
2673
- "step": 4320
2674
- },
2675
- {
2676
- "epoch": 1.45,
2677
- "learning_rate": 3.5240965676398215e-05,
2678
- "loss": 0.8863,
2679
- "step": 4330
2680
- },
2681
- {
2682
- "epoch": 1.46,
2683
- "learning_rate": 3.48193195609493e-05,
2684
- "loss": 0.8841,
2685
- "step": 4340
2686
- },
2687
- {
2688
- "epoch": 1.46,
2689
- "learning_rate": 3.439967870707807e-05,
2690
- "loss": 0.8755,
2691
- "step": 4350
2692
- },
2693
- {
2694
- "epoch": 1.46,
2695
- "learning_rate": 3.398205602489337e-05,
2696
- "loss": 0.8844,
2697
- "step": 4360
2698
- },
2699
- {
2700
- "epoch": 1.47,
2701
- "learning_rate": 3.356646436241573e-05,
2702
- "loss": 0.9026,
2703
- "step": 4370
2704
- },
2705
- {
2706
- "epoch": 1.47,
2707
- "learning_rate": 3.315291650518197e-05,
2708
- "loss": 0.877,
2709
- "step": 4380
2710
- },
2711
- {
2712
- "epoch": 1.47,
2713
- "learning_rate": 3.2741425175852056e-05,
2714
- "loss": 0.8969,
2715
- "step": 4390
2716
- },
2717
- {
2718
- "epoch": 1.48,
2719
- "learning_rate": 3.233200303381746e-05,
2720
- "loss": 0.8719,
2721
- "step": 4400
2722
- },
2723
- {
2724
- "epoch": 1.48,
2725
- "learning_rate": 3.1924662674811836e-05,
2726
- "loss": 0.8522,
2727
- "step": 4410
2728
- },
2729
- {
2730
- "epoch": 1.48,
2731
- "eval_loss": 1.1344659328460693,
2732
- "eval_runtime": 2.6026,
2733
- "eval_samples_per_second": 933.308,
2734
- "eval_steps_per_second": 116.808,
2735
- "step": 4419
2736
- },
2737
- {
2738
- "epoch": 1.48,
2739
- "learning_rate": 3.1519416630523444e-05,
2740
- "loss": 0.8861,
2741
- "step": 4420
2742
- },
2743
- {
2744
- "epoch": 1.49,
2745
- "learning_rate": 3.111627736820979e-05,
2746
- "loss": 0.8813,
2747
- "step": 4430
2748
- },
2749
- {
2750
- "epoch": 1.49,
2751
- "learning_rate": 3.071525729031384e-05,
2752
- "loss": 0.8727,
2753
- "step": 4440
2754
- },
2755
- {
2756
- "epoch": 1.49,
2757
- "learning_rate": 3.0316368734082533e-05,
2758
- "loss": 0.8966,
2759
- "step": 4450
2760
- },
2761
- {
2762
- "epoch": 1.5,
2763
- "learning_rate": 2.9919623971187415e-05,
2764
- "loss": 0.8636,
2765
- "step": 4460
2766
- },
2767
- {
2768
- "epoch": 1.5,
2769
- "learning_rate": 2.952503520734682e-05,
2770
- "loss": 0.8779,
2771
- "step": 4470
2772
- },
2773
- {
2774
- "epoch": 1.5,
2775
- "learning_rate": 2.913261458195051e-05,
2776
- "loss": 0.8666,
2777
- "step": 4480
2778
- },
2779
- {
2780
- "epoch": 1.51,
2781
- "learning_rate": 2.8742374167686313e-05,
2782
- "loss": 0.8625,
2783
- "step": 4490
2784
- },
2785
- {
2786
- "epoch": 1.51,
2787
- "learning_rate": 2.8354325970168484e-05,
2788
- "loss": 0.8835,
2789
- "step": 4500
2790
- },
2791
- {
2792
- "epoch": 1.51,
2793
- "learning_rate": 2.7968481927568514e-05,
2794
- "loss": 0.8778,
2795
- "step": 4510
2796
- },
2797
- {
2798
- "epoch": 1.52,
2799
- "learning_rate": 2.7584853910247767e-05,
2800
- "loss": 0.8682,
2801
- "step": 4520
2802
- },
2803
- {
2804
- "epoch": 1.52,
2805
- "learning_rate": 2.7203453720392458e-05,
2806
- "loss": 0.858,
2807
- "step": 4530
2808
- },
2809
- {
2810
- "epoch": 1.52,
2811
- "learning_rate": 2.682429309165033e-05,
2812
- "loss": 0.8964,
2813
- "step": 4540
2814
- },
2815
- {
2816
- "epoch": 1.53,
2817
- "learning_rate": 2.6447383688769846e-05,
2818
- "loss": 0.8856,
2819
- "step": 4550
2820
- },
2821
- {
2822
- "epoch": 1.53,
2823
- "learning_rate": 2.6072737107241208e-05,
2824
- "loss": 0.8832,
2825
- "step": 4560
2826
- },
2827
- {
2828
- "epoch": 1.53,
2829
- "learning_rate": 2.5700364872939718e-05,
2830
- "loss": 0.8549,
2831
- "step": 4570
2832
- },
2833
- {
2834
- "epoch": 1.54,
2835
- "learning_rate": 2.5330278441771226e-05,
2836
- "loss": 0.836,
2837
- "step": 4580
2838
- },
2839
- {
2840
- "epoch": 1.54,
2841
- "learning_rate": 2.4962489199319504e-05,
2842
- "loss": 0.8541,
2843
- "step": 4590
2844
- },
2845
- {
2846
- "epoch": 1.54,
2847
- "learning_rate": 2.4597008460496164e-05,
2848
- "loss": 0.8889,
2849
- "step": 4600
2850
- },
2851
- {
2852
- "epoch": 1.55,
2853
- "learning_rate": 2.4233847469192438e-05,
2854
- "loss": 0.8512,
2855
- "step": 4610
2856
- },
2857
- {
2858
- "epoch": 1.55,
2859
- "learning_rate": 2.3873017397933327e-05,
2860
- "loss": 0.8525,
2861
- "step": 4620
2862
- },
2863
- {
2864
- "epoch": 1.55,
2865
- "learning_rate": 2.351452934753394e-05,
2866
- "loss": 0.8887,
2867
- "step": 4630
2868
- },
2869
- {
2870
- "epoch": 1.56,
2871
- "learning_rate": 2.3158394346757784e-05,
2872
- "loss": 0.8911,
2873
- "step": 4640
2874
- },
2875
- {
2876
- "epoch": 1.56,
2877
- "learning_rate": 2.280462335197767e-05,
2878
- "loss": 0.905,
2879
- "step": 4650
2880
- },
2881
- {
2882
- "epoch": 1.56,
2883
- "learning_rate": 2.2453227246838538e-05,
2884
- "loss": 0.8785,
2885
- "step": 4660
2886
- },
2887
- {
2888
- "epoch": 1.57,
2889
- "learning_rate": 2.2104216841922633e-05,
2890
- "loss": 0.8762,
2891
- "step": 4670
2892
- },
2893
- {
2894
- "epoch": 1.57,
2895
- "learning_rate": 2.1757602874416993e-05,
2896
- "loss": 0.8723,
2897
- "step": 4680
2898
- },
2899
- {
2900
- "epoch": 1.57,
2901
- "learning_rate": 2.141339600778305e-05,
2902
- "loss": 0.8541,
2903
- "step": 4690
2904
- },
2905
- {
2906
- "epoch": 1.58,
2907
- "learning_rate": 2.1071606831428557e-05,
2908
- "loss": 0.8762,
2909
- "step": 4700
2910
- },
2911
- {
2912
- "epoch": 1.58,
2913
- "learning_rate": 2.0732245860381872e-05,
2914
- "loss": 0.8654,
2915
- "step": 4710
2916
- },
2917
- {
2918
- "epoch": 1.58,
2919
- "learning_rate": 2.0395323534968412e-05,
2920
- "loss": 0.8362,
2921
- "step": 4720
2922
- },
2923
- {
2924
- "epoch": 1.59,
2925
- "learning_rate": 2.0060850220489567e-05,
2926
- "loss": 0.8711,
2927
- "step": 4730
2928
- },
2929
- {
2930
- "epoch": 1.59,
2931
- "learning_rate": 1.9728836206903656e-05,
2932
- "loss": 0.8564,
2933
- "step": 4740
2934
- },
2935
- {
2936
- "epoch": 1.59,
2937
- "learning_rate": 1.9399291708509438e-05,
2938
- "loss": 0.8733,
2939
- "step": 4750
2940
- },
2941
- {
2942
- "epoch": 1.6,
2943
- "learning_rate": 1.907222686363188e-05,
2944
- "loss": 0.8519,
2945
- "step": 4760
2946
- },
2947
- {
2948
- "epoch": 1.6,
2949
- "learning_rate": 1.8747651734310314e-05,
2950
- "loss": 0.8923,
2951
- "step": 4770
2952
- },
2953
- {
2954
- "epoch": 1.6,
2955
- "learning_rate": 1.8425576305988734e-05,
2956
- "loss": 0.8662,
2957
- "step": 4780
2958
- },
2959
- {
2960
- "epoch": 1.61,
2961
- "learning_rate": 1.8106010487208713e-05,
2962
- "loss": 0.856,
2963
- "step": 4790
2964
- },
2965
- {
2966
- "epoch": 1.61,
2967
- "learning_rate": 1.7788964109304496e-05,
2968
- "loss": 0.8734,
2969
- "step": 4800
2970
- },
2971
- {
2972
- "epoch": 1.61,
2973
- "learning_rate": 1.7474446926100673e-05,
2974
- "loss": 0.8542,
2975
- "step": 4810
2976
- },
2977
- {
2978
- "epoch": 1.62,
2979
- "learning_rate": 1.7162468613611937e-05,
2980
- "loss": 0.8638,
2981
- "step": 4820
2982
- },
2983
- {
2984
- "epoch": 1.62,
2985
- "learning_rate": 1.6853038769745467e-05,
2986
- "loss": 0.8949,
2987
- "step": 4830
2988
- },
2989
- {
2990
- "epoch": 1.63,
2991
- "learning_rate": 1.6546166914005767e-05,
2992
- "loss": 0.861,
2993
- "step": 4840
2994
- },
2995
- {
2996
- "epoch": 1.63,
2997
- "learning_rate": 1.62418624872016e-05,
2998
- "loss": 0.8437,
2999
- "step": 4850
3000
- },
3001
- {
3002
- "epoch": 1.63,
3003
- "learning_rate": 1.5940134851155697e-05,
3004
- "loss": 0.8584,
3005
- "step": 4860
3006
- },
3007
- {
3008
- "epoch": 1.64,
3009
- "learning_rate": 1.564099328841665e-05,
3010
- "loss": 0.8736,
3011
- "step": 4870
3012
- },
3013
- {
3014
- "epoch": 1.64,
3015
- "learning_rate": 1.534444700197346e-05,
3016
- "loss": 0.8591,
3017
- "step": 4880
3018
- },
3019
- {
3020
- "epoch": 1.64,
3021
- "learning_rate": 1.505050511497227e-05,
3022
- "loss": 0.8684,
3023
- "step": 4890
3024
- },
3025
- {
3026
- "epoch": 1.65,
3027
- "learning_rate": 1.4759176670435748e-05,
3028
- "loss": 0.9093,
3029
- "step": 4900
3030
- },
3031
- {
3032
- "epoch": 1.65,
3033
- "learning_rate": 1.4470470630984911e-05,
3034
- "loss": 0.871,
3035
- "step": 4910
3036
- },
3037
- {
3038
- "epoch": 1.65,
3039
- "eval_loss": 1.1315988302230835,
3040
- "eval_runtime": 2.6531,
3041
- "eval_samples_per_second": 915.533,
3042
- "eval_steps_per_second": 114.583,
3043
- "step": 4910
3044
- },
3045
- {
3046
- "epoch": 1.65,
3047
- "learning_rate": 1.4184395878563395e-05,
3048
- "loss": 0.8653,
3049
- "step": 4920
3050
- },
3051
- {
3052
- "epoch": 1.66,
3053
- "learning_rate": 1.3900961214164144e-05,
3054
- "loss": 0.8912,
3055
- "step": 4930
3056
- },
3057
- {
3058
- "epoch": 1.66,
3059
- "learning_rate": 1.3620175357558706e-05,
3060
- "loss": 0.8725,
3061
- "step": 4940
3062
- },
3063
- {
3064
- "epoch": 1.66,
3065
- "learning_rate": 1.3342046947028896e-05,
3066
- "loss": 0.9122,
3067
- "step": 4950
3068
- },
3069
- {
3070
- "epoch": 1.67,
3071
- "learning_rate": 1.3066584539101223e-05,
3072
- "loss": 0.8874,
3073
- "step": 4960
3074
- },
3075
- {
3076
- "epoch": 1.67,
3077
- "learning_rate": 1.2793796608283426e-05,
3078
- "loss": 0.8628,
3079
- "step": 4970
3080
- },
3081
- {
3082
- "epoch": 1.67,
3083
- "learning_rate": 1.2523691546803873e-05,
3084
- "loss": 0.8751,
3085
- "step": 4980
3086
- },
3087
- {
3088
- "epoch": 1.68,
3089
- "learning_rate": 1.2256277664353411e-05,
3090
- "loss": 0.8485,
3091
- "step": 4990
3092
- },
3093
- {
3094
- "epoch": 1.68,
3095
- "learning_rate": 1.1991563187829591e-05,
3096
- "loss": 0.8741,
3097
- "step": 5000
3098
- },
3099
- {
3100
- "epoch": 1.68,
3101
- "learning_rate": 1.1729556261083784e-05,
3102
- "loss": 0.8194,
3103
- "step": 5010
3104
- },
3105
- {
3106
- "epoch": 1.69,
3107
- "learning_rate": 1.1470264944670383e-05,
3108
- "loss": 0.8649,
3109
- "step": 5020
3110
- },
3111
- {
3112
- "epoch": 1.69,
3113
- "learning_rate": 1.1213697215599017e-05,
3114
- "loss": 0.887,
3115
- "step": 5030
3116
- },
3117
- {
3118
- "epoch": 1.69,
3119
- "learning_rate": 1.0959860967089041e-05,
3120
- "loss": 0.8359,
3121
- "step": 5040
3122
- },
3123
- {
3124
- "epoch": 1.7,
3125
- "learning_rate": 1.0708764008326733e-05,
3126
- "loss": 0.8837,
3127
- "step": 5050
3128
- },
3129
- {
3130
- "epoch": 1.7,
3131
- "learning_rate": 1.04604140642251e-05,
3132
- "loss": 0.8914,
3133
- "step": 5060
3134
- },
3135
- {
3136
- "epoch": 1.7,
3137
- "learning_rate": 1.0214818775186108e-05,
3138
- "loss": 0.9028,
3139
- "step": 5070
3140
- },
3141
- {
3142
- "epoch": 1.71,
3143
- "learning_rate": 9.971985696865738e-06,
3144
- "loss": 0.8887,
3145
- "step": 5080
3146
- },
3147
- {
3148
- "epoch": 1.71,
3149
- "learning_rate": 9.731922299941454e-06,
3150
- "loss": 0.8613,
3151
- "step": 5090
3152
- },
3153
- {
3154
- "epoch": 1.71,
3155
- "learning_rate": 9.494635969882426e-06,
3156
- "loss": 0.8656,
3157
- "step": 5100
3158
- },
3159
- {
3160
- "epoch": 1.72,
3161
- "learning_rate": 9.260134006722332e-06,
3162
- "loss": 0.879,
3163
- "step": 5110
3164
- },
3165
- {
3166
- "epoch": 1.72,
3167
- "learning_rate": 9.028423624834704e-06,
3168
- "loss": 0.862,
3169
- "step": 5120
3170
- },
3171
- {
3172
- "epoch": 1.72,
3173
- "learning_rate": 8.799511952711025e-06,
3174
- "loss": 0.8987,
3175
- "step": 5130
3176
- },
3177
- {
3178
- "epoch": 1.73,
3179
- "learning_rate": 8.573406032741438e-06,
3180
- "loss": 0.8483,
3181
- "step": 5140
3182
- },
3183
- {
3184
- "epoch": 1.73,
3185
- "learning_rate": 8.350112820998035e-06,
3186
- "loss": 0.8875,
3187
- "step": 5150
3188
- },
3189
- {
3190
- "epoch": 1.73,
3191
- "learning_rate": 8.129639187020954e-06,
3192
- "loss": 0.8512,
3193
- "step": 5160
3194
- },
3195
- {
3196
- "epoch": 1.74,
3197
- "learning_rate": 7.911991913606932e-06,
3198
- "loss": 0.8735,
3199
- "step": 5170
3200
- },
3201
- {
3202
- "epoch": 1.74,
3203
- "learning_rate": 7.697177696600677e-06,
3204
- "loss": 0.8544,
3205
- "step": 5180
3206
- },
3207
- {
3208
- "epoch": 1.74,
3209
- "learning_rate": 7.4852031446888685e-06,
3210
- "loss": 0.8834,
3211
- "step": 5190
3212
- },
3213
- {
3214
- "epoch": 1.75,
3215
- "learning_rate": 7.276074779196906e-06,
3216
- "loss": 0.8757,
3217
- "step": 5200
3218
- },
3219
- {
3220
- "epoch": 1.75,
3221
- "learning_rate": 7.06979903388818e-06,
3222
- "loss": 0.8711,
3223
- "step": 5210
3224
- },
3225
- {
3226
- "epoch": 1.75,
3227
- "learning_rate": 6.866382254766157e-06,
3228
- "loss": 0.869,
3229
- "step": 5220
3230
- },
3231
- {
3232
- "epoch": 1.76,
3233
- "learning_rate": 6.665830699879261e-06,
3234
- "loss": 0.8468,
3235
- "step": 5230
3236
- },
3237
- {
3238
- "epoch": 1.76,
3239
- "learning_rate": 6.468150539128193e-06,
3240
- "loss": 0.8703,
3241
- "step": 5240
3242
- },
3243
- {
3244
- "epoch": 1.76,
3245
- "learning_rate": 6.273347854076195e-06,
3246
- "loss": 0.8728,
3247
- "step": 5250
3248
- },
3249
- {
3250
- "epoch": 1.77,
3251
- "learning_rate": 6.081428637761977e-06,
3252
- "loss": 0.8635,
3253
- "step": 5260
3254
- },
3255
- {
3256
- "epoch": 1.77,
3257
- "learning_rate": 5.892398794515286e-06,
3258
- "loss": 0.8632,
3259
- "step": 5270
3260
- },
3261
- {
3262
- "epoch": 1.77,
3263
- "learning_rate": 5.706264139775286e-06,
3264
- "loss": 0.8723,
3265
- "step": 5280
3266
- },
3267
- {
3268
- "epoch": 1.78,
3269
- "learning_rate": 5.523030399911611e-06,
3270
- "loss": 0.8804,
3271
- "step": 5290
3272
- },
3273
- {
3274
- "epoch": 1.78,
3275
- "learning_rate": 5.342703212048305e-06,
3276
- "loss": 0.855,
3277
- "step": 5300
3278
- },
3279
- {
3280
- "epoch": 1.78,
3281
- "learning_rate": 5.165288123890277e-06,
3282
- "loss": 0.8741,
3283
- "step": 5310
3284
- }
3285
- ],
3286
- "logging_steps": 10,
3287
- "max_steps": 5892,
3288
- "num_input_tokens_seen": 0,
3289
- "num_train_epochs": 2,
3290
- "save_steps": 590,
3291
- "total_flos": 1.865860038912901e+17,
3292
- "train_batch_size": 4,
3293
- "trial_name": null,
3294
- "trial_params": null
3295
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
checkpoint-5310/training_args.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:4af80da8f1766695fd422eeecbe8550d1bedd1fd1e544d9b291755c3ff7b7de4
3
- size 5304