SivilTaram commited on
Commit
fc9c587
·
verified ·
1 Parent(s): 3ce3730

Upload checkpoint-1000/trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. checkpoint-1000/trainer_state.json +773 -0
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,773 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.842468908337172,
5
+ "eval_steps": 200,
6
+ "global_step": 1000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.018424689083371717,
13
+ "grad_norm": 12.213397932615418,
14
+ "learning_rate": 1.2269938650306748e-07,
15
+ "loss": 0.7896,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.036849378166743434,
20
+ "grad_norm": 12.364034738650293,
21
+ "learning_rate": 2.4539877300613496e-07,
22
+ "loss": 0.7591,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.055274067250115154,
27
+ "grad_norm": 3.33541668449751,
28
+ "learning_rate": 3.6809815950920245e-07,
29
+ "loss": 0.68,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.07369875633348687,
34
+ "grad_norm": 1.8325651149958935,
35
+ "learning_rate": 4.907975460122699e-07,
36
+ "loss": 0.6151,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.09212344541685859,
41
+ "grad_norm": 1.1850978860914436,
42
+ "learning_rate": 6.134969325153374e-07,
43
+ "loss": 0.5901,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.11054813450023031,
48
+ "grad_norm": 1.0238439765458438,
49
+ "learning_rate": 7.361963190184049e-07,
50
+ "loss": 0.5866,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.12897282358360201,
55
+ "grad_norm": 1.0859904158012907,
56
+ "learning_rate": 8.588957055214724e-07,
57
+ "loss": 0.5872,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.14739751266697373,
62
+ "grad_norm": 0.9445030962726604,
63
+ "learning_rate": 9.815950920245399e-07,
64
+ "loss": 0.5908,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.16582220175034545,
69
+ "grad_norm": 0.902903286685186,
70
+ "learning_rate": 1.1042944785276073e-06,
71
+ "loss": 0.585,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.18424689083371718,
76
+ "grad_norm": 0.9406769634542373,
77
+ "learning_rate": 1.2269938650306748e-06,
78
+ "loss": 0.5759,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.2026715799170889,
83
+ "grad_norm": 0.9379530599920557,
84
+ "learning_rate": 1.3496932515337423e-06,
85
+ "loss": 0.5782,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.22109626900046062,
90
+ "grad_norm": 0.9806028234388734,
91
+ "learning_rate": 1.4723926380368098e-06,
92
+ "loss": 0.5599,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.23952095808383234,
97
+ "grad_norm": 1.0489169076559612,
98
+ "learning_rate": 1.595092024539877e-06,
99
+ "loss": 0.5482,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.25794564716720403,
104
+ "grad_norm": 1.1182127736456162,
105
+ "learning_rate": 1.7177914110429447e-06,
106
+ "loss": 0.548,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.2763703362505758,
111
+ "grad_norm": 1.0633010115096506,
112
+ "learning_rate": 1.8404907975460122e-06,
113
+ "loss": 0.5378,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.29479502533394747,
118
+ "grad_norm": 0.9818550023254589,
119
+ "learning_rate": 1.9631901840490797e-06,
120
+ "loss": 0.5336,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.3132197144173192,
125
+ "grad_norm": 1.1126051299502253,
126
+ "learning_rate": 1.9998870284726965e-06,
127
+ "loss": 0.5265,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.3316444035006909,
132
+ "grad_norm": 1.2009782036779326,
133
+ "learning_rate": 1.9993337599895922e-06,
134
+ "loss": 0.5248,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.35006909258406266,
139
+ "grad_norm": 1.0946468762708952,
140
+ "learning_rate": 1.9983196994710604e-06,
141
+ "loss": 0.5196,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.36849378166743435,
146
+ "grad_norm": 1.0764331817764232,
147
+ "learning_rate": 1.996845314499634e-06,
148
+ "loss": 0.5176,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.36849378166743435,
153
+ "eval_loss": 0.48491016030311584,
154
+ "eval_runtime": 420.0216,
155
+ "eval_samples_per_second": 18.373,
156
+ "eval_steps_per_second": 2.298,
157
+ "step": 200
158
+ },
159
+ {
160
+ "epoch": 0.3869184707508061,
161
+ "grad_norm": 1.1132997861385496,
162
+ "learning_rate": 1.9949112849131003e-06,
163
+ "loss": 0.5247,
164
+ "step": 210
165
+ },
166
+ {
167
+ "epoch": 0.4053431598341778,
168
+ "grad_norm": 1.2031973800065288,
169
+ "learning_rate": 1.9925185024910275e-06,
170
+ "loss": 0.4998,
171
+ "step": 220
172
+ },
173
+ {
174
+ "epoch": 0.42376784891754954,
175
+ "grad_norm": 1.3213263203662036,
176
+ "learning_rate": 1.9896680705435688e-06,
177
+ "loss": 0.4837,
178
+ "step": 230
179
+ },
180
+ {
181
+ "epoch": 0.44219253800092123,
182
+ "grad_norm": 1.2768991437360866,
183
+ "learning_rate": 1.9863613034027223e-06,
184
+ "loss": 0.5022,
185
+ "step": 240
186
+ },
187
+ {
188
+ "epoch": 0.460617227084293,
189
+ "grad_norm": 1.125849818584755,
190
+ "learning_rate": 1.9825997258162984e-06,
191
+ "loss": 0.4894,
192
+ "step": 250
193
+ },
194
+ {
195
+ "epoch": 0.47904191616766467,
196
+ "grad_norm": 1.2472284470521988,
197
+ "learning_rate": 1.9783850722448565e-06,
198
+ "loss": 0.4858,
199
+ "step": 260
200
+ },
201
+ {
202
+ "epoch": 0.49746660525103636,
203
+ "grad_norm": 1.2644974012951764,
204
+ "learning_rate": 1.9737192860619475e-06,
205
+ "loss": 0.4777,
206
+ "step": 270
207
+ },
208
+ {
209
+ "epoch": 0.5158912943344081,
210
+ "grad_norm": 1.3849217540568468,
211
+ "learning_rate": 1.9686045186580258e-06,
212
+ "loss": 0.4649,
213
+ "step": 280
214
+ },
215
+ {
216
+ "epoch": 0.5343159834177799,
217
+ "grad_norm": 1.3250529799695097,
218
+ "learning_rate": 1.9630431284484445e-06,
219
+ "loss": 0.4611,
220
+ "step": 290
221
+ },
222
+ {
223
+ "epoch": 0.5527406725011516,
224
+ "grad_norm": 1.2452480422858616,
225
+ "learning_rate": 1.9570376797859937e-06,
226
+ "loss": 0.4656,
227
+ "step": 300
228
+ },
229
+ {
230
+ "epoch": 0.5711653615845232,
231
+ "grad_norm": 1.3984211841107805,
232
+ "learning_rate": 1.9505909417784754e-06,
233
+ "loss": 0.4431,
234
+ "step": 310
235
+ },
236
+ {
237
+ "epoch": 0.5895900506678949,
238
+ "grad_norm": 1.223702182223073,
239
+ "learning_rate": 1.943705887011874e-06,
240
+ "loss": 0.4554,
241
+ "step": 320
242
+ },
243
+ {
244
+ "epoch": 0.6080147397512667,
245
+ "grad_norm": 1.3622081330675895,
246
+ "learning_rate": 1.936385690179698e-06,
247
+ "loss": 0.4411,
248
+ "step": 330
249
+ },
250
+ {
251
+ "epoch": 0.6264394288346384,
252
+ "grad_norm": 1.519239936734184,
253
+ "learning_rate": 1.928633726619129e-06,
254
+ "loss": 0.4272,
255
+ "step": 340
256
+ },
257
+ {
258
+ "epoch": 0.6448641179180101,
259
+ "grad_norm": 1.567409748872381,
260
+ "learning_rate": 1.92045357075466e-06,
261
+ "loss": 0.4171,
262
+ "step": 350
263
+ },
264
+ {
265
+ "epoch": 0.6632888070013818,
266
+ "grad_norm": 1.4683737283437361,
267
+ "learning_rate": 1.9118489944499284e-06,
268
+ "loss": 0.4282,
269
+ "step": 360
270
+ },
271
+ {
272
+ "epoch": 0.6817134960847536,
273
+ "grad_norm": 1.3169044491538355,
274
+ "learning_rate": 1.9028239652685126e-06,
275
+ "loss": 0.4202,
276
+ "step": 370
277
+ },
278
+ {
279
+ "epoch": 0.7001381851681253,
280
+ "grad_norm": 1.4350164646007288,
281
+ "learning_rate": 1.893382644644493e-06,
282
+ "loss": 0.4159,
283
+ "step": 380
284
+ },
285
+ {
286
+ "epoch": 0.718562874251497,
287
+ "grad_norm": 1.5280431086824893,
288
+ "learning_rate": 1.8835293859636177e-06,
289
+ "loss": 0.3997,
290
+ "step": 390
291
+ },
292
+ {
293
+ "epoch": 0.7369875633348687,
294
+ "grad_norm": 1.6589930030103581,
295
+ "learning_rate": 1.8732687325559566e-06,
296
+ "loss": 0.4079,
297
+ "step": 400
298
+ },
299
+ {
300
+ "epoch": 0.7369875633348687,
301
+ "eval_loss": 0.3841191828250885,
302
+ "eval_runtime": 422.8229,
303
+ "eval_samples_per_second": 18.251,
304
+ "eval_steps_per_second": 2.282,
305
+ "step": 400
306
+ },
307
+ {
308
+ "epoch": 0.7554122524182404,
309
+ "grad_norm": 1.6654473824863523,
310
+ "learning_rate": 1.8626054156009805e-06,
311
+ "loss": 0.404,
312
+ "step": 410
313
+ },
314
+ {
315
+ "epoch": 0.7738369415016122,
316
+ "grad_norm": 2.3979929087591527,
317
+ "learning_rate": 1.8515443519460136e-06,
318
+ "loss": 0.4003,
319
+ "step": 420
320
+ },
321
+ {
322
+ "epoch": 0.7922616305849839,
323
+ "grad_norm": 1.3611905196878527,
324
+ "learning_rate": 1.8400906418390804e-06,
325
+ "loss": 0.3799,
326
+ "step": 430
327
+ },
328
+ {
329
+ "epoch": 0.8106863196683556,
330
+ "grad_norm": 1.4783642041172977,
331
+ "learning_rate": 1.8282495665771862e-06,
332
+ "loss": 0.387,
333
+ "step": 440
334
+ },
335
+ {
336
+ "epoch": 0.8291110087517273,
337
+ "grad_norm": 1.3634768413482825,
338
+ "learning_rate": 1.8160265860711132e-06,
339
+ "loss": 0.3824,
340
+ "step": 450
341
+ },
342
+ {
343
+ "epoch": 0.8475356978350991,
344
+ "grad_norm": 1.6646421151787196,
345
+ "learning_rate": 1.803427336327861e-06,
346
+ "loss": 0.3875,
347
+ "step": 460
348
+ },
349
+ {
350
+ "epoch": 0.8659603869184708,
351
+ "grad_norm": 1.5372444888922498,
352
+ "learning_rate": 1.7904576268518884e-06,
353
+ "loss": 0.3906,
354
+ "step": 470
355
+ },
356
+ {
357
+ "epoch": 0.8843850760018425,
358
+ "grad_norm": 1.6088504472102723,
359
+ "learning_rate": 1.7771234379663542e-06,
360
+ "loss": 0.3712,
361
+ "step": 480
362
+ },
363
+ {
364
+ "epoch": 0.9028097650852142,
365
+ "grad_norm": 1.764162466911846,
366
+ "learning_rate": 1.763430918055595e-06,
367
+ "loss": 0.3606,
368
+ "step": 490
369
+ },
370
+ {
371
+ "epoch": 0.921234454168586,
372
+ "grad_norm": 1.71583814988475,
373
+ "learning_rate": 1.7493863807301113e-06,
374
+ "loss": 0.3596,
375
+ "step": 500
376
+ },
377
+ {
378
+ "epoch": 0.9396591432519577,
379
+ "grad_norm": 1.5688409159154797,
380
+ "learning_rate": 1.7349963019153635e-06,
381
+ "loss": 0.3574,
382
+ "step": 510
383
+ },
384
+ {
385
+ "epoch": 0.9580838323353293,
386
+ "grad_norm": 2.332890317163757,
387
+ "learning_rate": 1.7202673168657315e-06,
388
+ "loss": 0.3568,
389
+ "step": 520
390
+ },
391
+ {
392
+ "epoch": 0.976508521418701,
393
+ "grad_norm": 1.6627193812593573,
394
+ "learning_rate": 1.7052062171050005e-06,
395
+ "loss": 0.3806,
396
+ "step": 530
397
+ },
398
+ {
399
+ "epoch": 0.9949332105020727,
400
+ "grad_norm": 1.677719889885817,
401
+ "learning_rate": 1.6898199472947968e-06,
402
+ "loss": 0.3497,
403
+ "step": 540
404
+ },
405
+ {
406
+ "epoch": 1.0133578995854444,
407
+ "grad_norm": 1.9190946425252744,
408
+ "learning_rate": 1.6741156020324084e-06,
409
+ "loss": 0.3313,
410
+ "step": 550
411
+ },
412
+ {
413
+ "epoch": 1.0317825886688161,
414
+ "grad_norm": 1.7445670792303516,
415
+ "learning_rate": 1.6581004225794712e-06,
416
+ "loss": 0.3074,
417
+ "step": 560
418
+ },
419
+ {
420
+ "epoch": 1.050207277752188,
421
+ "grad_norm": 1.5838800006945206,
422
+ "learning_rate": 1.6417817935230316e-06,
423
+ "loss": 0.3225,
424
+ "step": 570
425
+ },
426
+ {
427
+ "epoch": 1.0686319668355597,
428
+ "grad_norm": 1.6389397510853179,
429
+ "learning_rate": 1.625167239370515e-06,
430
+ "loss": 0.3096,
431
+ "step": 580
432
+ },
433
+ {
434
+ "epoch": 1.0870566559189314,
435
+ "grad_norm": 1.7537419891849084,
436
+ "learning_rate": 1.6082644210801843e-06,
437
+ "loss": 0.3111,
438
+ "step": 590
439
+ },
440
+ {
441
+ "epoch": 1.105481345002303,
442
+ "grad_norm": 1.6608064872566262,
443
+ "learning_rate": 1.5910811325286765e-06,
444
+ "loss": 0.3062,
445
+ "step": 600
446
+ },
447
+ {
448
+ "epoch": 1.105481345002303,
449
+ "eval_loss": 0.31510841846466064,
450
+ "eval_runtime": 413.4567,
451
+ "eval_samples_per_second": 18.665,
452
+ "eval_steps_per_second": 2.334,
453
+ "step": 600
454
+ },
455
+ {
456
+ "epoch": 1.1239060340856748,
457
+ "grad_norm": 1.66378688831552,
458
+ "learning_rate": 1.573625296917252e-06,
459
+ "loss": 0.2989,
460
+ "step": 610
461
+ },
462
+ {
463
+ "epoch": 1.1423307231690465,
464
+ "grad_norm": 1.83094448775576,
465
+ "learning_rate": 1.5559049631184134e-06,
466
+ "loss": 0.2999,
467
+ "step": 620
468
+ },
469
+ {
470
+ "epoch": 1.1607554122524182,
471
+ "grad_norm": 1.6466825765437474,
472
+ "learning_rate": 1.5379283019645757e-06,
473
+ "loss": 0.2921,
474
+ "step": 630
475
+ },
476
+ {
477
+ "epoch": 1.1791801013357899,
478
+ "grad_norm": 1.9136384536335407,
479
+ "learning_rate": 1.5197036024805016e-06,
480
+ "loss": 0.2899,
481
+ "step": 640
482
+ },
483
+ {
484
+ "epoch": 1.1976047904191618,
485
+ "grad_norm": 1.903672049718007,
486
+ "learning_rate": 1.5012392680612408e-06,
487
+ "loss": 0.2949,
488
+ "step": 650
489
+ },
490
+ {
491
+ "epoch": 1.2160294795025335,
492
+ "grad_norm": 1.9119223368600173,
493
+ "learning_rate": 1.4825438125973263e-06,
494
+ "loss": 0.2882,
495
+ "step": 660
496
+ },
497
+ {
498
+ "epoch": 1.2344541685859052,
499
+ "grad_norm": 2.1449683362841587,
500
+ "learning_rate": 1.4636258565490303e-06,
501
+ "loss": 0.3033,
502
+ "step": 670
503
+ },
504
+ {
505
+ "epoch": 1.2528788576692769,
506
+ "grad_norm": 1.7831247878191976,
507
+ "learning_rate": 1.4444941229714758e-06,
508
+ "loss": 0.281,
509
+ "step": 680
510
+ },
511
+ {
512
+ "epoch": 1.2713035467526486,
513
+ "grad_norm": 1.95273338911629,
514
+ "learning_rate": 1.4251574334924393e-06,
515
+ "loss": 0.2816,
516
+ "step": 690
517
+ },
518
+ {
519
+ "epoch": 1.2897282358360203,
520
+ "grad_norm": 2.4288369199858857,
521
+ "learning_rate": 1.4056247042447093e-06,
522
+ "loss": 0.2805,
523
+ "step": 700
524
+ },
525
+ {
526
+ "epoch": 1.308152924919392,
527
+ "grad_norm": 1.89388118234306,
528
+ "learning_rate": 1.3859049417548618e-06,
529
+ "loss": 0.2735,
530
+ "step": 710
531
+ },
532
+ {
533
+ "epoch": 1.3265776140027636,
534
+ "grad_norm": 1.7524435130158207,
535
+ "learning_rate": 1.3660072387903579e-06,
536
+ "loss": 0.2768,
537
+ "step": 720
538
+ },
539
+ {
540
+ "epoch": 1.3450023030861353,
541
+ "grad_norm": 1.6591458938910955,
542
+ "learning_rate": 1.345940770166876e-06,
543
+ "loss": 0.2711,
544
+ "step": 730
545
+ },
546
+ {
547
+ "epoch": 1.363426992169507,
548
+ "grad_norm": 1.8640912895563528,
549
+ "learning_rate": 1.3257147885178123e-06,
550
+ "loss": 0.2748,
551
+ "step": 740
552
+ },
553
+ {
554
+ "epoch": 1.381851681252879,
555
+ "grad_norm": 1.8871016983908344,
556
+ "learning_rate": 1.3053386200278961e-06,
557
+ "loss": 0.2562,
558
+ "step": 750
559
+ },
560
+ {
561
+ "epoch": 1.4002763703362506,
562
+ "grad_norm": 1.8787377767418174,
563
+ "learning_rate": 1.2848216601328956e-06,
564
+ "loss": 0.2673,
565
+ "step": 760
566
+ },
567
+ {
568
+ "epoch": 1.4187010594196223,
569
+ "grad_norm": 1.808234271387127,
570
+ "learning_rate": 1.2641733691873882e-06,
571
+ "loss": 0.277,
572
+ "step": 770
573
+ },
574
+ {
575
+ "epoch": 1.437125748502994,
576
+ "grad_norm": 1.8743491223069482,
577
+ "learning_rate": 1.2434032681025986e-06,
578
+ "loss": 0.2578,
579
+ "step": 780
580
+ },
581
+ {
582
+ "epoch": 1.4555504375863657,
583
+ "grad_norm": 1.5665178489233025,
584
+ "learning_rate": 1.2225209339563143e-06,
585
+ "loss": 0.2583,
586
+ "step": 790
587
+ },
588
+ {
589
+ "epoch": 1.4739751266697374,
590
+ "grad_norm": 2.1334656844759325,
591
+ "learning_rate": 1.201535995576902e-06,
592
+ "loss": 0.2699,
593
+ "step": 800
594
+ },
595
+ {
596
+ "epoch": 1.4739751266697374,
597
+ "eval_loss": 0.2707812488079071,
598
+ "eval_runtime": 410.1156,
599
+ "eval_samples_per_second": 18.817,
600
+ "eval_steps_per_second": 2.353,
601
+ "step": 800
602
+ },
603
+ {
604
+ "epoch": 1.492399815753109,
605
+ "grad_norm": 1.9887460771971892,
606
+ "learning_rate": 1.1804581291034615e-06,
607
+ "loss": 0.2888,
608
+ "step": 810
609
+ },
610
+ {
611
+ "epoch": 1.510824504836481,
612
+ "grad_norm": 1.6772712284536069,
613
+ "learning_rate": 1.1592970535241667e-06,
614
+ "loss": 0.2612,
615
+ "step": 820
616
+ },
617
+ {
618
+ "epoch": 1.5292491939198527,
619
+ "grad_norm": 1.8994787118106975,
620
+ "learning_rate": 1.1380625261948457e-06,
621
+ "loss": 0.2693,
622
+ "step": 830
623
+ },
624
+ {
625
+ "epoch": 1.5476738830032244,
626
+ "grad_norm": 1.6305539275745158,
627
+ "learning_rate": 1.1167643383398744e-06,
628
+ "loss": 0.261,
629
+ "step": 840
630
+ },
631
+ {
632
+ "epoch": 1.566098572086596,
633
+ "grad_norm": 1.684538304608348,
634
+ "learning_rate": 1.0954123105374467e-06,
635
+ "loss": 0.2553,
636
+ "step": 850
637
+ },
638
+ {
639
+ "epoch": 1.5845232611699678,
640
+ "grad_norm": 1.853188540021309,
641
+ "learning_rate": 1.0740162881913164e-06,
642
+ "loss": 0.2597,
643
+ "step": 860
644
+ },
645
+ {
646
+ "epoch": 1.6029479502533395,
647
+ "grad_norm": 1.6601117418801663,
648
+ "learning_rate": 1.0525861369910876e-06,
649
+ "loss": 0.249,
650
+ "step": 870
651
+ },
652
+ {
653
+ "epoch": 1.6213726393367112,
654
+ "grad_norm": 1.7563715504893396,
655
+ "learning_rate": 1.031131738363153e-06,
656
+ "loss": 0.25,
657
+ "step": 880
658
+ },
659
+ {
660
+ "epoch": 1.6397973284200829,
661
+ "grad_norm": 2.082818010442183,
662
+ "learning_rate": 1.0096629849143757e-06,
663
+ "loss": 0.251,
664
+ "step": 890
665
+ },
666
+ {
667
+ "epoch": 1.6582220175034545,
668
+ "grad_norm": 1.6295520934299772,
669
+ "learning_rate": 9.881897758706154e-07,
670
+ "loss": 0.2363,
671
+ "step": 900
672
+ },
673
+ {
674
+ "epoch": 1.6766467065868262,
675
+ "grad_norm": 1.881860506349835,
676
+ "learning_rate": 9.667220125122042e-07,
677
+ "loss": 0.2322,
678
+ "step": 910
679
+ },
680
+ {
681
+ "epoch": 1.695071395670198,
682
+ "grad_norm": 1.710287346085137,
683
+ "learning_rate": 9.452695936084726e-07,
684
+ "loss": 0.2478,
685
+ "step": 920
686
+ },
687
+ {
688
+ "epoch": 1.7134960847535696,
689
+ "grad_norm": 1.9656481185455599,
690
+ "learning_rate": 9.238424108534332e-07,
691
+ "loss": 0.2478,
692
+ "step": 930
693
+ },
694
+ {
695
+ "epoch": 1.7319207738369415,
696
+ "grad_norm": 1.7567627102945302,
697
+ "learning_rate": 9.024503443047318e-07,
698
+ "loss": 0.2525,
699
+ "step": 940
700
+ },
701
+ {
702
+ "epoch": 1.7503454629203132,
703
+ "grad_norm": 2.1264042760141497,
704
+ "learning_rate": 8.811032578279568e-07,
705
+ "loss": 0.2377,
706
+ "step": 950
707
+ },
708
+ {
709
+ "epoch": 1.768770152003685,
710
+ "grad_norm": 1.8132528012747537,
711
+ "learning_rate": 8.598109945484207e-07,
712
+ "loss": 0.2343,
713
+ "step": 960
714
+ },
715
+ {
716
+ "epoch": 1.7871948410870566,
717
+ "grad_norm": 2.1145702458263735,
718
+ "learning_rate": 8.385833723125006e-07,
719
+ "loss": 0.2365,
720
+ "step": 970
721
+ },
722
+ {
723
+ "epoch": 1.8056195301704285,
724
+ "grad_norm": 1.8190541312769752,
725
+ "learning_rate": 8.174301791606384e-07,
726
+ "loss": 0.2582,
727
+ "step": 980
728
+ },
729
+ {
730
+ "epoch": 1.8240442192538002,
731
+ "grad_norm": 1.928487419366485,
732
+ "learning_rate": 7.963611688140813e-07,
733
+ "loss": 0.2394,
734
+ "step": 990
735
+ },
736
+ {
737
+ "epoch": 1.842468908337172,
738
+ "grad_norm": 1.7487120218475518,
739
+ "learning_rate": 7.753860561774494e-07,
740
+ "loss": 0.2312,
741
+ "step": 1000
742
+ },
743
+ {
744
+ "epoch": 1.842468908337172,
745
+ "eval_loss": 0.2388714849948883,
746
+ "eval_runtime": 408.9275,
747
+ "eval_samples_per_second": 18.871,
748
+ "eval_steps_per_second": 2.36,
749
+ "step": 1000
750
+ }
751
+ ],
752
+ "logging_steps": 10,
753
+ "max_steps": 1626,
754
+ "num_input_tokens_seen": 0,
755
+ "num_train_epochs": 3,
756
+ "save_steps": 500,
757
+ "stateful_callbacks": {
758
+ "TrainerControl": {
759
+ "args": {
760
+ "should_epoch_stop": false,
761
+ "should_evaluate": false,
762
+ "should_log": false,
763
+ "should_save": true,
764
+ "should_training_stop": false
765
+ },
766
+ "attributes": {}
767
+ }
768
+ },
769
+ "total_flos": 837309243064320.0,
770
+ "train_batch_size": 4,
771
+ "trial_name": null,
772
+ "trial_params": null
773
+ }