smithblack-0 commited on
Commit
da1e28b
·
verified ·
1 Parent(s): 7078c3e

Upload folder using huggingface_hub

Browse files
control_42/epoch1/metadata.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "epoch_num": 1,
3
+ "global_batch_num": 268,
4
+ "device": "cuda",
5
+ "dtype": "bfloat16"
6
+ }
control_42/epoch1/metrics.json ADDED
@@ -0,0 +1,844 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "training": {
3
+ "effective_batch_nums": [
4
+ 1,
5
+ 2,
6
+ 3,
7
+ 4,
8
+ 5,
9
+ 6,
10
+ 7,
11
+ 8,
12
+ 9,
13
+ 10,
14
+ 11,
15
+ 12,
16
+ 13,
17
+ 14,
18
+ 15,
19
+ 16,
20
+ 17,
21
+ 18,
22
+ 19,
23
+ 20,
24
+ 21,
25
+ 22,
26
+ 23,
27
+ 24,
28
+ 25,
29
+ 26,
30
+ 27,
31
+ 28,
32
+ 29,
33
+ 30,
34
+ 31,
35
+ 32,
36
+ 33,
37
+ 34,
38
+ 35,
39
+ 36,
40
+ 37,
41
+ 38,
42
+ 39,
43
+ 40,
44
+ 41,
45
+ 42,
46
+ 43,
47
+ 44,
48
+ 45,
49
+ 46,
50
+ 47,
51
+ 48,
52
+ 49,
53
+ 50,
54
+ 51,
55
+ 52,
56
+ 53,
57
+ 54,
58
+ 55,
59
+ 56,
60
+ 57,
61
+ 58,
62
+ 59,
63
+ 60,
64
+ 61,
65
+ 62,
66
+ 63,
67
+ 64,
68
+ 65,
69
+ 66,
70
+ 67,
71
+ 68,
72
+ 69,
73
+ 70,
74
+ 71,
75
+ 72,
76
+ 73,
77
+ 74,
78
+ 75,
79
+ 76,
80
+ 77,
81
+ 78,
82
+ 79,
83
+ 80,
84
+ 81,
85
+ 82,
86
+ 83,
87
+ 84,
88
+ 85,
89
+ 86,
90
+ 87,
91
+ 88,
92
+ 89,
93
+ 90,
94
+ 91,
95
+ 92,
96
+ 93,
97
+ 94,
98
+ 95,
99
+ 96,
100
+ 97,
101
+ 98,
102
+ 99,
103
+ 100,
104
+ 101,
105
+ 102,
106
+ 103,
107
+ 104,
108
+ 105,
109
+ 106,
110
+ 107,
111
+ 108,
112
+ 109,
113
+ 110,
114
+ 111,
115
+ 112,
116
+ 113,
117
+ 114,
118
+ 115,
119
+ 116,
120
+ 117,
121
+ 118,
122
+ 119,
123
+ 120,
124
+ 121,
125
+ 122,
126
+ 123,
127
+ 124,
128
+ 125,
129
+ 126,
130
+ 127,
131
+ 128,
132
+ 129,
133
+ 130,
134
+ 131,
135
+ 132,
136
+ 133,
137
+ 134,
138
+ 135,
139
+ 136,
140
+ 137,
141
+ 138,
142
+ 139,
143
+ 140,
144
+ 141,
145
+ 142,
146
+ 143,
147
+ 144,
148
+ 145,
149
+ 146,
150
+ 147,
151
+ 148,
152
+ 149,
153
+ 150,
154
+ 151,
155
+ 152,
156
+ 153,
157
+ 154,
158
+ 155,
159
+ 156,
160
+ 157,
161
+ 158,
162
+ 159,
163
+ 160,
164
+ 161,
165
+ 162,
166
+ 163,
167
+ 164,
168
+ 165,
169
+ 166,
170
+ 167,
171
+ 168,
172
+ 169,
173
+ 170,
174
+ 171,
175
+ 172,
176
+ 173,
177
+ 174,
178
+ 175,
179
+ 176,
180
+ 177,
181
+ 178,
182
+ 179,
183
+ 180,
184
+ 181,
185
+ 182,
186
+ 183,
187
+ 184,
188
+ 185,
189
+ 186,
190
+ 187,
191
+ 188,
192
+ 189,
193
+ 190,
194
+ 191,
195
+ 192,
196
+ 193,
197
+ 194,
198
+ 195,
199
+ 196,
200
+ 197,
201
+ 198,
202
+ 199,
203
+ 200,
204
+ 201,
205
+ 202,
206
+ 203,
207
+ 204,
208
+ 205,
209
+ 206,
210
+ 207,
211
+ 208,
212
+ 209,
213
+ 210,
214
+ 211,
215
+ 212,
216
+ 213,
217
+ 214,
218
+ 215,
219
+ 216,
220
+ 217,
221
+ 218,
222
+ 219,
223
+ 220,
224
+ 221,
225
+ 222,
226
+ 223,
227
+ 224,
228
+ 225,
229
+ 226,
230
+ 227,
231
+ 228,
232
+ 229,
233
+ 230,
234
+ 231,
235
+ 232,
236
+ 233,
237
+ 234,
238
+ 235,
239
+ 236,
240
+ 237,
241
+ 238,
242
+ 239,
243
+ 240,
244
+ 241,
245
+ 242,
246
+ 243,
247
+ 244,
248
+ 245,
249
+ 246,
250
+ 247,
251
+ 248,
252
+ 249,
253
+ 250,
254
+ 251,
255
+ 252,
256
+ 253,
257
+ 254,
258
+ 255,
259
+ 256,
260
+ 257,
261
+ 258,
262
+ 259,
263
+ 260,
264
+ 261,
265
+ 262,
266
+ 263,
267
+ 264,
268
+ 265,
269
+ 266,
270
+ 267,
271
+ 268
272
+ ],
273
+ "losses": [
274
+ 117.0,
275
+ 44.0,
276
+ 41.25,
277
+ 37.25,
278
+ 30.75,
279
+ 25.625,
280
+ 20.875,
281
+ 17.625,
282
+ 14.5,
283
+ 12.625,
284
+ 11.25,
285
+ 10.5,
286
+ 10.0,
287
+ 9.6875,
288
+ 9.5,
289
+ 9.375,
290
+ 9.375,
291
+ 9.3125,
292
+ 9.25,
293
+ 9.1875,
294
+ 9.125,
295
+ 9.0,
296
+ 8.9375,
297
+ 8.8125,
298
+ 8.875,
299
+ 8.75,
300
+ 8.75,
301
+ 8.6875,
302
+ 8.625,
303
+ 8.6875,
304
+ 8.625,
305
+ 8.6875,
306
+ 8.625,
307
+ 8.625,
308
+ 8.625,
309
+ 8.5625,
310
+ 8.5625,
311
+ 8.5625,
312
+ 8.5625,
313
+ 8.5,
314
+ 8.5625,
315
+ 8.5625,
316
+ 8.5625,
317
+ 8.5625,
318
+ 8.5,
319
+ 8.5,
320
+ 8.5625,
321
+ 8.5,
322
+ 8.5625,
323
+ 8.5,
324
+ 8.5,
325
+ 8.4375,
326
+ 8.5,
327
+ 8.4375,
328
+ 8.5,
329
+ 8.4375,
330
+ 8.4375,
331
+ 8.5,
332
+ 8.5,
333
+ 8.5625,
334
+ 8.5,
335
+ 8.4375,
336
+ 8.4375,
337
+ 8.5,
338
+ 8.5,
339
+ 8.5,
340
+ 8.5,
341
+ 8.5,
342
+ 8.4375,
343
+ 8.4375,
344
+ 8.4375,
345
+ 8.5,
346
+ 8.5,
347
+ 8.4375,
348
+ 8.5,
349
+ 8.4375,
350
+ 8.5,
351
+ 8.5,
352
+ 8.4375,
353
+ 8.4375,
354
+ 8.4375,
355
+ 8.4375,
356
+ 8.4375,
357
+ 8.4375,
358
+ 8.5,
359
+ 8.5,
360
+ 8.4375,
361
+ 8.4375,
362
+ 8.4375,
363
+ 8.5,
364
+ 8.5,
365
+ 8.5,
366
+ 8.4375,
367
+ 8.4375,
368
+ 8.5,
369
+ 8.4375,
370
+ 8.5,
371
+ 8.5,
372
+ 8.5,
373
+ 8.4375,
374
+ 8.4375,
375
+ 8.5,
376
+ 8.4375,
377
+ 8.4375,
378
+ 8.5,
379
+ 8.4375,
380
+ 8.4375,
381
+ 8.5,
382
+ 8.4375,
383
+ 8.4375,
384
+ 8.4375,
385
+ 8.4375,
386
+ 8.4375,
387
+ 8.4375,
388
+ 8.4375,
389
+ 8.5,
390
+ 8.4375,
391
+ 8.4375,
392
+ 8.4375,
393
+ 8.4375,
394
+ 8.4375,
395
+ 8.4375,
396
+ 8.5,
397
+ 8.5,
398
+ 8.5,
399
+ 8.5,
400
+ 8.5,
401
+ 8.4375,
402
+ 8.4375,
403
+ 8.5,
404
+ 8.5,
405
+ 8.5,
406
+ 8.5,
407
+ 8.4375,
408
+ 8.4375,
409
+ 8.5,
410
+ 8.4375,
411
+ 8.5,
412
+ 8.4375,
413
+ 8.4375,
414
+ 8.4375,
415
+ 8.4375,
416
+ 8.5,
417
+ 8.4375,
418
+ 8.4375,
419
+ 8.4375,
420
+ 8.4375,
421
+ 8.4375,
422
+ 8.4375,
423
+ 8.4375,
424
+ 8.4375,
425
+ 8.5,
426
+ 8.4375,
427
+ 8.4375,
428
+ 8.4375,
429
+ 8.4375,
430
+ 8.5,
431
+ 8.4375,
432
+ 8.4375,
433
+ 8.4375,
434
+ 8.4375,
435
+ 8.4375,
436
+ 8.4375,
437
+ 8.4375,
438
+ 8.5,
439
+ 8.4375,
440
+ 8.4375,
441
+ 8.4375,
442
+ 8.4375,
443
+ 8.5,
444
+ 8.4375,
445
+ 8.4375,
446
+ 8.5,
447
+ 8.4375,
448
+ 8.4375,
449
+ 8.4375,
450
+ 8.4375,
451
+ 8.5,
452
+ 8.4375,
453
+ 8.5,
454
+ 8.4375,
455
+ 8.4375,
456
+ 8.4375,
457
+ 8.4375,
458
+ 8.4375,
459
+ 8.5,
460
+ 8.4375,
461
+ 8.4375,
462
+ 8.5,
463
+ 8.4375,
464
+ 8.4375,
465
+ 8.375,
466
+ 8.4375,
467
+ 8.4375,
468
+ 8.4375,
469
+ 8.4375,
470
+ 8.5,
471
+ 8.4375,
472
+ 8.4375,
473
+ 8.4375,
474
+ 8.5,
475
+ 8.4375,
476
+ 8.5,
477
+ 8.4375,
478
+ 8.4375,
479
+ 8.5,
480
+ 8.4375,
481
+ 8.4375,
482
+ 8.4375,
483
+ 8.4375,
484
+ 8.4375,
485
+ 8.4375,
486
+ 8.4375,
487
+ 8.4375,
488
+ 8.4375,
489
+ 8.5,
490
+ 8.4375,
491
+ 8.4375,
492
+ 8.4375,
493
+ 8.4375,
494
+ 8.4375,
495
+ 8.5,
496
+ 8.4375,
497
+ 8.4375,
498
+ 8.4375,
499
+ 8.4375,
500
+ 8.4375,
501
+ 8.4375,
502
+ 8.4375,
503
+ 8.4375,
504
+ 8.4375,
505
+ 8.4375,
506
+ 8.4375,
507
+ 8.4375,
508
+ 8.4375,
509
+ 8.4375,
510
+ 8.5,
511
+ 8.4375,
512
+ 8.375,
513
+ 8.4375,
514
+ 8.4375,
515
+ 8.5,
516
+ 8.4375,
517
+ 8.4375,
518
+ 8.5,
519
+ 8.4375,
520
+ 8.4375,
521
+ 8.4375,
522
+ 8.4375,
523
+ 8.5,
524
+ 8.4375,
525
+ 8.4375,
526
+ 8.4375,
527
+ 8.4375,
528
+ 8.4375,
529
+ 8.4375,
530
+ 8.4375,
531
+ 8.4375,
532
+ 8.4375,
533
+ 8.4375,
534
+ 8.4375,
535
+ 8.4375,
536
+ 8.4375,
537
+ 8.4375,
538
+ 8.375,
539
+ 8.5,
540
+ 8.5,
541
+ 8.5
542
+ ],
543
+ "grad_norms": [
544
+ 32.0,
545
+ 17.75,
546
+ 9.9375,
547
+ 14.0,
548
+ 8.25,
549
+ 6.90625,
550
+ 5.1875,
551
+ 3.75,
552
+ 2.75,
553
+ 1.8984375,
554
+ 1.6015625,
555
+ 1.34375,
556
+ 1.6171875,
557
+ 1.5625,
558
+ 1.2109375,
559
+ 1.4609375,
560
+ 1.71875,
561
+ 1.921875,
562
+ 1.8984375,
563
+ 1.703125,
564
+ 1.3828125,
565
+ 1.0859375,
566
+ 0.95703125,
567
+ 1.0078125,
568
+ 1.15625,
569
+ 0.8984375,
570
+ 0.5703125,
571
+ 0.5703125,
572
+ 0.734375,
573
+ 0.73046875,
574
+ 0.66015625,
575
+ 0.404296875,
576
+ 0.462890625,
577
+ 0.62109375,
578
+ 0.546875,
579
+ 0.326171875,
580
+ 0.423828125,
581
+ 0.4140625,
582
+ 0.416015625,
583
+ 0.322265625,
584
+ 0.419921875,
585
+ 0.380859375,
586
+ 0.330078125,
587
+ 0.203125,
588
+ 0.314453125,
589
+ 0.38671875,
590
+ 0.265625,
591
+ 0.236328125,
592
+ 0.28125,
593
+ 0.255859375,
594
+ 0.189453125,
595
+ 0.220703125,
596
+ 0.1328125,
597
+ 0.1826171875,
598
+ 0.15625,
599
+ 0.2080078125,
600
+ 0.1806640625,
601
+ 0.189453125,
602
+ 0.1865234375,
603
+ 0.2177734375,
604
+ 0.2890625,
605
+ 0.197265625,
606
+ 0.2099609375,
607
+ 0.1923828125,
608
+ 0.1845703125,
609
+ 0.263671875,
610
+ 0.197265625,
611
+ 0.19921875,
612
+ 0.1796875,
613
+ 0.212890625,
614
+ 0.1875,
615
+ 0.1689453125,
616
+ 0.16796875,
617
+ 0.185546875,
618
+ 0.2080078125,
619
+ 0.17578125,
620
+ 0.1484375,
621
+ 0.166015625,
622
+ 0.1708984375,
623
+ 0.1806640625,
624
+ 0.12890625,
625
+ 0.18359375,
626
+ 0.1953125,
627
+ 0.224609375,
628
+ 0.1630859375,
629
+ 0.1435546875,
630
+ 0.1669921875,
631
+ 0.1728515625,
632
+ 0.1923828125,
633
+ 0.1845703125,
634
+ 0.150390625,
635
+ 0.1943359375,
636
+ 0.23828125,
637
+ 0.19140625,
638
+ 0.2490234375,
639
+ 0.181640625,
640
+ 0.1796875,
641
+ 0.2119140625,
642
+ 0.212890625,
643
+ 0.2275390625,
644
+ 0.150390625,
645
+ 0.1728515625,
646
+ 0.171875,
647
+ 0.255859375,
648
+ 0.2490234375,
649
+ 0.1630859375,
650
+ 0.216796875,
651
+ 0.25390625,
652
+ 0.17578125,
653
+ 0.224609375,
654
+ 0.2490234375,
655
+ 0.1904296875,
656
+ 0.2158203125,
657
+ 0.298828125,
658
+ 0.2138671875,
659
+ 0.2060546875,
660
+ 0.1787109375,
661
+ 0.185546875,
662
+ 0.23828125,
663
+ 0.1962890625,
664
+ 0.228515625,
665
+ 0.1875,
666
+ 0.208984375,
667
+ 0.169921875,
668
+ 0.212890625,
669
+ 0.2021484375,
670
+ 0.220703125,
671
+ 0.166015625,
672
+ 0.205078125,
673
+ 0.2451171875,
674
+ 0.1669921875,
675
+ 0.255859375,
676
+ 0.2275390625,
677
+ 0.189453125,
678
+ 0.181640625,
679
+ 0.275390625,
680
+ 0.185546875,
681
+ 0.2060546875,
682
+ 0.251953125,
683
+ 0.2578125,
684
+ 0.193359375,
685
+ 0.216796875,
686
+ 0.3046875,
687
+ 0.1337890625,
688
+ 0.2294921875,
689
+ 0.1796875,
690
+ 0.14453125,
691
+ 0.12451171875,
692
+ 0.13671875,
693
+ 0.1943359375,
694
+ 0.1728515625,
695
+ 0.1826171875,
696
+ 0.1611328125,
697
+ 0.1845703125,
698
+ 0.1669921875,
699
+ 0.255859375,
700
+ 0.212890625,
701
+ 0.1796875,
702
+ 0.2216796875,
703
+ 0.25390625,
704
+ 0.2294921875,
705
+ 0.1650390625,
706
+ 0.140625,
707
+ 0.150390625,
708
+ 0.212890625,
709
+ 0.1923828125,
710
+ 0.1630859375,
711
+ 0.166015625,
712
+ 0.15234375,
713
+ 0.2001953125,
714
+ 0.2021484375,
715
+ 0.2099609375,
716
+ 0.1982421875,
717
+ 0.162109375,
718
+ 0.1396484375,
719
+ 0.1611328125,
720
+ 0.2109375,
721
+ 0.2197265625,
722
+ 0.1904296875,
723
+ 0.1328125,
724
+ 0.1650390625,
725
+ 0.1953125,
726
+ 0.2236328125,
727
+ 0.1474609375,
728
+ 0.1806640625,
729
+ 0.181640625,
730
+ 0.205078125,
731
+ 0.1806640625,
732
+ 0.212890625,
733
+ 0.11669921875,
734
+ 0.251953125,
735
+ 0.2275390625,
736
+ 0.2392578125,
737
+ 0.212890625,
738
+ 0.1376953125,
739
+ 0.1826171875,
740
+ 0.1943359375,
741
+ 0.158203125,
742
+ 0.12890625,
743
+ 0.185546875,
744
+ 0.185546875,
745
+ 0.158203125,
746
+ 0.1796875,
747
+ 0.1787109375,
748
+ 0.2001953125,
749
+ 0.265625,
750
+ 0.1796875,
751
+ 0.2255859375,
752
+ 0.1435546875,
753
+ 0.1611328125,
754
+ 0.189453125,
755
+ 0.177734375,
756
+ 0.2216796875,
757
+ 0.166015625,
758
+ 0.2001953125,
759
+ 0.216796875,
760
+ 0.15625,
761
+ 0.1904296875,
762
+ 0.12890625,
763
+ 0.1396484375,
764
+ 0.140625,
765
+ 0.1806640625,
766
+ 0.201171875,
767
+ 0.1767578125,
768
+ 0.146484375,
769
+ 0.1455078125,
770
+ 0.2314453125,
771
+ 0.2109375,
772
+ 0.1650390625,
773
+ 0.1689453125,
774
+ 0.140625,
775
+ 0.1728515625,
776
+ 0.1474609375,
777
+ 0.1572265625,
778
+ 0.1689453125,
779
+ 0.1259765625,
780
+ 0.193359375,
781
+ 0.11083984375,
782
+ 0.17578125,
783
+ 0.1494140625,
784
+ 0.14453125,
785
+ 0.2451171875,
786
+ 0.2412109375,
787
+ 0.177734375,
788
+ 0.189453125,
789
+ 0.1474609375,
790
+ 0.1376953125,
791
+ 0.1650390625,
792
+ 0.2021484375,
793
+ 0.162109375,
794
+ 0.18359375,
795
+ 0.150390625,
796
+ 0.169921875,
797
+ 0.2216796875,
798
+ 0.15625,
799
+ 0.2001953125,
800
+ 0.2060546875,
801
+ 0.1904296875,
802
+ 0.216796875,
803
+ 0.181640625,
804
+ 0.22265625,
805
+ 0.142578125,
806
+ 0.267578125,
807
+ 0.212890625,
808
+ 0.162109375,
809
+ 0.2421875,
810
+ 0.2119140625,
811
+ 0.29296875
812
+ ]
813
+ },
814
+ "eval": {
815
+ "effective_batch_nums": [
816
+ 50,
817
+ 100,
818
+ 150,
819
+ 200,
820
+ 250
821
+ ],
822
+ "losses": [
823
+ 8.525735294117647,
824
+ 8.5,
825
+ 8.496323529411764,
826
+ 8.485294117647058,
827
+ 8.488970588235293
828
+ ],
829
+ "perplexities": [
830
+ 5065.411764705882,
831
+ 4931.764705882353,
832
+ 4912.941176470588,
833
+ 4862.117647058823,
834
+ 4880.941176470588
835
+ ],
836
+ "accuracies": [
837
+ 0.050909109305596306,
838
+ 0.050909109305596306,
839
+ 0.050909109305596306,
840
+ 0.050909109305596306,
841
+ 0.05846225451013325
842
+ ]
843
+ }
844
+ }
control_42/epoch1/model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6dc35c8ae8aac4d84f9259fc9a595f32d15d4183c0e370f6d145f666c2eeedf4
3
+ size 18206151
control_42/epoch1/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfa13c66141bc0e9b3c86261f6ca6209220fcaf990ea922fde6336e1d1a4f1a3
3
+ size 36450891
control_42/epoch1/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:079708af01390ffdfc72a13ac3d31671cf218fbc15732cc6a33035717297dd52
3
+ size 1465
control_42/epoch1/tokenizer/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
control_42/epoch1/tokenizer/special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "pad_token": "<|endoftext|>",
5
+ "unk_token": "<|endoftext|>"
6
+ }
control_42/epoch1/tokenizer/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
control_42/epoch1/tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "50256": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ }
12
+ },
13
+ "bos_token": "<|endoftext|>",
14
+ "clean_up_tokenization_spaces": false,
15
+ "eos_token": "<|endoftext|>",
16
+ "extra_special_tokens": {},
17
+ "model_max_length": 1024,
18
+ "pad_token": "<|endoftext|>",
19
+ "tokenizer_class": "GPT2Tokenizer",
20
+ "unk_token": "<|endoftext|>"
21
+ }
control_42/epoch1/tokenizer/vocab.json ADDED
The diff for this file is too large to render. See raw diff