skrd3 commited on
Commit
51d0f78
·
verified ·
1 Parent(s): 276141d

Upload best model checkpoint (step 93, loss 0.9836)

Browse files
Files changed (3) hide show
  1. generation_config.json +8 -0
  2. loss.txt +1 -0
  3. trainer_state.json +701 -0
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 1,
3
+ "do_sample": true,
4
+ "eos_token_id": 2,
5
+ "max_length": 2048,
6
+ "pad_token_id": 0,
7
+ "transformers_version": "4.51.3"
8
+ }
loss.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ 93,0.9836028218269348
trainer_state.json ADDED
@@ -0,0 +1,701 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.0536254865215511,
6
+ "eval_steps": 1734,
7
+ "global_step": 93,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0005766181346403345,
14
+ "grad_norm": 2.40625,
15
+ "learning_rate": 0.0,
16
+ "loss": 1.2938,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.0005766181346403345,
21
+ "eval_loss": 1.3029676675796509,
22
+ "eval_runtime": 1.482,
23
+ "eval_samples_per_second": 125.506,
24
+ "eval_steps_per_second": 2.699,
25
+ "step": 1
26
+ },
27
+ {
28
+ "epoch": 0.001153236269280669,
29
+ "grad_norm": 2.59375,
30
+ "learning_rate": 1e-05,
31
+ "loss": 1.2719,
32
+ "step": 2
33
+ },
34
+ {
35
+ "epoch": 0.0017298544039210033,
36
+ "grad_norm": 1.7265625,
37
+ "learning_rate": 2e-05,
38
+ "loss": 1.2501,
39
+ "step": 3
40
+ },
41
+ {
42
+ "epoch": 0.002306472538561338,
43
+ "grad_norm": 1.3359375,
44
+ "learning_rate": 3e-05,
45
+ "loss": 1.1749,
46
+ "step": 4
47
+ },
48
+ {
49
+ "epoch": 0.0028830906732016724,
50
+ "grad_norm": 1.203125,
51
+ "learning_rate": 4e-05,
52
+ "loss": 1.2028,
53
+ "step": 5
54
+ },
55
+ {
56
+ "epoch": 0.0034597088078420065,
57
+ "grad_norm": 1.1953125,
58
+ "learning_rate": 5e-05,
59
+ "loss": 1.1511,
60
+ "step": 6
61
+ },
62
+ {
63
+ "epoch": 0.004036326942482341,
64
+ "grad_norm": 1.1875,
65
+ "learning_rate": 6e-05,
66
+ "loss": 1.1037,
67
+ "step": 7
68
+ },
69
+ {
70
+ "epoch": 0.004612945077122676,
71
+ "grad_norm": 1.0390625,
72
+ "learning_rate": 7e-05,
73
+ "loss": 1.0619,
74
+ "step": 8
75
+ },
76
+ {
77
+ "epoch": 0.00518956321176301,
78
+ "grad_norm": 0.86328125,
79
+ "learning_rate": 8e-05,
80
+ "loss": 1.1328,
81
+ "step": 9
82
+ },
83
+ {
84
+ "epoch": 0.005766181346403345,
85
+ "grad_norm": 0.90625,
86
+ "learning_rate": 9e-05,
87
+ "loss": 1.0916,
88
+ "step": 10
89
+ },
90
+ {
91
+ "epoch": 0.0063427994810436785,
92
+ "grad_norm": 0.83984375,
93
+ "learning_rate": 0.0001,
94
+ "loss": 1.0373,
95
+ "step": 11
96
+ },
97
+ {
98
+ "epoch": 0.006919417615684013,
99
+ "grad_norm": 0.9140625,
100
+ "learning_rate": 9.999999359279803e-05,
101
+ "loss": 1.0594,
102
+ "step": 12
103
+ },
104
+ {
105
+ "epoch": 0.007496035750324348,
106
+ "grad_norm": 0.7734375,
107
+ "learning_rate": 9.999997437119453e-05,
108
+ "loss": 1.0309,
109
+ "step": 13
110
+ },
111
+ {
112
+ "epoch": 0.008072653884964682,
113
+ "grad_norm": 0.8046875,
114
+ "learning_rate": 9.999994233519648e-05,
115
+ "loss": 1.0731,
116
+ "step": 14
117
+ },
118
+ {
119
+ "epoch": 0.008649272019605017,
120
+ "grad_norm": 0.79296875,
121
+ "learning_rate": 9.999989748481562e-05,
122
+ "loss": 1.0455,
123
+ "step": 15
124
+ },
125
+ {
126
+ "epoch": 0.009225890154245351,
127
+ "grad_norm": 0.7421875,
128
+ "learning_rate": 9.99998398200684e-05,
129
+ "loss": 1.0654,
130
+ "step": 16
131
+ },
132
+ {
133
+ "epoch": 0.009802508288885686,
134
+ "grad_norm": 0.7109375,
135
+ "learning_rate": 9.99997693409759e-05,
136
+ "loss": 1.0141,
137
+ "step": 17
138
+ },
139
+ {
140
+ "epoch": 0.01037912642352602,
141
+ "grad_norm": 0.71875,
142
+ "learning_rate": 9.999968604756393e-05,
143
+ "loss": 1.0353,
144
+ "step": 18
145
+ },
146
+ {
147
+ "epoch": 0.010955744558166355,
148
+ "grad_norm": 0.7109375,
149
+ "learning_rate": 9.999958993986302e-05,
150
+ "loss": 1.0461,
151
+ "step": 19
152
+ },
153
+ {
154
+ "epoch": 0.01153236269280669,
155
+ "grad_norm": 0.69140625,
156
+ "learning_rate": 9.999948101790833e-05,
157
+ "loss": 1.0137,
158
+ "step": 20
159
+ },
160
+ {
161
+ "epoch": 0.012108980827447022,
162
+ "grad_norm": 0.71484375,
163
+ "learning_rate": 9.999935928173971e-05,
164
+ "loss": 1.0216,
165
+ "step": 21
166
+ },
167
+ {
168
+ "epoch": 0.012685598962087357,
169
+ "grad_norm": 0.6953125,
170
+ "learning_rate": 9.999922473140177e-05,
171
+ "loss": 1.0957,
172
+ "step": 22
173
+ },
174
+ {
175
+ "epoch": 0.013262217096727692,
176
+ "grad_norm": 0.73046875,
177
+ "learning_rate": 9.999907736694379e-05,
178
+ "loss": 1.0038,
179
+ "step": 23
180
+ },
181
+ {
182
+ "epoch": 0.013838835231368026,
183
+ "grad_norm": 0.67578125,
184
+ "learning_rate": 9.99989171884197e-05,
185
+ "loss": 1.0123,
186
+ "step": 24
187
+ },
188
+ {
189
+ "epoch": 0.01441545336600836,
190
+ "grad_norm": 0.68359375,
191
+ "learning_rate": 9.99987441958881e-05,
192
+ "loss": 0.9884,
193
+ "step": 25
194
+ },
195
+ {
196
+ "epoch": 0.014992071500648695,
197
+ "grad_norm": 0.6953125,
198
+ "learning_rate": 9.999855838941241e-05,
199
+ "loss": 0.9824,
200
+ "step": 26
201
+ },
202
+ {
203
+ "epoch": 0.01556868963528903,
204
+ "grad_norm": 0.6875,
205
+ "learning_rate": 9.99983597690606e-05,
206
+ "loss": 1.0282,
207
+ "step": 27
208
+ },
209
+ {
210
+ "epoch": 0.016145307769929364,
211
+ "grad_norm": 0.6875,
212
+ "learning_rate": 9.999814833490542e-05,
213
+ "loss": 1.0501,
214
+ "step": 28
215
+ },
216
+ {
217
+ "epoch": 0.0167219259045697,
218
+ "grad_norm": 0.671875,
219
+ "learning_rate": 9.999792408702424e-05,
220
+ "loss": 1.0105,
221
+ "step": 29
222
+ },
223
+ {
224
+ "epoch": 0.017298544039210034,
225
+ "grad_norm": 0.74609375,
226
+ "learning_rate": 9.999768702549923e-05,
227
+ "loss": 1.0676,
228
+ "step": 30
229
+ },
230
+ {
231
+ "epoch": 0.017875162173850368,
232
+ "grad_norm": 0.6875,
233
+ "learning_rate": 9.999743715041714e-05,
234
+ "loss": 1.021,
235
+ "step": 31
236
+ },
237
+ {
238
+ "epoch": 0.018451780308490703,
239
+ "grad_norm": 0.67578125,
240
+ "learning_rate": 9.999717446186944e-05,
241
+ "loss": 0.9625,
242
+ "step": 32
243
+ },
244
+ {
245
+ "epoch": 0.019028398443131037,
246
+ "grad_norm": 0.71484375,
247
+ "learning_rate": 9.999689895995237e-05,
248
+ "loss": 1.0373,
249
+ "step": 33
250
+ },
251
+ {
252
+ "epoch": 0.019605016577771372,
253
+ "grad_norm": 0.68359375,
254
+ "learning_rate": 9.999661064476674e-05,
255
+ "loss": 1.0124,
256
+ "step": 34
257
+ },
258
+ {
259
+ "epoch": 0.020181634712411706,
260
+ "grad_norm": 0.66796875,
261
+ "learning_rate": 9.999630951641812e-05,
262
+ "loss": 0.9176,
263
+ "step": 35
264
+ },
265
+ {
266
+ "epoch": 0.02075825284705204,
267
+ "grad_norm": 0.68359375,
268
+ "learning_rate": 9.999599557501678e-05,
269
+ "loss": 1.0163,
270
+ "step": 36
271
+ },
272
+ {
273
+ "epoch": 0.021334870981692376,
274
+ "grad_norm": 0.68359375,
275
+ "learning_rate": 9.999566882067764e-05,
276
+ "loss": 1.0296,
277
+ "step": 37
278
+ },
279
+ {
280
+ "epoch": 0.02191148911633271,
281
+ "grad_norm": 0.671875,
282
+ "learning_rate": 9.999532925352036e-05,
283
+ "loss": 1.0034,
284
+ "step": 38
285
+ },
286
+ {
287
+ "epoch": 0.022488107250973045,
288
+ "grad_norm": 0.640625,
289
+ "learning_rate": 9.999497687366923e-05,
290
+ "loss": 0.9524,
291
+ "step": 39
292
+ },
293
+ {
294
+ "epoch": 0.02306472538561338,
295
+ "grad_norm": 0.6875,
296
+ "learning_rate": 9.999461168125327e-05,
297
+ "loss": 0.9762,
298
+ "step": 40
299
+ },
300
+ {
301
+ "epoch": 0.02364134352025371,
302
+ "grad_norm": 0.66015625,
303
+ "learning_rate": 9.999423367640626e-05,
304
+ "loss": 0.9876,
305
+ "step": 41
306
+ },
307
+ {
308
+ "epoch": 0.024217961654894045,
309
+ "grad_norm": 0.66796875,
310
+ "learning_rate": 9.99938428592665e-05,
311
+ "loss": 1.0121,
312
+ "step": 42
313
+ },
314
+ {
315
+ "epoch": 0.02479457978953438,
316
+ "grad_norm": 0.64453125,
317
+ "learning_rate": 9.99934392299771e-05,
318
+ "loss": 0.9703,
319
+ "step": 43
320
+ },
321
+ {
322
+ "epoch": 0.025371197924174714,
323
+ "grad_norm": 0.64453125,
324
+ "learning_rate": 9.999302278868587e-05,
325
+ "loss": 0.9806,
326
+ "step": 44
327
+ },
328
+ {
329
+ "epoch": 0.02594781605881505,
330
+ "grad_norm": 0.65625,
331
+ "learning_rate": 9.999259353554528e-05,
332
+ "loss": 1.0016,
333
+ "step": 45
334
+ },
335
+ {
336
+ "epoch": 0.026524434193455383,
337
+ "grad_norm": 0.6796875,
338
+ "learning_rate": 9.999215147071245e-05,
339
+ "loss": 1.0187,
340
+ "step": 46
341
+ },
342
+ {
343
+ "epoch": 0.027101052328095718,
344
+ "grad_norm": 0.6484375,
345
+ "learning_rate": 9.999169659434929e-05,
346
+ "loss": 0.9999,
347
+ "step": 47
348
+ },
349
+ {
350
+ "epoch": 0.027677670462736052,
351
+ "grad_norm": 0.6484375,
352
+ "learning_rate": 9.99912289066223e-05,
353
+ "loss": 0.994,
354
+ "step": 48
355
+ },
356
+ {
357
+ "epoch": 0.028254288597376387,
358
+ "grad_norm": 0.68359375,
359
+ "learning_rate": 9.999074840770268e-05,
360
+ "loss": 0.9996,
361
+ "step": 49
362
+ },
363
+ {
364
+ "epoch": 0.02883090673201672,
365
+ "grad_norm": 0.640625,
366
+ "learning_rate": 9.999025509776645e-05,
367
+ "loss": 0.9744,
368
+ "step": 50
369
+ },
370
+ {
371
+ "epoch": 0.029407524866657056,
372
+ "grad_norm": 0.6328125,
373
+ "learning_rate": 9.998974897699413e-05,
374
+ "loss": 0.9966,
375
+ "step": 51
376
+ },
377
+ {
378
+ "epoch": 0.02998414300129739,
379
+ "grad_norm": 0.6484375,
380
+ "learning_rate": 9.99892300455711e-05,
381
+ "loss": 0.9773,
382
+ "step": 52
383
+ },
384
+ {
385
+ "epoch": 0.030560761135937725,
386
+ "grad_norm": 0.6640625,
387
+ "learning_rate": 9.998869830368729e-05,
388
+ "loss": 0.993,
389
+ "step": 53
390
+ },
391
+ {
392
+ "epoch": 0.03113737927057806,
393
+ "grad_norm": 0.62890625,
394
+ "learning_rate": 9.99881537515374e-05,
395
+ "loss": 0.9538,
396
+ "step": 54
397
+ },
398
+ {
399
+ "epoch": 0.031713997405218394,
400
+ "grad_norm": 0.671875,
401
+ "learning_rate": 9.998759638932085e-05,
402
+ "loss": 0.9988,
403
+ "step": 55
404
+ },
405
+ {
406
+ "epoch": 0.03229061553985873,
407
+ "grad_norm": 0.65234375,
408
+ "learning_rate": 9.998702621724162e-05,
409
+ "loss": 1.0345,
410
+ "step": 56
411
+ },
412
+ {
413
+ "epoch": 0.03286723367449906,
414
+ "grad_norm": 0.65234375,
415
+ "learning_rate": 9.998644323550856e-05,
416
+ "loss": 1.0087,
417
+ "step": 57
418
+ },
419
+ {
420
+ "epoch": 0.0334438518091394,
421
+ "grad_norm": 0.65234375,
422
+ "learning_rate": 9.998584744433506e-05,
423
+ "loss": 0.9365,
424
+ "step": 58
425
+ },
426
+ {
427
+ "epoch": 0.03402046994377973,
428
+ "grad_norm": 0.63671875,
429
+ "learning_rate": 9.998523884393927e-05,
430
+ "loss": 0.9741,
431
+ "step": 59
432
+ },
433
+ {
434
+ "epoch": 0.03459708807842007,
435
+ "grad_norm": 0.6171875,
436
+ "learning_rate": 9.998461743454399e-05,
437
+ "loss": 0.9498,
438
+ "step": 60
439
+ },
440
+ {
441
+ "epoch": 0.0351737062130604,
442
+ "grad_norm": 0.703125,
443
+ "learning_rate": 9.998398321637676e-05,
444
+ "loss": 1.0063,
445
+ "step": 61
446
+ },
447
+ {
448
+ "epoch": 0.035750324347700736,
449
+ "grad_norm": 0.66015625,
450
+ "learning_rate": 9.998333618966978e-05,
451
+ "loss": 0.9967,
452
+ "step": 62
453
+ },
454
+ {
455
+ "epoch": 0.03632694248234107,
456
+ "grad_norm": 0.62890625,
457
+ "learning_rate": 9.998267635465992e-05,
458
+ "loss": 0.9099,
459
+ "step": 63
460
+ },
461
+ {
462
+ "epoch": 0.036903560616981405,
463
+ "grad_norm": 0.6640625,
464
+ "learning_rate": 9.998200371158881e-05,
465
+ "loss": 0.9456,
466
+ "step": 64
467
+ },
468
+ {
469
+ "epoch": 0.03748017875162174,
470
+ "grad_norm": 0.64453125,
471
+ "learning_rate": 9.998131826070267e-05,
472
+ "loss": 1.0032,
473
+ "step": 65
474
+ },
475
+ {
476
+ "epoch": 0.038056796886262075,
477
+ "grad_norm": 0.6484375,
478
+ "learning_rate": 9.99806200022525e-05,
479
+ "loss": 0.9238,
480
+ "step": 66
481
+ },
482
+ {
483
+ "epoch": 0.03863341502090241,
484
+ "grad_norm": 0.69921875,
485
+ "learning_rate": 9.997990893649394e-05,
486
+ "loss": 1.0059,
487
+ "step": 67
488
+ },
489
+ {
490
+ "epoch": 0.039210033155542744,
491
+ "grad_norm": 0.625,
492
+ "learning_rate": 9.997918506368732e-05,
493
+ "loss": 0.9957,
494
+ "step": 68
495
+ },
496
+ {
497
+ "epoch": 0.03978665129018308,
498
+ "grad_norm": 0.64453125,
499
+ "learning_rate": 9.997844838409767e-05,
500
+ "loss": 0.9871,
501
+ "step": 69
502
+ },
503
+ {
504
+ "epoch": 0.04036326942482341,
505
+ "grad_norm": 0.67578125,
506
+ "learning_rate": 9.99776988979947e-05,
507
+ "loss": 0.9906,
508
+ "step": 70
509
+ },
510
+ {
511
+ "epoch": 0.04093988755946375,
512
+ "grad_norm": 0.61328125,
513
+ "learning_rate": 9.997693660565284e-05,
514
+ "loss": 0.9593,
515
+ "step": 71
516
+ },
517
+ {
518
+ "epoch": 0.04151650569410408,
519
+ "grad_norm": 0.6640625,
520
+ "learning_rate": 9.997616150735119e-05,
521
+ "loss": 0.943,
522
+ "step": 72
523
+ },
524
+ {
525
+ "epoch": 0.042093123828744416,
526
+ "grad_norm": 0.64453125,
527
+ "learning_rate": 9.997537360337348e-05,
528
+ "loss": 0.988,
529
+ "step": 73
530
+ },
531
+ {
532
+ "epoch": 0.04266974196338475,
533
+ "grad_norm": 0.64453125,
534
+ "learning_rate": 9.997457289400825e-05,
535
+ "loss": 0.9691,
536
+ "step": 74
537
+ },
538
+ {
539
+ "epoch": 0.043246360098025086,
540
+ "grad_norm": 0.6640625,
541
+ "learning_rate": 9.997375937954861e-05,
542
+ "loss": 0.9682,
543
+ "step": 75
544
+ },
545
+ {
546
+ "epoch": 0.04382297823266542,
547
+ "grad_norm": 0.64453125,
548
+ "learning_rate": 9.997293306029243e-05,
549
+ "loss": 0.9736,
550
+ "step": 76
551
+ },
552
+ {
553
+ "epoch": 0.044399596367305755,
554
+ "grad_norm": 0.64453125,
555
+ "learning_rate": 9.997209393654223e-05,
556
+ "loss": 0.9804,
557
+ "step": 77
558
+ },
559
+ {
560
+ "epoch": 0.04497621450194609,
561
+ "grad_norm": 0.63671875,
562
+ "learning_rate": 9.997124200860528e-05,
563
+ "loss": 0.9651,
564
+ "step": 78
565
+ },
566
+ {
567
+ "epoch": 0.045552832636586424,
568
+ "grad_norm": 0.62109375,
569
+ "learning_rate": 9.997037727679343e-05,
570
+ "loss": 0.943,
571
+ "step": 79
572
+ },
573
+ {
574
+ "epoch": 0.04612945077122676,
575
+ "grad_norm": 0.66015625,
576
+ "learning_rate": 9.996949974142334e-05,
577
+ "loss": 0.9878,
578
+ "step": 80
579
+ },
580
+ {
581
+ "epoch": 0.046706068905867086,
582
+ "grad_norm": 0.66796875,
583
+ "learning_rate": 9.996860940281624e-05,
584
+ "loss": 0.956,
585
+ "step": 81
586
+ },
587
+ {
588
+ "epoch": 0.04728268704050742,
589
+ "grad_norm": 0.6484375,
590
+ "learning_rate": 9.996770626129817e-05,
591
+ "loss": 0.9741,
592
+ "step": 82
593
+ },
594
+ {
595
+ "epoch": 0.047859305175147755,
596
+ "grad_norm": 0.69140625,
597
+ "learning_rate": 9.996679031719973e-05,
598
+ "loss": 0.9553,
599
+ "step": 83
600
+ },
601
+ {
602
+ "epoch": 0.04843592330978809,
603
+ "grad_norm": 0.66015625,
604
+ "learning_rate": 9.996586157085632e-05,
605
+ "loss": 0.9911,
606
+ "step": 84
607
+ },
608
+ {
609
+ "epoch": 0.049012541444428424,
610
+ "grad_norm": 0.66015625,
611
+ "learning_rate": 9.996492002260794e-05,
612
+ "loss": 0.97,
613
+ "step": 85
614
+ },
615
+ {
616
+ "epoch": 0.04958915957906876,
617
+ "grad_norm": 0.66796875,
618
+ "learning_rate": 9.996396567279933e-05,
619
+ "loss": 0.9272,
620
+ "step": 86
621
+ },
622
+ {
623
+ "epoch": 0.050165777713709093,
624
+ "grad_norm": 0.64453125,
625
+ "learning_rate": 9.996299852177992e-05,
626
+ "loss": 1.0005,
627
+ "step": 87
628
+ },
629
+ {
630
+ "epoch": 0.05074239584834943,
631
+ "grad_norm": 0.6484375,
632
+ "learning_rate": 9.996201856990381e-05,
633
+ "loss": 0.9845,
634
+ "step": 88
635
+ },
636
+ {
637
+ "epoch": 0.05131901398298976,
638
+ "grad_norm": 0.6796875,
639
+ "learning_rate": 9.996102581752976e-05,
640
+ "loss": 1.0048,
641
+ "step": 89
642
+ },
643
+ {
644
+ "epoch": 0.0518956321176301,
645
+ "grad_norm": 0.61328125,
646
+ "learning_rate": 9.996002026502125e-05,
647
+ "loss": 0.9377,
648
+ "step": 90
649
+ },
650
+ {
651
+ "epoch": 0.05247225025227043,
652
+ "grad_norm": 0.640625,
653
+ "learning_rate": 9.995900191274643e-05,
654
+ "loss": 0.9595,
655
+ "step": 91
656
+ },
657
+ {
658
+ "epoch": 0.053048868386910766,
659
+ "grad_norm": 0.67578125,
660
+ "learning_rate": 9.995797076107818e-05,
661
+ "loss": 0.9824,
662
+ "step": 92
663
+ },
664
+ {
665
+ "epoch": 0.0536254865215511,
666
+ "grad_norm": 0.61328125,
667
+ "learning_rate": 9.995692681039396e-05,
668
+ "loss": 0.9639,
669
+ "step": 93
670
+ },
671
+ {
672
+ "epoch": 0.0536254865215511,
673
+ "eval_loss": 0.9836028218269348,
674
+ "eval_runtime": 1.427,
675
+ "eval_samples_per_second": 130.347,
676
+ "eval_steps_per_second": 2.803,
677
+ "step": 93
678
+ }
679
+ ],
680
+ "logging_steps": 1,
681
+ "max_steps": 5202,
682
+ "num_input_tokens_seen": 0,
683
+ "num_train_epochs": 3,
684
+ "save_steps": 1734,
685
+ "stateful_callbacks": {
686
+ "TrainerControl": {
687
+ "args": {
688
+ "should_epoch_stop": false,
689
+ "should_evaluate": false,
690
+ "should_log": false,
691
+ "should_save": true,
692
+ "should_training_stop": false
693
+ },
694
+ "attributes": {}
695
+ }
696
+ },
697
+ "total_flos": 1.4185419194027213e+17,
698
+ "train_batch_size": 60,
699
+ "trial_name": null,
700
+ "trial_params": null
701
+ }