francesca9805 commited on
Commit
abfcb6e
·
verified ·
1 Parent(s): 06077ec

Training checkpoint: step 500

Browse files
checkpoint-500/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:25404087543d39778552df8c48ca227537b494e8ddd664b7e9b26bc795737b1f
3
  size 249556672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ca4f96444cf6400ce2c82082e3e155c5d4232c8d34753905219f9f74475ab50
3
  size 249556672
checkpoint-500/trainer_state.json CHANGED
@@ -10,7 +10,7 @@
10
  "is_world_process_zero": true,
11
  "log_history": [
12
  {
13
- "entropy": 10.692000007629394,
14
  "epoch": 0.00408997955010225,
15
  "grad_norm": 12.6875,
16
  "learning_rate": 2e-06,
@@ -20,1004 +20,1004 @@
20
  "step": 5
21
  },
22
  {
23
- "entropy": 10.691901969909669,
24
  "epoch": 0.0081799591002045,
25
  "grad_norm": 12.5,
26
  "learning_rate": 4.5e-06,
27
  "loss": 10.9107,
28
- "mean_token_accuracy": 0.0001713533652946353,
29
  "num_tokens": 162815.0,
30
  "step": 10
31
  },
32
  {
33
- "entropy": 10.69065294265747,
34
  "epoch": 0.012269938650306749,
35
  "grad_norm": 8.0,
36
  "learning_rate": 7e-06,
37
  "loss": 10.5862,
38
- "mean_token_accuracy": 0.02130717276595533,
39
  "num_tokens": 244175.0,
40
  "step": 15
41
  },
42
  {
43
- "entropy": 10.679887104034425,
44
  "epoch": 0.016359918200409,
45
  "grad_norm": 4.46875,
46
  "learning_rate": 9.5e-06,
47
  "loss": 10.2158,
48
- "mean_token_accuracy": 0.03475592397153378,
49
  "num_tokens": 324928.0,
50
  "step": 20
51
  },
52
  {
53
- "entropy": 10.636293029785156,
54
  "epoch": 0.02044989775051125,
55
  "grad_norm": 2.796875,
56
  "learning_rate": 1.2e-05,
57
  "loss": 9.8963,
58
- "mean_token_accuracy": 0.0389385461807251,
59
  "num_tokens": 406579.0,
60
  "step": 25
61
  },
62
  {
63
- "entropy": 10.604208278656007,
64
  "epoch": 0.024539877300613498,
65
  "grad_norm": 2.4375,
66
  "learning_rate": 1.4500000000000002e-05,
67
  "loss": 9.7363,
68
- "mean_token_accuracy": 0.04344259537756443,
69
  "num_tokens": 487674.0,
70
  "step": 30
71
  },
72
  {
73
- "entropy": 10.598915004730225,
74
  "epoch": 0.028629856850715747,
75
  "grad_norm": 2.296875,
76
  "learning_rate": 1.7000000000000003e-05,
77
  "loss": 9.6794,
78
- "mean_token_accuracy": 0.04643490575253963,
79
  "num_tokens": 569303.0,
80
  "step": 35
81
  },
82
  {
83
- "entropy": 10.59752607345581,
84
  "epoch": 0.032719836400818,
85
  "grad_norm": 2.375,
86
  "learning_rate": 1.95e-05,
87
  "loss": 9.6149,
88
- "mean_token_accuracy": 0.0586068470031023,
89
  "num_tokens": 650813.0,
90
  "step": 40
91
  },
92
  {
93
- "entropy": 10.537735271453858,
94
  "epoch": 0.03680981595092025,
95
- "grad_norm": 2.5,
96
  "learning_rate": 2.2e-05,
97
- "loss": 9.4929,
98
- "mean_token_accuracy": 0.07020987533032894,
99
  "num_tokens": 732280.0,
100
  "step": 45
101
  },
102
  {
103
- "entropy": 10.376041603088378,
104
  "epoch": 0.0408997955010225,
105
  "grad_norm": 2.4375,
106
  "learning_rate": 2.4500000000000003e-05,
107
- "loss": 9.3923,
108
- "mean_token_accuracy": 0.07574649453163147,
109
  "num_tokens": 813564.0,
110
  "step": 50
111
  },
112
  {
113
- "entropy": 10.325905418395996,
114
  "epoch": 0.044989775051124746,
115
  "grad_norm": 2.1875,
116
  "learning_rate": 2.7e-05,
117
- "loss": 9.3309,
118
- "mean_token_accuracy": 0.07549078464508056,
119
  "num_tokens": 895017.0,
120
  "step": 55
121
  },
122
  {
123
- "entropy": 10.389178371429443,
124
  "epoch": 0.049079754601226995,
125
  "grad_norm": 2.140625,
126
  "learning_rate": 2.95e-05,
127
- "loss": 9.2898,
128
- "mean_token_accuracy": 0.07733634486794472,
129
  "num_tokens": 976507.0,
130
  "step": 60
131
  },
132
  {
133
- "entropy": 10.312945556640624,
134
  "epoch": 0.053169734151329244,
135
  "grad_norm": 1.921875,
136
  "learning_rate": 3.2e-05,
137
- "loss": 9.1382,
138
- "mean_token_accuracy": 0.08499589711427688,
139
  "num_tokens": 1057953.0,
140
  "step": 65
141
  },
142
  {
143
- "entropy": 10.195150089263915,
144
  "epoch": 0.05725971370143149,
145
  "grad_norm": 2.046875,
146
  "learning_rate": 3.4500000000000005e-05,
147
  "loss": 8.9981,
148
- "mean_token_accuracy": 0.09292872175574303,
149
  "num_tokens": 1138725.0,
150
  "step": 70
151
  },
152
  {
153
- "entropy": 10.178875064849853,
154
  "epoch": 0.06134969325153374,
155
  "grad_norm": 2.421875,
156
  "learning_rate": 3.7e-05,
157
- "loss": 8.9207,
158
- "mean_token_accuracy": 0.09175906777381897,
159
  "num_tokens": 1219783.0,
160
  "step": 75
161
  },
162
  {
163
- "entropy": 10.17201852798462,
164
  "epoch": 0.065439672801636,
165
  "grad_norm": 1.9921875,
166
  "learning_rate": 3.95e-05,
167
  "loss": 8.7942,
168
- "mean_token_accuracy": 0.09474919810891151,
169
  "num_tokens": 1300892.0,
170
  "step": 80
171
  },
172
  {
173
- "entropy": 10.115837478637696,
174
  "epoch": 0.06952965235173825,
175
  "grad_norm": 1.84375,
176
  "learning_rate": 4.2000000000000004e-05,
177
- "loss": 8.6973,
178
- "mean_token_accuracy": 0.09508359506726265,
179
  "num_tokens": 1382196.0,
180
  "step": 85
181
  },
182
  {
183
- "entropy": 10.115576171875,
184
  "epoch": 0.0736196319018405,
185
  "grad_norm": 2.046875,
186
  "learning_rate": 4.45e-05,
187
- "loss": 8.5861,
188
- "mean_token_accuracy": 0.09681595638394355,
189
  "num_tokens": 1463433.0,
190
  "step": 90
191
  },
192
  {
193
- "entropy": 10.04802074432373,
194
  "epoch": 0.07770961145194274,
195
  "grad_norm": 1.765625,
196
  "learning_rate": 4.7000000000000004e-05,
197
- "loss": 8.4928,
198
- "mean_token_accuracy": 0.09577249959111214,
199
  "num_tokens": 1544967.0,
200
  "step": 95
201
  },
202
  {
203
- "entropy": 10.002210521697998,
204
  "epoch": 0.081799591002045,
205
- "grad_norm": 1.859375,
206
  "learning_rate": 4.9500000000000004e-05,
207
- "loss": 8.349,
208
- "mean_token_accuracy": 0.1003450721502304,
209
  "num_tokens": 1626565.0,
210
  "step": 100
211
  },
212
  {
213
- "entropy": 9.913407230377198,
214
  "epoch": 0.08588957055214724,
215
- "grad_norm": 1.765625,
216
  "learning_rate": 5.2e-05,
217
- "loss": 8.2142,
218
- "mean_token_accuracy": 0.10156635493040085,
219
  "num_tokens": 1707817.0,
220
  "step": 105
221
  },
222
  {
223
- "entropy": 9.820742130279541,
224
  "epoch": 0.08997955010224949,
225
  "grad_norm": 1.6484375,
226
  "learning_rate": 5.45e-05,
227
- "loss": 8.0762,
228
- "mean_token_accuracy": 0.10722621455788613,
229
  "num_tokens": 1789495.0,
230
  "step": 110
231
  },
232
  {
233
- "entropy": 9.724610519409179,
234
  "epoch": 0.09406952965235174,
235
- "grad_norm": 1.421875,
236
  "learning_rate": 5.7e-05,
237
- "loss": 7.9729,
238
- "mean_token_accuracy": 0.10886133536696434,
239
  "num_tokens": 1871084.0,
240
  "step": 115
241
  },
242
  {
243
- "entropy": 9.586759853363038,
244
  "epoch": 0.09815950920245399,
245
  "grad_norm": 1.3203125,
246
  "learning_rate": 5.9499999999999996e-05,
247
  "loss": 7.8413,
248
- "mean_token_accuracy": 0.10867177844047546,
249
  "num_tokens": 1952298.0,
250
  "step": 120
251
  },
252
  {
253
- "entropy": 9.416104125976563,
254
  "epoch": 0.10224948875255624,
255
  "grad_norm": 1.4765625,
256
  "learning_rate": 6.2e-05,
257
  "loss": 7.6389,
258
- "mean_token_accuracy": 0.1137846939265728,
259
  "num_tokens": 2033946.0,
260
  "step": 125
261
  },
262
  {
263
- "entropy": 9.18623046875,
264
  "epoch": 0.10633946830265849,
265
  "grad_norm": 1.171875,
266
  "learning_rate": 6.450000000000001e-05,
267
  "loss": 7.5292,
268
- "mean_token_accuracy": 0.11570263877511025,
269
  "num_tokens": 2115355.0,
270
  "step": 130
271
  },
272
  {
273
- "entropy": 8.975252056121827,
274
  "epoch": 0.11042944785276074,
275
  "grad_norm": 1.2421875,
276
  "learning_rate": 6.7e-05,
277
- "loss": 7.4295,
278
- "mean_token_accuracy": 0.11486484259366989,
279
  "num_tokens": 2195805.0,
280
  "step": 135
281
  },
282
  {
283
- "entropy": 8.759253120422363,
284
  "epoch": 0.11451942740286299,
285
  "grad_norm": 1.1640625,
286
  "learning_rate": 6.950000000000001e-05,
287
  "loss": 7.3127,
288
- "mean_token_accuracy": 0.11689863055944442,
289
  "num_tokens": 2277317.0,
290
  "step": 140
291
  },
292
  {
293
- "entropy": 8.507745265960693,
294
  "epoch": 0.11860940695296524,
295
  "grad_norm": 1.0703125,
296
  "learning_rate": 7.2e-05,
297
- "loss": 7.2512,
298
- "mean_token_accuracy": 0.12101163789629936,
299
  "num_tokens": 2358952.0,
300
  "step": 145
301
  },
302
  {
303
- "entropy": 8.293633270263673,
304
  "epoch": 0.12269938650306748,
305
  "grad_norm": 1.078125,
306
  "learning_rate": 7.45e-05,
307
- "loss": 7.134,
308
- "mean_token_accuracy": 0.12058043330907822,
309
  "num_tokens": 2440370.0,
310
  "step": 150
311
  },
312
  {
313
- "entropy": 8.062254238128663,
314
  "epoch": 0.12678936605316973,
315
- "grad_norm": 0.921875,
316
  "learning_rate": 7.7e-05,
317
- "loss": 7.1057,
318
- "mean_token_accuracy": 0.12724015042185782,
319
  "num_tokens": 2521468.0,
320
  "step": 155
321
  },
322
  {
323
- "entropy": 7.97076096534729,
324
  "epoch": 0.130879345603272,
325
- "grad_norm": 1.0546875,
326
  "learning_rate": 7.950000000000001e-05,
327
  "loss": 7.0757,
328
- "mean_token_accuracy": 0.1224481850862503,
329
  "num_tokens": 2601986.0,
330
  "step": 160
331
  },
332
  {
333
- "entropy": 7.803533411026001,
334
  "epoch": 0.13496932515337423,
335
- "grad_norm": 0.9609375,
336
  "learning_rate": 8.2e-05,
337
  "loss": 6.9867,
338
- "mean_token_accuracy": 0.12610725834965705,
339
  "num_tokens": 2683683.0,
340
  "step": 165
341
  },
342
  {
343
- "entropy": 7.705219650268555,
344
  "epoch": 0.1390593047034765,
345
- "grad_norm": 1.4375,
346
  "learning_rate": 8.450000000000001e-05,
347
- "loss": 6.9849,
348
- "mean_token_accuracy": 0.12802302911877633,
349
  "num_tokens": 2765248.0,
350
  "step": 170
351
  },
352
  {
353
- "entropy": 7.637285327911377,
354
  "epoch": 0.14314928425357873,
355
- "grad_norm": 1.1015625,
356
  "learning_rate": 8.7e-05,
357
- "loss": 6.9514,
358
- "mean_token_accuracy": 0.12579782903194428,
359
  "num_tokens": 2846905.0,
360
  "step": 175
361
  },
362
  {
363
- "entropy": 7.632810878753662,
364
  "epoch": 0.147239263803681,
365
- "grad_norm": 0.9921875,
366
  "learning_rate": 8.95e-05,
367
- "loss": 6.9527,
368
- "mean_token_accuracy": 0.1277453050017357,
369
  "num_tokens": 2927747.0,
370
  "step": 180
371
  },
372
  {
373
- "entropy": 7.515249443054199,
374
  "epoch": 0.15132924335378323,
375
- "grad_norm": 0.81640625,
376
  "learning_rate": 9.2e-05,
377
- "loss": 6.8672,
378
- "mean_token_accuracy": 0.13292431309819222,
379
  "num_tokens": 3008816.0,
380
  "step": 185
381
  },
382
  {
383
- "entropy": 7.475357294082642,
384
  "epoch": 0.1554192229038855,
385
  "grad_norm": 1.140625,
386
  "learning_rate": 9.45e-05,
387
- "loss": 6.8262,
388
- "mean_token_accuracy": 0.13385789096355438,
389
  "num_tokens": 3090309.0,
390
  "step": 190
391
  },
392
  {
393
- "entropy": 7.440796279907227,
394
  "epoch": 0.15950920245398773,
395
- "grad_norm": 0.84375,
396
  "learning_rate": 9.7e-05,
397
- "loss": 6.8687,
398
- "mean_token_accuracy": 0.13352703005075456,
399
  "num_tokens": 3171767.0,
400
  "step": 195
401
  },
402
  {
403
- "entropy": 7.414162540435791,
404
  "epoch": 0.16359918200409,
405
- "grad_norm": 0.89453125,
406
  "learning_rate": 9.95e-05,
407
- "loss": 6.8278,
408
- "mean_token_accuracy": 0.13457100391387938,
409
  "num_tokens": 3253322.0,
410
  "step": 200
411
  },
412
  {
413
- "entropy": 7.404553985595703,
414
  "epoch": 0.16768916155419222,
415
- "grad_norm": 1.1171875,
416
  "learning_rate": 0.000102,
417
- "loss": 6.8333,
418
- "mean_token_accuracy": 0.13599402010440825,
419
  "num_tokens": 3334617.0,
420
  "step": 205
421
  },
422
  {
423
- "entropy": 7.334035396575928,
424
  "epoch": 0.17177914110429449,
425
  "grad_norm": 1.0625,
426
  "learning_rate": 0.00010449999999999999,
427
- "loss": 6.8057,
428
- "mean_token_accuracy": 0.13515702188014983,
429
  "num_tokens": 3416224.0,
430
  "step": 210
431
  },
432
  {
433
- "entropy": 7.291983222961425,
434
  "epoch": 0.17586912065439672,
435
- "grad_norm": 0.85546875,
436
  "learning_rate": 0.000107,
437
- "loss": 6.7844,
438
- "mean_token_accuracy": 0.13452038168907166,
439
  "num_tokens": 3497574.0,
440
  "step": 215
441
  },
442
  {
443
- "entropy": 7.307376432418823,
444
  "epoch": 0.17995910020449898,
445
  "grad_norm": 1.0078125,
446
  "learning_rate": 0.0001095,
447
- "loss": 6.7816,
448
- "mean_token_accuracy": 0.13682517260313035,
449
  "num_tokens": 3578525.0,
450
  "step": 220
451
  },
452
  {
453
- "entropy": 7.249621200561523,
454
  "epoch": 0.18404907975460122,
455
- "grad_norm": 0.828125,
456
  "learning_rate": 0.000112,
457
- "loss": 6.7702,
458
- "mean_token_accuracy": 0.1377768710255623,
459
  "num_tokens": 3660181.0,
460
  "step": 225
461
  },
462
  {
463
- "entropy": 7.2665845394134525,
464
  "epoch": 0.18813905930470348,
465
- "grad_norm": 0.76953125,
466
  "learning_rate": 0.0001145,
467
- "loss": 6.7515,
468
- "mean_token_accuracy": 0.136446313560009,
469
  "num_tokens": 3741809.0,
470
  "step": 230
471
  },
472
  {
473
- "entropy": 7.236224603652954,
474
  "epoch": 0.19222903885480572,
475
- "grad_norm": 1.0234375,
476
  "learning_rate": 0.00011700000000000001,
477
- "loss": 6.7086,
478
- "mean_token_accuracy": 0.13772396147251129,
479
  "num_tokens": 3823389.0,
480
  "step": 235
481
  },
482
  {
483
- "entropy": 7.190951490402222,
484
  "epoch": 0.19631901840490798,
485
  "grad_norm": 0.9140625,
486
  "learning_rate": 0.00011949999999999999,
487
- "loss": 6.7268,
488
- "mean_token_accuracy": 0.13577288687229155,
489
  "num_tokens": 3904714.0,
490
  "step": 240
491
  },
492
  {
493
- "entropy": 7.149959564208984,
494
  "epoch": 0.20040899795501022,
495
- "grad_norm": 1.0390625,
496
  "learning_rate": 0.000122,
497
- "loss": 6.6735,
498
- "mean_token_accuracy": 0.13873157054185867,
499
  "num_tokens": 3986263.0,
500
  "step": 245
501
  },
502
  {
503
- "entropy": 7.147679090499878,
504
  "epoch": 0.20449897750511248,
505
- "grad_norm": 0.9140625,
506
  "learning_rate": 0.0001245,
507
- "loss": 6.6771,
508
- "mean_token_accuracy": 0.13870261535048484,
509
  "num_tokens": 4067896.0,
510
  "step": 250
511
  },
512
  {
513
- "entropy": 7.115998220443726,
514
  "epoch": 0.2085889570552147,
515
- "grad_norm": 0.92578125,
516
  "learning_rate": 0.000127,
517
- "loss": 6.6749,
518
- "mean_token_accuracy": 0.14127641916275024,
519
  "num_tokens": 4149422.0,
520
  "step": 255
521
  },
522
  {
523
- "entropy": 7.07224407196045,
524
  "epoch": 0.21267893660531698,
525
- "grad_norm": 0.84765625,
526
  "learning_rate": 0.0001295,
527
- "loss": 6.5745,
528
- "mean_token_accuracy": 0.14302785694599152,
529
  "num_tokens": 4230807.0,
530
  "step": 260
531
  },
532
  {
533
- "entropy": 7.013430881500244,
534
  "epoch": 0.2167689161554192,
535
- "grad_norm": 1.0703125,
536
  "learning_rate": 0.000132,
537
- "loss": 6.6014,
538
- "mean_token_accuracy": 0.14430382549762727,
539
  "num_tokens": 4312299.0,
540
  "step": 265
541
  },
542
  {
543
- "entropy": 7.082822227478028,
544
  "epoch": 0.22085889570552147,
545
- "grad_norm": 0.890625,
546
  "learning_rate": 0.00013450000000000002,
547
- "loss": 6.6315,
548
- "mean_token_accuracy": 0.14351730197668075,
549
  "num_tokens": 4393563.0,
550
  "step": 270
551
  },
552
  {
553
- "entropy": 7.030689144134522,
554
  "epoch": 0.2249488752556237,
555
  "grad_norm": 0.8046875,
556
  "learning_rate": 0.00013700000000000002,
557
- "loss": 6.655,
558
- "mean_token_accuracy": 0.1369350366294384,
559
  "num_tokens": 4474907.0,
560
  "step": 275
561
  },
562
  {
563
- "entropy": 7.044711637496948,
564
  "epoch": 0.22903885480572597,
565
- "grad_norm": 0.890625,
566
  "learning_rate": 0.0001395,
567
  "loss": 6.5587,
568
- "mean_token_accuracy": 0.14312482625246048,
569
  "num_tokens": 4556026.0,
570
  "step": 280
571
  },
572
  {
573
- "entropy": 7.011560773849487,
574
  "epoch": 0.2331288343558282,
575
- "grad_norm": 1.296875,
576
  "learning_rate": 0.00014199999999999998,
577
- "loss": 6.5626,
578
- "mean_token_accuracy": 0.14314131289720536,
579
  "num_tokens": 4637280.0,
580
  "step": 285
581
  },
582
  {
583
- "entropy": 6.985124301910401,
584
  "epoch": 0.23721881390593047,
585
  "grad_norm": 0.8671875,
586
  "learning_rate": 0.0001445,
587
- "loss": 6.5747,
588
- "mean_token_accuracy": 0.14551239907741548,
589
  "num_tokens": 4718900.0,
590
  "step": 290
591
  },
592
  {
593
- "entropy": 6.987762880325318,
594
  "epoch": 0.24130879345603273,
595
- "grad_norm": 0.9140625,
596
  "learning_rate": 0.000147,
597
- "loss": 6.5737,
598
- "mean_token_accuracy": 0.14291551113128662,
599
  "num_tokens": 4800507.0,
600
  "step": 295
601
  },
602
  {
603
- "entropy": 6.900343227386474,
604
  "epoch": 0.24539877300613497,
605
  "grad_norm": 0.859375,
606
  "learning_rate": 0.0001495,
607
- "loss": 6.5154,
608
- "mean_token_accuracy": 0.14588727355003356,
609
  "num_tokens": 4881971.0,
610
  "step": 300
611
  },
612
  {
613
- "entropy": 6.930733013153076,
614
  "epoch": 0.24948875255623723,
615
- "grad_norm": 0.90625,
616
  "learning_rate": 0.000152,
617
  "loss": 6.5192,
618
- "mean_token_accuracy": 0.14804027676582338,
619
  "num_tokens": 4963516.0,
620
  "step": 305
621
  },
622
  {
623
- "entropy": 6.871261930465698,
624
  "epoch": 0.25357873210633947,
625
- "grad_norm": 0.97265625,
626
  "learning_rate": 0.00015450000000000001,
627
- "loss": 6.4974,
628
- "mean_token_accuracy": 0.1481480434536934,
629
  "num_tokens": 5045190.0,
630
  "step": 310
631
  },
632
  {
633
- "entropy": 6.9122120380401615,
634
  "epoch": 0.25766871165644173,
635
- "grad_norm": 0.92578125,
636
  "learning_rate": 0.000157,
637
- "loss": 6.5102,
638
- "mean_token_accuracy": 0.14717467725276948,
639
  "num_tokens": 5126730.0,
640
  "step": 315
641
  },
642
  {
643
- "entropy": 6.862851333618164,
644
  "epoch": 0.261758691206544,
645
- "grad_norm": 0.8671875,
646
  "learning_rate": 0.0001595,
647
  "loss": 6.5129,
648
- "mean_token_accuracy": 0.14673057496547698,
649
  "num_tokens": 5207888.0,
650
  "step": 320
651
  },
652
  {
653
- "entropy": 6.848834371566772,
654
  "epoch": 0.2658486707566462,
655
- "grad_norm": 0.82421875,
656
  "learning_rate": 0.000162,
657
- "loss": 6.4373,
658
- "mean_token_accuracy": 0.15214565843343736,
659
  "num_tokens": 5289484.0,
660
  "step": 325
661
  },
662
  {
663
- "entropy": 6.819906520843506,
664
  "epoch": 0.26993865030674846,
665
  "grad_norm": 0.75,
666
  "learning_rate": 0.00016450000000000001,
667
- "loss": 6.4051,
668
- "mean_token_accuracy": 0.15237780809402465,
669
  "num_tokens": 5371086.0,
670
  "step": 330
671
  },
672
  {
673
- "entropy": 6.868084287643432,
674
  "epoch": 0.2740286298568507,
675
  "grad_norm": 0.77734375,
676
  "learning_rate": 0.00016700000000000002,
677
- "loss": 6.5088,
678
- "mean_token_accuracy": 0.14460985064506532,
679
  "num_tokens": 5452574.0,
680
  "step": 335
681
  },
682
  {
683
- "entropy": 6.770962238311768,
684
  "epoch": 0.278118609406953,
685
  "grad_norm": 0.89453125,
686
  "learning_rate": 0.00016950000000000003,
687
- "loss": 6.4104,
688
- "mean_token_accuracy": 0.15459631383419037,
689
  "num_tokens": 5534091.0,
690
  "step": 340
691
  },
692
  {
693
- "entropy": 6.788186645507812,
694
  "epoch": 0.2822085889570552,
695
- "grad_norm": 0.87890625,
696
  "learning_rate": 0.00017199999999999998,
697
- "loss": 6.456,
698
- "mean_token_accuracy": 0.14959714710712432,
699
  "num_tokens": 5614709.0,
700
  "step": 345
701
  },
702
  {
703
- "entropy": 6.779982662200927,
704
  "epoch": 0.28629856850715746,
705
- "grad_norm": 0.97265625,
706
  "learning_rate": 0.00017449999999999999,
707
- "loss": 6.3974,
708
- "mean_token_accuracy": 0.15147258043289186,
709
  "num_tokens": 5695945.0,
710
  "step": 350
711
  },
712
  {
713
- "entropy": 6.722852182388306,
714
  "epoch": 0.2903885480572597,
715
- "grad_norm": 0.88671875,
716
  "learning_rate": 0.000177,
717
- "loss": 6.3872,
718
- "mean_token_accuracy": 0.1556071251630783,
719
  "num_tokens": 5776894.0,
720
  "step": 355
721
  },
722
  {
723
- "entropy": 6.682234907150269,
724
  "epoch": 0.294478527607362,
725
- "grad_norm": 0.83203125,
726
  "learning_rate": 0.0001795,
727
- "loss": 6.3415,
728
- "mean_token_accuracy": 0.15552895963191987,
729
  "num_tokens": 5858426.0,
730
  "step": 360
731
  },
732
  {
733
- "entropy": 6.751620960235596,
734
  "epoch": 0.2985685071574642,
735
- "grad_norm": 0.79296875,
736
  "learning_rate": 0.000182,
737
- "loss": 6.4349,
738
- "mean_token_accuracy": 0.1517845869064331,
739
  "num_tokens": 5939990.0,
740
  "step": 365
741
  },
742
  {
743
- "entropy": 6.6951488018035885,
744
  "epoch": 0.30265848670756645,
745
- "grad_norm": 0.76171875,
746
  "learning_rate": 0.0001845,
747
  "loss": 6.3261,
748
- "mean_token_accuracy": 0.15971007198095322,
749
  "num_tokens": 6021209.0,
750
  "step": 370
751
  },
752
  {
753
- "entropy": 6.727404975891114,
754
  "epoch": 0.3067484662576687,
755
  "grad_norm": 0.89453125,
756
  "learning_rate": 0.000187,
757
  "loss": 6.3882,
758
- "mean_token_accuracy": 0.1535564124584198,
759
  "num_tokens": 6102085.0,
760
  "step": 375
761
  },
762
  {
763
- "entropy": 6.668899202346802,
764
  "epoch": 0.310838445807771,
765
- "grad_norm": 0.90625,
766
  "learning_rate": 0.0001895,
767
- "loss": 6.3176,
768
- "mean_token_accuracy": 0.15698710680007935,
769
  "num_tokens": 6183631.0,
770
  "step": 380
771
  },
772
  {
773
- "entropy": 6.686964225769043,
774
  "epoch": 0.3149284253578732,
775
  "grad_norm": 0.72265625,
776
  "learning_rate": 0.000192,
777
- "loss": 6.347,
778
- "mean_token_accuracy": 0.15678837448358535,
779
  "num_tokens": 6264174.0,
780
  "step": 385
781
  },
782
  {
783
- "entropy": 6.649777507781982,
784
  "epoch": 0.31901840490797545,
785
  "grad_norm": 0.87109375,
786
  "learning_rate": 0.0001945,
787
- "loss": 6.2918,
788
- "mean_token_accuracy": 0.16049820333719253,
789
  "num_tokens": 6345616.0,
790
  "step": 390
791
  },
792
  {
793
- "entropy": 6.642700815200806,
794
  "epoch": 0.3231083844580777,
795
- "grad_norm": 0.8125,
796
  "learning_rate": 0.00019700000000000002,
797
  "loss": 6.2962,
798
- "mean_token_accuracy": 0.15763863176107407,
799
  "num_tokens": 6426371.0,
800
  "step": 395
801
  },
802
  {
803
- "entropy": 6.669038724899292,
804
  "epoch": 0.32719836400818,
805
  "grad_norm": 0.95703125,
806
  "learning_rate": 0.00019950000000000002,
807
- "loss": 6.3489,
808
- "mean_token_accuracy": 0.15595990121364595,
809
  "num_tokens": 6508017.0,
810
  "step": 400
811
  },
812
  {
813
- "entropy": 6.626407861709595,
814
  "epoch": 0.3312883435582822,
815
- "grad_norm": 1.03125,
816
  "learning_rate": 0.000202,
817
  "loss": 6.3398,
818
- "mean_token_accuracy": 0.15653544664382935,
819
  "num_tokens": 6589627.0,
820
  "step": 405
821
  },
822
  {
823
- "entropy": 6.614445304870605,
824
  "epoch": 0.33537832310838445,
825
  "grad_norm": 0.90625,
826
  "learning_rate": 0.00020449999999999998,
827
  "loss": 6.2965,
828
- "mean_token_accuracy": 0.1605309635400772,
829
  "num_tokens": 6670965.0,
830
  "step": 410
831
  },
832
  {
833
- "entropy": 6.663286781311035,
834
  "epoch": 0.3394683026584867,
835
- "grad_norm": 0.9296875,
836
  "learning_rate": 0.000207,
837
  "loss": 6.3392,
838
- "mean_token_accuracy": 0.15533859878778458,
839
  "num_tokens": 6752603.0,
840
  "step": 415
841
  },
842
  {
843
- "entropy": 6.556842565536499,
844
  "epoch": 0.34355828220858897,
845
- "grad_norm": 0.8984375,
846
  "learning_rate": 0.0002095,
847
- "loss": 6.2426,
848
- "mean_token_accuracy": 0.16100438684225082,
849
  "num_tokens": 6833529.0,
850
  "step": 420
851
  },
852
  {
853
- "entropy": 6.608948469161987,
854
  "epoch": 0.3476482617586912,
855
- "grad_norm": 0.82421875,
856
  "learning_rate": 0.000212,
857
- "loss": 6.2617,
858
- "mean_token_accuracy": 0.1630496382713318,
859
  "num_tokens": 6915141.0,
860
  "step": 425
861
  },
862
  {
863
- "entropy": 6.538837718963623,
864
  "epoch": 0.35173824130879344,
865
- "grad_norm": 0.8125,
866
  "learning_rate": 0.0002145,
867
  "loss": 6.2681,
868
- "mean_token_accuracy": 0.15852631628513336,
869
  "num_tokens": 6996145.0,
870
  "step": 430
871
  },
872
  {
873
- "entropy": 6.586381292343139,
874
  "epoch": 0.3558282208588957,
875
  "grad_norm": 0.74609375,
876
  "learning_rate": 0.00021700000000000002,
877
- "loss": 6.2617,
878
- "mean_token_accuracy": 0.1601476714015007,
879
  "num_tokens": 7077692.0,
880
  "step": 435
881
  },
882
  {
883
- "entropy": 6.546258163452149,
884
  "epoch": 0.35991820040899797,
885
- "grad_norm": 0.81640625,
886
  "learning_rate": 0.0002195,
887
- "loss": 6.2647,
888
- "mean_token_accuracy": 0.16120847016572953,
889
  "num_tokens": 7159364.0,
890
  "step": 440
891
  },
892
  {
893
- "entropy": 6.547860431671142,
894
  "epoch": 0.36400817995910023,
895
  "grad_norm": 0.76171875,
896
  "learning_rate": 0.000222,
897
- "loss": 6.2299,
898
- "mean_token_accuracy": 0.16258099377155305,
899
  "num_tokens": 7240330.0,
900
  "step": 445
901
  },
902
  {
903
- "entropy": 6.473711585998535,
904
  "epoch": 0.36809815950920244,
905
- "grad_norm": 0.875,
906
  "learning_rate": 0.0002245,
907
- "loss": 6.1577,
908
- "mean_token_accuracy": 0.1646140530705452,
909
  "num_tokens": 7320583.0,
910
  "step": 450
911
  },
912
  {
913
- "entropy": 6.495396137237549,
914
  "epoch": 0.3721881390593047,
915
- "grad_norm": 0.75390625,
916
  "learning_rate": 0.00022700000000000002,
917
- "loss": 6.2007,
918
- "mean_token_accuracy": 0.16422612816095353,
919
  "num_tokens": 7401685.0,
920
  "step": 455
921
  },
922
  {
923
- "entropy": 6.505111026763916,
924
  "epoch": 0.37627811860940696,
925
- "grad_norm": 0.78125,
926
  "learning_rate": 0.00022950000000000002,
927
  "loss": 6.1459,
928
- "mean_token_accuracy": 0.16756232380867003,
929
  "num_tokens": 7482873.0,
930
  "step": 460
931
  },
932
  {
933
- "entropy": 6.463589715957641,
934
  "epoch": 0.3803680981595092,
935
- "grad_norm": 0.7421875,
936
  "learning_rate": 0.00023200000000000003,
937
  "loss": 6.157,
938
- "mean_token_accuracy": 0.16520966142416,
939
  "num_tokens": 7564543.0,
940
  "step": 465
941
  },
942
  {
943
- "entropy": 6.440634059906006,
944
  "epoch": 0.38445807770961143,
945
- "grad_norm": 0.8359375,
946
  "learning_rate": 0.00023449999999999998,
947
- "loss": 6.1861,
948
- "mean_token_accuracy": 0.16435976326465607,
949
  "num_tokens": 7645311.0,
950
  "step": 470
951
  },
952
  {
953
- "entropy": 6.524266099929809,
954
  "epoch": 0.3885480572597137,
955
- "grad_norm": 0.875,
956
  "learning_rate": 0.000237,
957
- "loss": 6.2155,
958
- "mean_token_accuracy": 0.1614104002714157,
959
  "num_tokens": 7726050.0,
960
  "step": 475
961
  },
962
  {
963
- "entropy": 6.451401805877685,
964
  "epoch": 0.39263803680981596,
965
  "grad_norm": 0.765625,
966
  "learning_rate": 0.0002395,
967
- "loss": 6.193,
968
- "mean_token_accuracy": 0.16441368460655212,
969
  "num_tokens": 7807452.0,
970
  "step": 480
971
  },
972
  {
973
- "entropy": 6.459316158294678,
974
  "epoch": 0.3967280163599182,
975
- "grad_norm": 0.73046875,
976
  "learning_rate": 0.000242,
977
- "loss": 6.1736,
978
- "mean_token_accuracy": 0.16316543221473695,
979
  "num_tokens": 7889159.0,
980
  "step": 485
981
  },
982
  {
983
- "entropy": 6.512886714935303,
984
  "epoch": 0.40081799591002043,
985
  "grad_norm": 0.765625,
986
  "learning_rate": 0.0002445,
987
- "loss": 6.2242,
988
- "mean_token_accuracy": 0.1614704579114914,
989
  "num_tokens": 7970803.0,
990
  "step": 490
991
  },
992
  {
993
- "entropy": 6.389158916473389,
994
  "epoch": 0.4049079754601227,
995
  "grad_norm": 0.734375,
996
  "learning_rate": 0.000247,
997
- "loss": 6.0863,
998
- "mean_token_accuracy": 0.1714015007019043,
999
  "num_tokens": 8051777.0,
1000
  "step": 495
1001
  },
1002
  {
1003
- "entropy": 6.496681356430054,
1004
  "epoch": 0.40899795501022496,
1005
  "grad_norm": 0.79296875,
1006
  "learning_rate": 0.0002495,
1007
- "loss": 6.1939,
1008
- "mean_token_accuracy": 0.16548911184072496,
1009
  "num_tokens": 8133204.0,
1010
  "step": 500
1011
  },
1012
  {
1013
  "epoch": 0.40899795501022496,
1014
- "eval_entropy": 6.274000018720001,
1015
- "eval_loss": 6.0819621086120605,
1016
- "eval_mean_token_accuracy": 0.17133162983308428,
1017
  "eval_num_tokens": 8133204.0,
1018
- "eval_runtime": 25.6732,
1019
- "eval_samples_per_second": 161.141,
1020
- "eval_steps_per_second": 20.177,
1021
  "step": 500
1022
  }
1023
  ],
 
10
  "is_world_process_zero": true,
11
  "log_history": [
12
  {
13
+ "entropy": 10.692000198364259,
14
  "epoch": 0.00408997955010225,
15
  "grad_norm": 12.6875,
16
  "learning_rate": 2e-06,
 
20
  "step": 5
21
  },
22
  {
23
+ "entropy": 10.691902351379394,
24
  "epoch": 0.0081799591002045,
25
  "grad_norm": 12.5,
26
  "learning_rate": 4.5e-06,
27
  "loss": 10.9107,
28
+ "mean_token_accuracy": 0.0001590834290254861,
29
  "num_tokens": 162815.0,
30
  "step": 10
31
  },
32
  {
33
+ "entropy": 10.690652847290039,
34
  "epoch": 0.012269938650306749,
35
  "grad_norm": 8.0,
36
  "learning_rate": 7e-06,
37
  "loss": 10.5862,
38
+ "mean_token_accuracy": 0.021183770173229276,
39
  "num_tokens": 244175.0,
40
  "step": 15
41
  },
42
  {
43
+ "entropy": 10.67988691329956,
44
  "epoch": 0.016359918200409,
45
  "grad_norm": 4.46875,
46
  "learning_rate": 9.5e-06,
47
  "loss": 10.2158,
48
+ "mean_token_accuracy": 0.03468580432236194,
49
  "num_tokens": 324928.0,
50
  "step": 20
51
  },
52
  {
53
+ "entropy": 10.636291885375977,
54
  "epoch": 0.02044989775051125,
55
  "grad_norm": 2.796875,
56
  "learning_rate": 1.2e-05,
57
  "loss": 9.8963,
58
+ "mean_token_accuracy": 0.039024268835783006,
59
  "num_tokens": 406579.0,
60
  "step": 25
61
  },
62
  {
63
+ "entropy": 10.604203414916991,
64
  "epoch": 0.024539877300613498,
65
  "grad_norm": 2.4375,
66
  "learning_rate": 1.4500000000000002e-05,
67
  "loss": 9.7363,
68
+ "mean_token_accuracy": 0.04346781931817532,
69
  "num_tokens": 487674.0,
70
  "step": 30
71
  },
72
  {
73
+ "entropy": 10.598912239074707,
74
  "epoch": 0.028629856850715747,
75
  "grad_norm": 2.296875,
76
  "learning_rate": 1.7000000000000003e-05,
77
  "loss": 9.6794,
78
+ "mean_token_accuracy": 0.04639807604253292,
79
  "num_tokens": 569303.0,
80
  "step": 35
81
  },
82
  {
83
+ "entropy": 10.597527599334716,
84
  "epoch": 0.032719836400818,
85
  "grad_norm": 2.375,
86
  "learning_rate": 1.95e-05,
87
  "loss": 9.6149,
88
+ "mean_token_accuracy": 0.05854539461433887,
89
  "num_tokens": 650813.0,
90
  "step": 40
91
  },
92
  {
93
+ "entropy": 10.537779712677002,
94
  "epoch": 0.03680981595092025,
95
+ "grad_norm": 2.484375,
96
  "learning_rate": 2.2e-05,
97
+ "loss": 9.493,
98
+ "mean_token_accuracy": 0.07012398801743984,
99
  "num_tokens": 732280.0,
100
  "step": 45
101
  },
102
  {
103
+ "entropy": 10.376054859161377,
104
  "epoch": 0.0408997955010225,
105
  "grad_norm": 2.4375,
106
  "learning_rate": 2.4500000000000003e-05,
107
+ "loss": 9.3922,
108
+ "mean_token_accuracy": 0.0757705345749855,
109
  "num_tokens": 813564.0,
110
  "step": 50
111
  },
112
  {
113
+ "entropy": 10.325862598419189,
114
  "epoch": 0.044989775051124746,
115
  "grad_norm": 2.1875,
116
  "learning_rate": 2.7e-05,
117
+ "loss": 9.3308,
118
+ "mean_token_accuracy": 0.07539180815219879,
119
  "num_tokens": 895017.0,
120
  "step": 55
121
  },
122
  {
123
+ "entropy": 10.389144897460938,
124
  "epoch": 0.049079754601226995,
125
  "grad_norm": 2.140625,
126
  "learning_rate": 2.95e-05,
127
+ "loss": 9.2897,
128
+ "mean_token_accuracy": 0.07736093625426292,
129
  "num_tokens": 976507.0,
130
  "step": 60
131
  },
132
  {
133
+ "entropy": 10.31296672821045,
134
  "epoch": 0.053169734151329244,
135
  "grad_norm": 1.921875,
136
  "learning_rate": 3.2e-05,
137
+ "loss": 9.1381,
138
+ "mean_token_accuracy": 0.08499588370323181,
139
  "num_tokens": 1057953.0,
140
  "step": 65
141
  },
142
  {
143
+ "entropy": 10.195121669769287,
144
  "epoch": 0.05725971370143149,
145
  "grad_norm": 2.046875,
146
  "learning_rate": 3.4500000000000005e-05,
147
  "loss": 8.9981,
148
+ "mean_token_accuracy": 0.0928544394671917,
149
  "num_tokens": 1138725.0,
150
  "step": 70
151
  },
152
  {
153
+ "entropy": 10.178840732574463,
154
  "epoch": 0.06134969325153374,
155
  "grad_norm": 2.421875,
156
  "learning_rate": 3.7e-05,
157
+ "loss": 8.9206,
158
+ "mean_token_accuracy": 0.09163383767008781,
159
  "num_tokens": 1219783.0,
160
  "step": 75
161
  },
162
  {
163
+ "entropy": 10.1720477104187,
164
  "epoch": 0.065439672801636,
165
  "grad_norm": 1.9921875,
166
  "learning_rate": 3.95e-05,
167
  "loss": 8.7942,
168
+ "mean_token_accuracy": 0.09474985226988793,
169
  "num_tokens": 1300892.0,
170
  "step": 80
171
  },
172
  {
173
+ "entropy": 10.115754795074462,
174
  "epoch": 0.06952965235173825,
175
  "grad_norm": 1.84375,
176
  "learning_rate": 4.2000000000000004e-05,
177
+ "loss": 8.6972,
178
+ "mean_token_accuracy": 0.09509614557027816,
179
  "num_tokens": 1382196.0,
180
  "step": 85
181
  },
182
  {
183
+ "entropy": 10.115432357788086,
184
  "epoch": 0.0736196319018405,
185
  "grad_norm": 2.046875,
186
  "learning_rate": 4.45e-05,
187
+ "loss": 8.5859,
188
+ "mean_token_accuracy": 0.09679216295480728,
189
  "num_tokens": 1463433.0,
190
  "step": 90
191
  },
192
  {
193
+ "entropy": 10.048014736175537,
194
  "epoch": 0.07770961145194274,
195
  "grad_norm": 1.765625,
196
  "learning_rate": 4.7000000000000004e-05,
197
+ "loss": 8.4925,
198
+ "mean_token_accuracy": 0.09578462094068527,
199
  "num_tokens": 1544967.0,
200
  "step": 95
201
  },
202
  {
203
+ "entropy": 10.00209379196167,
204
  "epoch": 0.081799591002045,
205
+ "grad_norm": 1.8515625,
206
  "learning_rate": 4.9500000000000004e-05,
207
+ "loss": 8.3486,
208
+ "mean_token_accuracy": 0.10033286660909653,
209
  "num_tokens": 1626565.0,
210
  "step": 100
211
  },
212
  {
213
+ "entropy": 9.9131498336792,
214
  "epoch": 0.08588957055214724,
215
+ "grad_norm": 1.78125,
216
  "learning_rate": 5.2e-05,
217
+ "loss": 8.2137,
218
+ "mean_token_accuracy": 0.10160388499498367,
219
  "num_tokens": 1707817.0,
220
  "step": 105
221
  },
222
  {
223
+ "entropy": 9.82022647857666,
224
  "epoch": 0.08997955010224949,
225
  "grad_norm": 1.6484375,
226
  "learning_rate": 5.45e-05,
227
+ "loss": 8.0758,
228
+ "mean_token_accuracy": 0.10720168352127075,
229
  "num_tokens": 1789495.0,
230
  "step": 110
231
  },
232
  {
233
+ "entropy": 9.724441719055175,
234
  "epoch": 0.09406952965235174,
235
+ "grad_norm": 1.4140625,
236
  "learning_rate": 5.7e-05,
237
+ "loss": 7.9726,
238
+ "mean_token_accuracy": 0.10897153615951538,
239
  "num_tokens": 1871084.0,
240
  "step": 115
241
  },
242
  {
243
+ "entropy": 9.586730098724365,
244
  "epoch": 0.09815950920245399,
245
  "grad_norm": 1.3203125,
246
  "learning_rate": 5.9499999999999996e-05,
247
  "loss": 7.8413,
248
+ "mean_token_accuracy": 0.10868334248661995,
249
  "num_tokens": 1952298.0,
250
  "step": 120
251
  },
252
  {
253
+ "entropy": 9.416004085540772,
254
  "epoch": 0.10224948875255624,
255
  "grad_norm": 1.4765625,
256
  "learning_rate": 6.2e-05,
257
  "loss": 7.6389,
258
+ "mean_token_accuracy": 0.11382156684994697,
259
  "num_tokens": 2033946.0,
260
  "step": 125
261
  },
262
  {
263
+ "entropy": 9.186109161376953,
264
  "epoch": 0.10633946830265849,
265
  "grad_norm": 1.171875,
266
  "learning_rate": 6.450000000000001e-05,
267
  "loss": 7.5292,
268
+ "mean_token_accuracy": 0.11576363742351532,
269
  "num_tokens": 2115355.0,
270
  "step": 130
271
  },
272
  {
273
+ "entropy": 8.975370025634765,
274
  "epoch": 0.11042944785276074,
275
  "grad_norm": 1.2421875,
276
  "learning_rate": 6.7e-05,
277
+ "loss": 7.4296,
278
+ "mean_token_accuracy": 0.11505270451307296,
279
  "num_tokens": 2195805.0,
280
  "step": 135
281
  },
282
  {
283
+ "entropy": 8.75938024520874,
284
  "epoch": 0.11451942740286299,
285
  "grad_norm": 1.1640625,
286
  "learning_rate": 6.950000000000001e-05,
287
  "loss": 7.3127,
288
+ "mean_token_accuracy": 0.11694714352488518,
289
  "num_tokens": 2277317.0,
290
  "step": 140
291
  },
292
  {
293
+ "entropy": 8.507802486419678,
294
  "epoch": 0.11860940695296524,
295
  "grad_norm": 1.0703125,
296
  "learning_rate": 7.2e-05,
297
+ "loss": 7.2511,
298
+ "mean_token_accuracy": 0.1210237443447113,
299
  "num_tokens": 2358952.0,
300
  "step": 145
301
  },
302
  {
303
+ "entropy": 8.29345064163208,
304
  "epoch": 0.12269938650306748,
305
  "grad_norm": 1.078125,
306
  "learning_rate": 7.45e-05,
307
+ "loss": 7.1339,
308
+ "mean_token_accuracy": 0.12062957882881165,
309
  "num_tokens": 2440370.0,
310
  "step": 150
311
  },
312
  {
313
+ "entropy": 8.062093591690063,
314
  "epoch": 0.12678936605316973,
315
+ "grad_norm": 0.92578125,
316
  "learning_rate": 7.7e-05,
317
+ "loss": 7.1056,
318
+ "mean_token_accuracy": 0.1271037481725216,
319
  "num_tokens": 2521468.0,
320
  "step": 155
321
  },
322
  {
323
+ "entropy": 7.970836448669433,
324
  "epoch": 0.130879345603272,
325
+ "grad_norm": 1.0625,
326
  "learning_rate": 7.950000000000001e-05,
327
  "loss": 7.0757,
328
+ "mean_token_accuracy": 0.1223635084927082,
329
  "num_tokens": 2601986.0,
330
  "step": 160
331
  },
332
  {
333
+ "entropy": 7.8029450416564945,
334
  "epoch": 0.13496932515337423,
335
+ "grad_norm": 0.96484375,
336
  "learning_rate": 8.2e-05,
337
  "loss": 6.9867,
338
+ "mean_token_accuracy": 0.12624209150671958,
339
  "num_tokens": 2683683.0,
340
  "step": 165
341
  },
342
  {
343
+ "entropy": 7.705253219604492,
344
  "epoch": 0.1390593047034765,
345
+ "grad_norm": 1.4609375,
346
  "learning_rate": 8.450000000000001e-05,
347
+ "loss": 6.985,
348
+ "mean_token_accuracy": 0.12793700769543648,
349
  "num_tokens": 2765248.0,
350
  "step": 170
351
  },
352
  {
353
+ "entropy": 7.637160062789917,
354
  "epoch": 0.14314928425357873,
355
+ "grad_norm": 1.0703125,
356
  "learning_rate": 8.7e-05,
357
+ "loss": 6.9519,
358
+ "mean_token_accuracy": 0.12585932165384292,
359
  "num_tokens": 2846905.0,
360
  "step": 175
361
  },
362
  {
363
+ "entropy": 7.6360515594482425,
364
  "epoch": 0.147239263803681,
365
+ "grad_norm": 1.03125,
366
  "learning_rate": 8.95e-05,
367
+ "loss": 6.953,
368
+ "mean_token_accuracy": 0.1275423213839531,
369
  "num_tokens": 2927747.0,
370
  "step": 180
371
  },
372
  {
373
+ "entropy": 7.516210556030273,
374
  "epoch": 0.15132924335378323,
375
+ "grad_norm": 0.828125,
376
  "learning_rate": 9.2e-05,
377
+ "loss": 6.8677,
378
+ "mean_token_accuracy": 0.13259214907884598,
379
  "num_tokens": 3008816.0,
380
  "step": 185
381
  },
382
  {
383
+ "entropy": 7.476730108261108,
384
  "epoch": 0.1554192229038855,
385
  "grad_norm": 1.140625,
386
  "learning_rate": 9.45e-05,
387
+ "loss": 6.8269,
388
+ "mean_token_accuracy": 0.13380919247865677,
389
  "num_tokens": 3090309.0,
390
  "step": 190
391
  },
392
  {
393
+ "entropy": 7.442372560501099,
394
  "epoch": 0.15950920245398773,
395
+ "grad_norm": 0.86328125,
396
  "learning_rate": 9.7e-05,
397
+ "loss": 6.8692,
398
+ "mean_token_accuracy": 0.13351524621248245,
399
  "num_tokens": 3171767.0,
400
  "step": 195
401
  },
402
  {
403
+ "entropy": 7.416834020614624,
404
  "epoch": 0.16359918200409,
405
+ "grad_norm": 0.90625,
406
  "learning_rate": 9.95e-05,
407
+ "loss": 6.8286,
408
+ "mean_token_accuracy": 0.13438718765974045,
409
  "num_tokens": 3253322.0,
410
  "step": 200
411
  },
412
  {
413
+ "entropy": 7.4071942329406735,
414
  "epoch": 0.16768916155419222,
415
+ "grad_norm": 1.109375,
416
  "learning_rate": 0.000102,
417
+ "loss": 6.8339,
418
+ "mean_token_accuracy": 0.13587063997983934,
419
  "num_tokens": 3334617.0,
420
  "step": 205
421
  },
422
  {
423
+ "entropy": 7.334885931015014,
424
  "epoch": 0.17177914110429449,
425
  "grad_norm": 1.0625,
426
  "learning_rate": 0.00010449999999999999,
427
+ "loss": 6.8066,
428
+ "mean_token_accuracy": 0.1350346751511097,
429
  "num_tokens": 3416224.0,
430
  "step": 210
431
  },
432
  {
433
+ "entropy": 7.294337224960327,
434
  "epoch": 0.17586912065439672,
435
+ "grad_norm": 0.859375,
436
  "learning_rate": 0.000107,
437
+ "loss": 6.7851,
438
+ "mean_token_accuracy": 0.13475466817617415,
439
  "num_tokens": 3497574.0,
440
  "step": 215
441
  },
442
  {
443
+ "entropy": 7.308300495147705,
444
  "epoch": 0.17995910020449898,
445
  "grad_norm": 1.0078125,
446
  "learning_rate": 0.0001095,
447
+ "loss": 6.7822,
448
+ "mean_token_accuracy": 0.13686141669750213,
449
  "num_tokens": 3578525.0,
450
  "step": 220
451
  },
452
  {
453
+ "entropy": 7.253078174591065,
454
  "epoch": 0.18404907975460122,
455
+ "grad_norm": 0.84375,
456
  "learning_rate": 0.000112,
457
+ "loss": 6.7709,
458
+ "mean_token_accuracy": 0.1378871127963066,
459
  "num_tokens": 3660181.0,
460
  "step": 225
461
  },
462
  {
463
+ "entropy": 7.265198373794556,
464
  "epoch": 0.18813905930470348,
465
+ "grad_norm": 0.765625,
466
  "learning_rate": 0.0001145,
467
+ "loss": 6.7519,
468
+ "mean_token_accuracy": 0.13652033507823944,
469
  "num_tokens": 3741809.0,
470
  "step": 230
471
  },
472
  {
473
+ "entropy": 7.23785982131958,
474
  "epoch": 0.19222903885480572,
475
+ "grad_norm": 0.98046875,
476
  "learning_rate": 0.00011700000000000001,
477
+ "loss": 6.7088,
478
+ "mean_token_accuracy": 0.13769972324371338,
479
  "num_tokens": 3823389.0,
480
  "step": 235
481
  },
482
  {
483
+ "entropy": 7.1915041446685795,
484
  "epoch": 0.19631901840490798,
485
  "grad_norm": 0.9140625,
486
  "learning_rate": 0.00011949999999999999,
487
+ "loss": 6.7275,
488
+ "mean_token_accuracy": 0.13537857830524444,
489
  "num_tokens": 3904714.0,
490
  "step": 240
491
  },
492
  {
493
+ "entropy": 7.14861159324646,
494
  "epoch": 0.20040899795501022,
495
+ "grad_norm": 1.046875,
496
  "learning_rate": 0.000122,
497
+ "loss": 6.6742,
498
+ "mean_token_accuracy": 0.1384981632232666,
499
  "num_tokens": 3986263.0,
500
  "step": 245
501
  },
502
  {
503
+ "entropy": 7.149033546447754,
504
  "epoch": 0.20449897750511248,
505
+ "grad_norm": 0.9375,
506
  "learning_rate": 0.0001245,
507
+ "loss": 6.6775,
508
+ "mean_token_accuracy": 0.13888683989644052,
509
  "num_tokens": 4067896.0,
510
  "step": 250
511
  },
512
  {
513
+ "entropy": 7.114649248123169,
514
  "epoch": 0.2085889570552147,
515
+ "grad_norm": 0.9375,
516
  "learning_rate": 0.000127,
517
+ "loss": 6.6754,
518
+ "mean_token_accuracy": 0.1410924032330513,
519
  "num_tokens": 4149422.0,
520
  "step": 255
521
  },
522
  {
523
+ "entropy": 7.072041988372803,
524
  "epoch": 0.21267893660531698,
525
+ "grad_norm": 0.84375,
526
  "learning_rate": 0.0001295,
527
+ "loss": 6.5748,
528
+ "mean_token_accuracy": 0.14269786030054094,
529
  "num_tokens": 4230807.0,
530
  "step": 260
531
  },
532
  {
533
+ "entropy": 7.01435227394104,
534
  "epoch": 0.2167689161554192,
535
+ "grad_norm": 1.0546875,
536
  "learning_rate": 0.000132,
537
+ "loss": 6.6012,
538
+ "mean_token_accuracy": 0.14459822773933412,
539
  "num_tokens": 4312299.0,
540
  "step": 265
541
  },
542
  {
543
+ "entropy": 7.081182289123535,
544
  "epoch": 0.22085889570552147,
545
+ "grad_norm": 0.8984375,
546
  "learning_rate": 0.00013450000000000002,
547
+ "loss": 6.6316,
548
+ "mean_token_accuracy": 0.1434426262974739,
549
  "num_tokens": 4393563.0,
550
  "step": 270
551
  },
552
  {
553
+ "entropy": 7.031969118118286,
554
  "epoch": 0.2249488752556237,
555
  "grad_norm": 0.8046875,
556
  "learning_rate": 0.00013700000000000002,
557
+ "loss": 6.6549,
558
+ "mean_token_accuracy": 0.13694730624556542,
559
  "num_tokens": 4474907.0,
560
  "step": 275
561
  },
562
  {
563
+ "entropy": 7.04373688697815,
564
  "epoch": 0.22903885480572597,
565
+ "grad_norm": 0.87890625,
566
  "learning_rate": 0.0001395,
567
  "loss": 6.5587,
568
+ "mean_token_accuracy": 0.14282752722501754,
569
  "num_tokens": 4556026.0,
570
  "step": 280
571
  },
572
  {
573
+ "entropy": 7.012068223953247,
574
  "epoch": 0.2331288343558282,
575
+ "grad_norm": 1.28125,
576
  "learning_rate": 0.00014199999999999998,
577
+ "loss": 6.5625,
578
+ "mean_token_accuracy": 0.142834110558033,
579
  "num_tokens": 4637280.0,
580
  "step": 285
581
  },
582
  {
583
+ "entropy": 6.986003494262695,
584
  "epoch": 0.23721881390593047,
585
  "grad_norm": 0.8671875,
586
  "learning_rate": 0.0001445,
587
+ "loss": 6.5745,
588
+ "mean_token_accuracy": 0.14551258087158203,
589
  "num_tokens": 4718900.0,
590
  "step": 290
591
  },
592
  {
593
+ "entropy": 6.98618369102478,
594
  "epoch": 0.24130879345603273,
595
+ "grad_norm": 0.90625,
596
  "learning_rate": 0.000147,
597
+ "loss": 6.5734,
598
+ "mean_token_accuracy": 0.142940554022789,
599
  "num_tokens": 4800507.0,
600
  "step": 295
601
  },
602
  {
603
+ "entropy": 6.899418592453003,
604
  "epoch": 0.24539877300613497,
605
  "grad_norm": 0.859375,
606
  "learning_rate": 0.0001495,
607
+ "loss": 6.5152,
608
+ "mean_token_accuracy": 0.14577620029449462,
609
  "num_tokens": 4881971.0,
610
  "step": 300
611
  },
612
  {
613
+ "entropy": 6.931239128112793,
614
  "epoch": 0.24948875255623723,
615
+ "grad_norm": 0.91015625,
616
  "learning_rate": 0.000152,
617
  "loss": 6.5192,
618
+ "mean_token_accuracy": 0.1479423984885216,
619
  "num_tokens": 4963516.0,
620
  "step": 305
621
  },
622
  {
623
+ "entropy": 6.870612716674804,
624
  "epoch": 0.25357873210633947,
625
+ "grad_norm": 0.96875,
626
  "learning_rate": 0.00015450000000000001,
627
+ "loss": 6.4973,
628
+ "mean_token_accuracy": 0.14800097942352294,
629
  "num_tokens": 5045190.0,
630
  "step": 310
631
  },
632
  {
633
+ "entropy": 6.911778402328491,
634
  "epoch": 0.25766871165644173,
635
+ "grad_norm": 0.9375,
636
  "learning_rate": 0.000157,
637
+ "loss": 6.5103,
638
+ "mean_token_accuracy": 0.14730968773365022,
639
  "num_tokens": 5126730.0,
640
  "step": 315
641
  },
642
  {
643
+ "entropy": 6.862739562988281,
644
  "epoch": 0.261758691206544,
645
+ "grad_norm": 0.86328125,
646
  "learning_rate": 0.0001595,
647
  "loss": 6.5129,
648
+ "mean_token_accuracy": 0.14664446264505387,
649
  "num_tokens": 5207888.0,
650
  "step": 320
651
  },
652
  {
653
+ "entropy": 6.849811601638794,
654
  "epoch": 0.2658486707566462,
655
+ "grad_norm": 0.81640625,
656
  "learning_rate": 0.000162,
657
+ "loss": 6.437,
658
+ "mean_token_accuracy": 0.1522682175040245,
659
  "num_tokens": 5289484.0,
660
  "step": 325
661
  },
662
  {
663
+ "entropy": 6.818774795532226,
664
  "epoch": 0.26993865030674846,
665
  "grad_norm": 0.75,
666
  "learning_rate": 0.00016450000000000001,
667
+ "loss": 6.405,
668
+ "mean_token_accuracy": 0.15261074155569077,
669
  "num_tokens": 5371086.0,
670
  "step": 330
671
  },
672
  {
673
+ "entropy": 6.868287801742554,
674
  "epoch": 0.2740286298568507,
675
  "grad_norm": 0.77734375,
676
  "learning_rate": 0.00016700000000000002,
677
+ "loss": 6.5087,
678
+ "mean_token_accuracy": 0.1447703406214714,
679
  "num_tokens": 5452574.0,
680
  "step": 335
681
  },
682
  {
683
+ "entropy": 6.7703859329223635,
684
  "epoch": 0.278118609406953,
685
  "grad_norm": 0.89453125,
686
  "learning_rate": 0.00016950000000000003,
687
+ "loss": 6.4103,
688
+ "mean_token_accuracy": 0.1545592203736305,
689
  "num_tokens": 5534091.0,
690
  "step": 340
691
  },
692
  {
693
+ "entropy": 6.787615823745727,
694
  "epoch": 0.2822085889570552,
695
+ "grad_norm": 0.87109375,
696
  "learning_rate": 0.00017199999999999998,
697
+ "loss": 6.4559,
698
+ "mean_token_accuracy": 0.14936767816543578,
699
  "num_tokens": 5614709.0,
700
  "step": 345
701
  },
702
  {
703
+ "entropy": 6.780069017410279,
704
  "epoch": 0.28629856850715746,
705
+ "grad_norm": 0.96484375,
706
  "learning_rate": 0.00017449999999999999,
707
+ "loss": 6.3975,
708
+ "mean_token_accuracy": 0.15142018646001815,
709
  "num_tokens": 5695945.0,
710
  "step": 350
711
  },
712
  {
713
+ "entropy": 6.7227113246917725,
714
  "epoch": 0.2903885480572597,
715
+ "grad_norm": 0.890625,
716
  "learning_rate": 0.000177,
717
+ "loss": 6.3873,
718
+ "mean_token_accuracy": 0.15569809824228287,
719
  "num_tokens": 5776894.0,
720
  "step": 355
721
  },
722
  {
723
+ "entropy": 6.681255960464478,
724
  "epoch": 0.294478527607362,
725
+ "grad_norm": 0.83984375,
726
  "learning_rate": 0.0001795,
727
+ "loss": 6.3418,
728
+ "mean_token_accuracy": 0.1554790660738945,
729
  "num_tokens": 5858426.0,
730
  "step": 360
731
  },
732
  {
733
+ "entropy": 6.75174880027771,
734
  "epoch": 0.2985685071574642,
735
+ "grad_norm": 0.7890625,
736
  "learning_rate": 0.000182,
737
+ "loss": 6.435,
738
+ "mean_token_accuracy": 0.1517605572938919,
739
  "num_tokens": 5939990.0,
740
  "step": 365
741
  },
742
  {
743
+ "entropy": 6.69475622177124,
744
  "epoch": 0.30265848670756645,
745
+ "grad_norm": 0.7578125,
746
  "learning_rate": 0.0001845,
747
  "loss": 6.3261,
748
+ "mean_token_accuracy": 0.15980565249919892,
749
  "num_tokens": 6021209.0,
750
  "step": 370
751
  },
752
  {
753
+ "entropy": 6.72758207321167,
754
  "epoch": 0.3067484662576687,
755
  "grad_norm": 0.89453125,
756
  "learning_rate": 0.000187,
757
  "loss": 6.3882,
758
+ "mean_token_accuracy": 0.15352147072553635,
759
  "num_tokens": 6102085.0,
760
  "step": 375
761
  },
762
  {
763
+ "entropy": 6.668760585784912,
764
  "epoch": 0.310838445807771,
765
+ "grad_norm": 0.90234375,
766
  "learning_rate": 0.0001895,
767
+ "loss": 6.3174,
768
+ "mean_token_accuracy": 0.1569987043738365,
769
  "num_tokens": 6183631.0,
770
  "step": 380
771
  },
772
  {
773
+ "entropy": 6.686910724639892,
774
  "epoch": 0.3149284253578732,
775
  "grad_norm": 0.72265625,
776
  "learning_rate": 0.000192,
777
+ "loss": 6.3469,
778
+ "mean_token_accuracy": 0.15672450959682466,
779
  "num_tokens": 6264174.0,
780
  "step": 385
781
  },
782
  {
783
+ "entropy": 6.64973726272583,
784
  "epoch": 0.31901840490797545,
785
  "grad_norm": 0.87109375,
786
  "learning_rate": 0.0001945,
787
+ "loss": 6.2917,
788
+ "mean_token_accuracy": 0.1605835884809494,
789
  "num_tokens": 6345616.0,
790
  "step": 390
791
  },
792
  {
793
+ "entropy": 6.6427412033081055,
794
  "epoch": 0.3231083844580777,
795
+ "grad_norm": 0.81640625,
796
  "learning_rate": 0.00019700000000000002,
797
  "loss": 6.2962,
798
+ "mean_token_accuracy": 0.15768682956695557,
799
  "num_tokens": 6426371.0,
800
  "step": 395
801
  },
802
  {
803
+ "entropy": 6.668876314163208,
804
  "epoch": 0.32719836400818,
805
  "grad_norm": 0.95703125,
806
  "learning_rate": 0.00019950000000000002,
807
+ "loss": 6.349,
808
+ "mean_token_accuracy": 0.15605812668800353,
809
  "num_tokens": 6508017.0,
810
  "step": 400
811
  },
812
  {
813
+ "entropy": 6.626045894622803,
814
  "epoch": 0.3312883435582822,
815
+ "grad_norm": 1.0234375,
816
  "learning_rate": 0.000202,
817
  "loss": 6.3398,
818
+ "mean_token_accuracy": 0.15648607313632965,
819
  "num_tokens": 6589627.0,
820
  "step": 405
821
  },
822
  {
823
+ "entropy": 6.614638900756836,
824
  "epoch": 0.33537832310838445,
825
  "grad_norm": 0.90625,
826
  "learning_rate": 0.00020449999999999998,
827
  "loss": 6.2965,
828
+ "mean_token_accuracy": 0.16072938144207,
829
  "num_tokens": 6670965.0,
830
  "step": 410
831
  },
832
  {
833
+ "entropy": 6.662922954559326,
834
  "epoch": 0.3394683026584867,
835
+ "grad_norm": 0.92578125,
836
  "learning_rate": 0.000207,
837
  "loss": 6.3392,
838
+ "mean_token_accuracy": 0.15543674528598786,
839
  "num_tokens": 6752603.0,
840
  "step": 415
841
  },
842
  {
843
+ "entropy": 6.557052993774414,
844
  "epoch": 0.34355828220858897,
845
+ "grad_norm": 0.89453125,
846
  "learning_rate": 0.0002095,
847
+ "loss": 6.2425,
848
+ "mean_token_accuracy": 0.1611531674861908,
849
  "num_tokens": 6833529.0,
850
  "step": 420
851
  },
852
  {
853
+ "entropy": 6.608387994766235,
854
  "epoch": 0.3476482617586912,
855
+ "grad_norm": 0.828125,
856
  "learning_rate": 0.000212,
857
+ "loss": 6.2616,
858
+ "mean_token_accuracy": 0.1630495846271515,
859
  "num_tokens": 6915141.0,
860
  "step": 425
861
  },
862
  {
863
+ "entropy": 6.538517665863037,
864
  "epoch": 0.35173824130879344,
865
+ "grad_norm": 0.81640625,
866
  "learning_rate": 0.0002145,
867
  "loss": 6.2681,
868
+ "mean_token_accuracy": 0.158408260345459,
869
  "num_tokens": 6996145.0,
870
  "step": 430
871
  },
872
  {
873
+ "entropy": 6.586181640625,
874
  "epoch": 0.3558282208588957,
875
  "grad_norm": 0.74609375,
876
  "learning_rate": 0.00021700000000000002,
877
+ "loss": 6.2616,
878
+ "mean_token_accuracy": 0.1599885493516922,
879
  "num_tokens": 7077692.0,
880
  "step": 435
881
  },
882
  {
883
+ "entropy": 6.546446752548218,
884
  "epoch": 0.35991820040899797,
885
+ "grad_norm": 0.8125,
886
  "learning_rate": 0.0002195,
887
+ "loss": 6.2648,
888
+ "mean_token_accuracy": 0.16150236129760742,
889
  "num_tokens": 7159364.0,
890
  "step": 440
891
  },
892
  {
893
+ "entropy": 6.547263431549072,
894
  "epoch": 0.36400817995910023,
895
  "grad_norm": 0.76171875,
896
  "learning_rate": 0.000222,
897
+ "loss": 6.23,
898
+ "mean_token_accuracy": 0.16269229352474213,
899
  "num_tokens": 7240330.0,
900
  "step": 445
901
  },
902
  {
903
+ "entropy": 6.473912286758423,
904
  "epoch": 0.36809815950920244,
905
+ "grad_norm": 0.87890625,
906
  "learning_rate": 0.0002245,
907
+ "loss": 6.1576,
908
+ "mean_token_accuracy": 0.16461296379566193,
909
  "num_tokens": 7320583.0,
910
  "step": 450
911
  },
912
  {
913
+ "entropy": 6.49521541595459,
914
  "epoch": 0.3721881390593047,
915
+ "grad_norm": 0.75,
916
  "learning_rate": 0.00022700000000000002,
917
+ "loss": 6.2006,
918
+ "mean_token_accuracy": 0.16421556025743483,
919
  "num_tokens": 7401685.0,
920
  "step": 455
921
  },
922
  {
923
+ "entropy": 6.5050506591796875,
924
  "epoch": 0.37627811860940696,
925
+ "grad_norm": 0.78515625,
926
  "learning_rate": 0.00022950000000000002,
927
  "loss": 6.1459,
928
+ "mean_token_accuracy": 0.1676486998796463,
929
  "num_tokens": 7482873.0,
930
  "step": 460
931
  },
932
  {
933
+ "entropy": 6.463202667236328,
934
  "epoch": 0.3803680981595092,
935
+ "grad_norm": 0.73828125,
936
  "learning_rate": 0.00023200000000000003,
937
  "loss": 6.157,
938
+ "mean_token_accuracy": 0.1651605486869812,
939
  "num_tokens": 7564543.0,
940
  "step": 465
941
  },
942
  {
943
+ "entropy": 6.4412964344024655,
944
  "epoch": 0.38445807770961143,
945
+ "grad_norm": 0.83203125,
946
  "learning_rate": 0.00023449999999999998,
947
+ "loss": 6.186,
948
+ "mean_token_accuracy": 0.16442248970270157,
949
  "num_tokens": 7645311.0,
950
  "step": 470
951
  },
952
  {
953
+ "entropy": 6.522923421859741,
954
  "epoch": 0.3885480572597137,
955
+ "grad_norm": 0.87109375,
956
  "learning_rate": 0.000237,
957
+ "loss": 6.2154,
958
+ "mean_token_accuracy": 0.16138581335544586,
959
  "num_tokens": 7726050.0,
960
  "step": 475
961
  },
962
  {
963
+ "entropy": 6.4519061088562015,
964
  "epoch": 0.39263803680981596,
965
  "grad_norm": 0.765625,
966
  "learning_rate": 0.0002395,
967
+ "loss": 6.1928,
968
+ "mean_token_accuracy": 0.1644761323928833,
969
  "num_tokens": 7807452.0,
970
  "step": 480
971
  },
972
  {
973
+ "entropy": 6.45891547203064,
974
  "epoch": 0.3967280163599182,
975
+ "grad_norm": 0.7265625,
976
  "learning_rate": 0.000242,
977
+ "loss": 6.1734,
978
+ "mean_token_accuracy": 0.16323913782835006,
979
  "num_tokens": 7889159.0,
980
  "step": 485
981
  },
982
  {
983
+ "entropy": 6.512608671188355,
984
  "epoch": 0.40081799591002043,
985
  "grad_norm": 0.765625,
986
  "learning_rate": 0.0002445,
987
+ "loss": 6.224,
988
+ "mean_token_accuracy": 0.16140950471162796,
989
  "num_tokens": 7970803.0,
990
  "step": 490
991
  },
992
  {
993
+ "entropy": 6.389305877685547,
994
  "epoch": 0.4049079754601227,
995
  "grad_norm": 0.734375,
996
  "learning_rate": 0.000247,
997
+ "loss": 6.0861,
998
+ "mean_token_accuracy": 0.17114942967891694,
999
  "num_tokens": 8051777.0,
1000
  "step": 495
1001
  },
1002
  {
1003
+ "entropy": 6.496857643127441,
1004
  "epoch": 0.40899795501022496,
1005
  "grad_norm": 0.79296875,
1006
  "learning_rate": 0.0002495,
1007
+ "loss": 6.1937,
1008
+ "mean_token_accuracy": 0.16533919274806977,
1009
  "num_tokens": 8133204.0,
1010
  "step": 500
1011
  },
1012
  {
1013
  "epoch": 0.40899795501022496,
1014
+ "eval_entropy": 6.272721568589965,
1015
+ "eval_loss": 6.081885814666748,
1016
+ "eval_mean_token_accuracy": 0.1713385213756193,
1017
  "eval_num_tokens": 8133204.0,
1018
+ "eval_runtime": 66.7776,
1019
+ "eval_samples_per_second": 61.952,
1020
+ "eval_steps_per_second": 7.757,
1021
  "step": 500
1022
  }
1023
  ],
checkpoint-500/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:64eb4021ab14c58c80e547ae8f0605e3e0fcbb6a34b9609a3d3fbf345d405491
3
  size 5880
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de941e54df8fba69544c603308a97f7145f30276328bd1a1169098c9a5ab7dc1
3
  size 5880