File size: 16,316 Bytes
eef2cee
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9962034927866363,
  "eval_steps": 2000000,
  "global_step": 164,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 803.0022583007812,
      "epoch": 0.006074411541381929,
      "grad_norm": 0.14656181705800808,
      "learning_rate": 5.88235294117647e-08,
      "loss": 0.0479,
      "num_tokens": 918402.0,
      "reward": 0.9815848618745804,
      "reward_std": 0.23756355978548527,
      "rewards/accuracy_reward": 0.4933035746216774,
      "rewards/format_reward": 0.9765624925494194,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 732.8228569030762,
      "epoch": 0.030372057706909643,
      "grad_norm": 0.1360498469331386,
      "learning_rate": 2.941176470588235e-07,
      "loss": 0.0275,
      "num_tokens": 4253375.0,
      "reward": 1.0891462452709675,
      "reward_std": 0.2294587183278054,
      "rewards/accuracy_reward": 0.5954241026192904,
      "rewards/format_reward": 0.9874441903084517,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 733.9051712036132,
      "epoch": 0.060744115413819286,
      "grad_norm": 2.020390330382682,
      "learning_rate": 5.88235294117647e-07,
      "loss": 0.0394,
      "num_tokens": 8424966.0,
      "reward": 1.074776826798916,
      "reward_std": 0.2325986440293491,
      "rewards/accuracy_reward": 0.5812499992549419,
      "rewards/format_reward": 0.9870535641908645,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 730.850700378418,
      "epoch": 0.09111617312072894,
      "grad_norm": 0.6485429367394874,
      "learning_rate": 8.823529411764705e-07,
      "loss": 0.0319,
      "num_tokens": 12651361.0,
      "reward": 1.068080399930477,
      "reward_std": 0.22116196975111962,
      "rewards/accuracy_reward": 0.5736607149243355,
      "rewards/format_reward": 0.988839277625084,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 749.2339599609375,
      "epoch": 0.12148823082763857,
      "grad_norm": 0.15121241250002537,
      "learning_rate": 9.989726963751682e-07,
      "loss": 0.0412,
      "num_tokens": 16924113.0,
      "reward": 1.0832589775323869,
      "reward_std": 0.24419224373996257,
      "rewards/accuracy_reward": 0.590625,
      "rewards/format_reward": 0.9852678462862968,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 733.0969100952149,
      "epoch": 0.15186028853454822,
      "grad_norm": 0.1625406073177716,
      "learning_rate": 9.927100106776212e-07,
      "loss": 0.0385,
      "num_tokens": 21115051.0,
      "reward": 1.1054687976837159,
      "reward_std": 0.22408477105200292,
      "rewards/accuracy_reward": 0.6136160686612129,
      "rewards/format_reward": 0.9837053477764129,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 720.6406631469727,
      "epoch": 0.18223234624145787,
      "grad_norm": 2.666467566757727,
      "learning_rate": 9.808267184205181e-07,
      "loss": 0.0239,
      "num_tokens": 25258505.0,
      "reward": 1.0918527334928512,
      "reward_std": 0.22341929338872432,
      "rewards/accuracy_reward": 0.5984375,
      "rewards/format_reward": 0.9868303462862968,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 701.581282043457,
      "epoch": 0.2126044039483675,
      "grad_norm": 0.2580879090307691,
      "learning_rate": 9.634583786730108e-07,
      "loss": 0.0261,
      "num_tokens": 29330221.0,
      "reward": 1.1060268327593803,
      "reward_std": 0.21581739820539952,
      "rewards/accuracy_reward": 0.613616070151329,
      "rewards/format_reward": 0.984821417927742,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 707.9558319091797,
      "epoch": 0.24297646165527714,
      "grad_norm": 0.8886692354776214,
      "learning_rate": 9.408031213740044e-07,
      "loss": 0.0272,
      "num_tokens": 33442287.0,
      "reward": 1.0679688140749932,
      "reward_std": 0.21454464346170427,
      "rewards/accuracy_reward": 0.5720982141792774,
      "rewards/format_reward": 0.9917410612106323,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 699.1395431518555,
      "epoch": 0.2733485193621868,
      "grad_norm": 0.1839147765074166,
      "learning_rate": 9.131193871579974e-07,
      "loss": 0.0273,
      "num_tokens": 37510936.0,
      "reward": 1.0806920111179352,
      "reward_std": 0.20978465098887683,
      "rewards/accuracy_reward": 0.5868303567171097,
      "rewards/format_reward": 0.987723208963871,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 688.0076187133789,
      "epoch": 0.30372057706909644,
      "grad_norm": 2.39827459380332,
      "learning_rate": 8.807229791845671e-07,
      "loss": 0.0254,
      "num_tokens": 41532546.0,
      "reward": 1.0955357611179353,
      "reward_std": 0.2291064240038395,
      "rewards/accuracy_reward": 0.6026785708963871,
      "rewards/format_reward": 0.9857142746448517,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 705.5748062133789,
      "epoch": 0.3340926347760061,
      "grad_norm": 6.336128416350602,
      "learning_rate": 8.439834606028593e-07,
      "loss": 0.0312,
      "num_tokens": 45638753.0,
      "reward": 1.0700893342494964,
      "reward_std": 0.22525846362113952,
      "rewards/accuracy_reward": 0.5787946447730065,
      "rewards/format_reward": 0.9825892791152,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 699.0966857910156,
      "epoch": 0.36446469248291574,
      "grad_norm": 2.8773101008786597,
      "learning_rate": 8.033199387471276e-07,
      "loss": 0.0257,
      "num_tokens": 49682178.0,
      "reward": 1.0860491678118707,
      "reward_std": 0.19614009652286768,
      "rewards/accuracy_reward": 0.5910714276134967,
      "rewards/format_reward": 0.9899553537368775,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 676.4727966308594,
      "epoch": 0.39483675018982534,
      "grad_norm": 0.1564093879360907,
      "learning_rate": 7.591962841552626e-07,
      "loss": 0.0148,
      "num_tokens": 53632632.0,
      "reward": 1.104575951397419,
      "reward_std": 0.1936511306092143,
      "rewards/accuracy_reward": 0.6069196410477161,
      "rewards/format_reward": 0.9953124940395355,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 693.0926666259766,
      "epoch": 0.425208807896735,
      "grad_norm": 0.21433648094895766,
      "learning_rate": 7.121158389495185e-07,
      "loss": 0.0179,
      "num_tokens": 57653679.0,
      "reward": 1.096875049173832,
      "reward_std": 0.19877330139279364,
      "rewards/accuracy_reward": 0.6008928552269935,
      "rewards/format_reward": 0.9919642791152,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 677.7578414916992,
      "epoch": 0.45558086560364464,
      "grad_norm": 0.13423679492808266,
      "learning_rate": 6.626156749437736e-07,
      "loss": 0.023,
      "num_tokens": 61583418.0,
      "reward": 1.0986607670783997,
      "reward_std": 0.19030132256448268,
      "rewards/accuracy_reward": 0.6033482141792774,
      "rewards/format_reward": 0.9906249925494194,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 677.3216796875,
      "epoch": 0.4859529233105543,
      "grad_norm": 0.12374303041671109,
      "learning_rate": 6.112604669781572e-07,
      "loss": 0.0139,
      "num_tokens": 65523531.0,
      "reward": 1.0838170185685159,
      "reward_std": 0.20196059457957743,
      "rewards/accuracy_reward": 0.5859375014901161,
      "rewards/format_reward": 0.9957589223980904,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 661.0935562133789,
      "epoch": 0.5163249810174639,
      "grad_norm": 0.24633881049994122,
      "learning_rate": 5.586360513712009e-07,
      "loss": 0.0146,
      "num_tokens": 69412862.0,
      "reward": 1.1289062976837159,
      "reward_std": 0.19050537403672935,
      "rewards/accuracy_reward": 0.6308035746216774,
      "rewards/format_reward": 0.9962053522467613,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 684.6299407958984,
      "epoch": 0.5466970387243736,
      "grad_norm": 0.20823664365461947,
      "learning_rate": 5.053427429716866e-07,
      "loss": 0.0183,
      "num_tokens": 73405012.0,
      "reward": 1.127790230512619,
      "reward_std": 0.1836160296574235,
      "rewards/accuracy_reward": 0.6328468464314938,
      "rewards/format_reward": 0.9921874940395355,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 679.512525177002,
      "epoch": 0.5770690964312832,
      "grad_norm": 0.17725373837889702,
      "learning_rate": 4.519884870461591e-07,
      "loss": 0.013,
      "num_tokens": 77357644.0,
      "reward": 1.1372768431901932,
      "reward_std": 0.1831109957769513,
      "rewards/accuracy_reward": 0.6390624985098838,
      "rewards/format_reward": 0.9964285656809807,
      "step": 95
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 691.0065063476562,
      "epoch": 0.6074411541381929,
      "grad_norm": 0.18903103595690637,
      "learning_rate": 3.991819241221835e-07,
      "loss": 0.015,
      "num_tokens": 81377353.0,
      "reward": 1.1179687947034835,
      "reward_std": 0.18047410659492016,
      "rewards/accuracy_reward": 0.6209821425378322,
      "rewards/format_reward": 0.9939732044935227,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 678.3850769042969,
      "epoch": 0.6378132118451025,
      "grad_norm": 0.20779542924679995,
      "learning_rate": 3.4752544690038643e-07,
      "loss": 0.0112,
      "num_tokens": 85350694.0,
      "reward": 1.1294643417000771,
      "reward_std": 0.18618108071386813,
      "rewards/accuracy_reward": 0.6323660708963871,
      "rewards/format_reward": 0.9941964238882065,
      "step": 105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 688.9138717651367,
      "epoch": 0.6681852695520122,
      "grad_norm": 0.2676944642662668,
      "learning_rate": 2.976083284388031e-07,
      "loss": 0.0163,
      "num_tokens": 89382916.0,
      "reward": 1.101339329779148,
      "reward_std": 0.1832605952396989,
      "rewards/accuracy_reward": 0.6040178589522839,
      "rewards/format_reward": 0.9946428507566452,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 709.0390884399415,
      "epoch": 0.6985573272589218,
      "grad_norm": 0.1723246791980809,
      "learning_rate": 2.500000000000001e-07,
      "loss": 0.0124,
      "num_tokens": 93498363.0,
      "reward": 1.123660759627819,
      "reward_std": 0.18700322844088077,
      "rewards/accuracy_reward": 0.6258928567171097,
      "rewards/format_reward": 0.9955357074737549,
      "step": 115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 674.1076217651367,
      "epoch": 0.7289293849658315,
      "grad_norm": 0.16904917847238007,
      "learning_rate": 2.0524355524417015e-07,
      "loss": 0.0146,
      "num_tokens": 97437677.0,
      "reward": 1.131361649930477,
      "reward_std": 0.19161452651023864,
      "rewards/accuracy_reward": 0.6328125,
      "rewards/format_reward": 0.997098208963871,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 682.0279342651368,
      "epoch": 0.7593014426727411,
      "grad_norm": 0.17946193842059094,
      "learning_rate": 1.6384955486934154e-07,
      "loss": 0.0112,
      "num_tokens": 101433570.0,
      "reward": 1.1439732566475869,
      "reward_std": 0.1751913372427225,
      "rewards/accuracy_reward": 0.6462053626775741,
      "rewards/format_reward": 0.9955357074737549,
      "step": 125
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 695.6339569091797,
      "epoch": 0.7896735003796507,
      "grad_norm": 0.1905658306829658,
      "learning_rate": 1.262902023724824e-07,
      "loss": 0.0081,
      "num_tokens": 105473282.0,
      "reward": 1.1212054073810578,
      "reward_std": 0.17985087335109712,
      "rewards/accuracy_reward": 0.6225446417927742,
      "rewards/format_reward": 0.9973214253783226,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 689.9069519042969,
      "epoch": 0.8200455580865603,
      "grad_norm": 0.18740313988650945,
      "learning_rate": 9.299395737170757e-08,
      "loss": 0.0183,
      "num_tokens": 109523129.0,
      "reward": 1.0936384439468383,
      "reward_std": 0.20308975782245398,
      "rewards/accuracy_reward": 0.5962053589522839,
      "rewards/format_reward": 0.9948660627007484,
      "step": 135
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 689.1723495483399,
      "epoch": 0.85041761579347,
      "grad_norm": 0.17643493300658672,
      "learning_rate": 6.43406479383053e-08,
      "loss": 0.0089,
      "num_tokens": 113562189.0,
      "reward": 1.1333705842494965,
      "reward_std": 0.18020046018064023,
      "rewards/accuracy_reward": 0.6352678582072258,
      "rewards/format_reward": 0.9962053537368775,
      "step": 140
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 673.6042694091797,
      "epoch": 0.8807896735003796,
      "grad_norm": 3.201259451192229,
      "learning_rate": 4.065713769482082e-08,
      "loss": 0.0138,
      "num_tokens": 117482928.0,
      "reward": 1.1319196924567223,
      "reward_std": 0.1873009530827403,
      "rewards/accuracy_reward": 0.6339285716414451,
      "rewards/format_reward": 0.9959821388125419,
      "step": 145
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 687.8794952392578,
      "epoch": 0.9111617312072893,
      "grad_norm": 2.7680554916216296,
      "learning_rate": 2.2213597106929605e-08,
      "loss": 0.0152,
      "num_tokens": 121487468.0,
      "reward": 1.1334821954369545,
      "reward_std": 0.1863085398450494,
      "rewards/accuracy_reward": 0.634821429848671,
      "rewards/format_reward": 0.9973214238882064,
      "step": 150
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 691.2263702392578,
      "epoch": 0.9415337889141989,
      "grad_norm": 0.16790447183243948,
      "learning_rate": 9.22042150446728e-09,
      "loss": 0.0089,
      "num_tokens": 125492538.0,
      "reward": 1.1498884424567222,
      "reward_std": 0.1872717458754778,
      "rewards/accuracy_reward": 0.6511160723865033,
      "rewards/format_reward": 0.997544638812542,
      "step": 155
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 689.0906539916992,
      "epoch": 0.9719058466211086,
      "grad_norm": 0.8754711405794506,
      "learning_rate": 1.8258309893965374e-09,
      "loss": 0.0103,
      "num_tokens": 129525872.0,
      "reward": 1.152567020058632,
      "reward_std": 0.189371613971889,
      "rewards/accuracy_reward": 0.6555803567171097,
      "rewards/format_reward": 0.9939732134342194,
      "step": 160
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 673.1839790344238,
      "epoch": 0.9962034927866363,
      "num_tokens": 132694223.0,
      "reward": 1.1247210260480642,
      "reward_std": 0.1711470059817657,
      "rewards/accuracy_reward": 0.6278545688837767,
      "rewards/format_reward": 0.997209819033742,
      "step": 164,
      "total_flos": 0.0,
      "train_loss": 0.019764071750659043,
      "train_runtime": 31261.7596,
      "train_samples_per_second": 0.59,
      "train_steps_per_second": 0.005
    }
  ],
  "logging_steps": 5,
  "max_steps": 164,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}