54data commited on
Commit
9471140
1 Parent(s): 0223c81

End of training

Browse files
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "train_loss": 2.6959276412548974,
4
+ "train_runtime": 6590.4176,
5
+ "train_samples": 42367,
6
+ "train_samples_per_second": 32.143,
7
+ "train_steps_per_second": 2.009
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "train_loss": 2.6959276412548974,
4
+ "train_runtime": 6590.4176,
5
+ "train_samples": 42367,
6
+ "train_samples_per_second": 32.143,
7
+ "train_steps_per_second": 2.009
8
+ }
train_results.txt ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ epoch = 5.0
2
+ train_loss = 2.6959276412548974
3
+ train_runtime = 6590.4176
4
+ train_samples = 42367
5
+ train_samples_per_second = 32.143
6
+ train_steps_per_second = 2.009
trainer_state.json ADDED
@@ -0,0 +1,921 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 3.9967169761657715,
3
+ "best_model_checkpoint": "./models/gpt_test6/checkpoint-10000",
4
+ "epoch": 5.0,
5
+ "global_step": 13240,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.04,
12
+ "learning_rate": 4.2296072507552865e-06,
13
+ "loss": 3.8813,
14
+ "step": 100
15
+ },
16
+ {
17
+ "epoch": 0.08,
18
+ "learning_rate": 8.459214501510573e-06,
19
+ "loss": 3.4212,
20
+ "step": 200
21
+ },
22
+ {
23
+ "epoch": 0.11,
24
+ "learning_rate": 1.2688821752265861e-05,
25
+ "loss": 3.2721,
26
+ "step": 300
27
+ },
28
+ {
29
+ "epoch": 0.15,
30
+ "learning_rate": 1.6918429003021146e-05,
31
+ "loss": 3.2861,
32
+ "step": 400
33
+ },
34
+ {
35
+ "epoch": 0.19,
36
+ "learning_rate": 2.1148036253776434e-05,
37
+ "loss": 3.2347,
38
+ "step": 500
39
+ },
40
+ {
41
+ "epoch": 0.23,
42
+ "learning_rate": 2.5377643504531723e-05,
43
+ "loss": 3.1735,
44
+ "step": 600
45
+ },
46
+ {
47
+ "epoch": 0.26,
48
+ "learning_rate": 2.960725075528701e-05,
49
+ "loss": 3.2013,
50
+ "step": 700
51
+ },
52
+ {
53
+ "epoch": 0.3,
54
+ "learning_rate": 3.383685800604229e-05,
55
+ "loss": 3.1709,
56
+ "step": 800
57
+ },
58
+ {
59
+ "epoch": 0.34,
60
+ "learning_rate": 3.8066465256797584e-05,
61
+ "loss": 3.1716,
62
+ "step": 900
63
+ },
64
+ {
65
+ "epoch": 0.38,
66
+ "learning_rate": 4.229607250755287e-05,
67
+ "loss": 3.1407,
68
+ "step": 1000
69
+ },
70
+ {
71
+ "epoch": 0.38,
72
+ "eval_loss": 4.086385250091553,
73
+ "eval_runtime": 4.2648,
74
+ "eval_samples_per_second": 100.59,
75
+ "eval_steps_per_second": 6.331,
76
+ "step": 1000
77
+ },
78
+ {
79
+ "epoch": 0.42,
80
+ "learning_rate": 4.6525679758308154e-05,
81
+ "loss": 3.1629,
82
+ "step": 1100
83
+ },
84
+ {
85
+ "epoch": 0.45,
86
+ "learning_rate": 5.0755287009063445e-05,
87
+ "loss": 3.1884,
88
+ "step": 1200
89
+ },
90
+ {
91
+ "epoch": 0.49,
92
+ "learning_rate": 5.498489425981873e-05,
93
+ "loss": 3.2097,
94
+ "step": 1300
95
+ },
96
+ {
97
+ "epoch": 0.53,
98
+ "learning_rate": 5.564283316549178e-05,
99
+ "loss": 3.2055,
100
+ "step": 1400
101
+ },
102
+ {
103
+ "epoch": 0.57,
104
+ "learning_rate": 5.517287680429674e-05,
105
+ "loss": 3.1601,
106
+ "step": 1500
107
+ },
108
+ {
109
+ "epoch": 0.6,
110
+ "learning_rate": 5.4702920443101714e-05,
111
+ "loss": 3.138,
112
+ "step": 1600
113
+ },
114
+ {
115
+ "epoch": 0.64,
116
+ "learning_rate": 5.423296408190668e-05,
117
+ "loss": 3.1692,
118
+ "step": 1700
119
+ },
120
+ {
121
+ "epoch": 0.68,
122
+ "learning_rate": 5.376300772071164e-05,
123
+ "loss": 3.1219,
124
+ "step": 1800
125
+ },
126
+ {
127
+ "epoch": 0.72,
128
+ "learning_rate": 5.3293051359516615e-05,
129
+ "loss": 3.1056,
130
+ "step": 1900
131
+ },
132
+ {
133
+ "epoch": 0.76,
134
+ "learning_rate": 5.2823094998321586e-05,
135
+ "loss": 3.1639,
136
+ "step": 2000
137
+ },
138
+ {
139
+ "epoch": 0.76,
140
+ "eval_loss": 4.086682319641113,
141
+ "eval_runtime": 4.2525,
142
+ "eval_samples_per_second": 100.882,
143
+ "eval_steps_per_second": 6.349,
144
+ "step": 2000
145
+ },
146
+ {
147
+ "epoch": 0.79,
148
+ "learning_rate": 5.235313863712655e-05,
149
+ "loss": 3.1307,
150
+ "step": 2100
151
+ },
152
+ {
153
+ "epoch": 0.83,
154
+ "learning_rate": 5.188318227593152e-05,
155
+ "loss": 3.1382,
156
+ "step": 2200
157
+ },
158
+ {
159
+ "epoch": 0.87,
160
+ "learning_rate": 5.1413225914736487e-05,
161
+ "loss": 3.0843,
162
+ "step": 2300
163
+ },
164
+ {
165
+ "epoch": 0.91,
166
+ "learning_rate": 5.094326955354146e-05,
167
+ "loss": 3.094,
168
+ "step": 2400
169
+ },
170
+ {
171
+ "epoch": 0.94,
172
+ "learning_rate": 5.047331319234642e-05,
173
+ "loss": 3.0955,
174
+ "step": 2500
175
+ },
176
+ {
177
+ "epoch": 0.98,
178
+ "learning_rate": 5.000335683115139e-05,
179
+ "loss": 3.0562,
180
+ "step": 2600
181
+ },
182
+ {
183
+ "epoch": 1.02,
184
+ "learning_rate": 4.9533400469956365e-05,
185
+ "loss": 2.9527,
186
+ "step": 2700
187
+ },
188
+ {
189
+ "epoch": 1.06,
190
+ "learning_rate": 4.906344410876133e-05,
191
+ "loss": 2.8783,
192
+ "step": 2800
193
+ },
194
+ {
195
+ "epoch": 1.1,
196
+ "learning_rate": 4.8593487747566294e-05,
197
+ "loss": 2.8926,
198
+ "step": 2900
199
+ },
200
+ {
201
+ "epoch": 1.13,
202
+ "learning_rate": 4.8123531386371266e-05,
203
+ "loss": 2.87,
204
+ "step": 3000
205
+ },
206
+ {
207
+ "epoch": 1.13,
208
+ "eval_loss": 4.06110143661499,
209
+ "eval_runtime": 4.2609,
210
+ "eval_samples_per_second": 100.683,
211
+ "eval_steps_per_second": 6.337,
212
+ "step": 3000
213
+ },
214
+ {
215
+ "epoch": 1.17,
216
+ "learning_rate": 4.765357502517623e-05,
217
+ "loss": 2.8732,
218
+ "step": 3100
219
+ },
220
+ {
221
+ "epoch": 1.21,
222
+ "learning_rate": 4.71836186639812e-05,
223
+ "loss": 2.8985,
224
+ "step": 3200
225
+ },
226
+ {
227
+ "epoch": 1.25,
228
+ "learning_rate": 4.6713662302786166e-05,
229
+ "loss": 2.8827,
230
+ "step": 3300
231
+ },
232
+ {
233
+ "epoch": 1.28,
234
+ "learning_rate": 4.624370594159114e-05,
235
+ "loss": 2.8261,
236
+ "step": 3400
237
+ },
238
+ {
239
+ "epoch": 1.32,
240
+ "learning_rate": 4.577374958039611e-05,
241
+ "loss": 2.9161,
242
+ "step": 3500
243
+ },
244
+ {
245
+ "epoch": 1.36,
246
+ "learning_rate": 4.5303793219201074e-05,
247
+ "loss": 2.8551,
248
+ "step": 3600
249
+ },
250
+ {
251
+ "epoch": 1.4,
252
+ "learning_rate": 4.483383685800604e-05,
253
+ "loss": 2.88,
254
+ "step": 3700
255
+ },
256
+ {
257
+ "epoch": 1.44,
258
+ "learning_rate": 4.436388049681101e-05,
259
+ "loss": 2.8753,
260
+ "step": 3800
261
+ },
262
+ {
263
+ "epoch": 1.47,
264
+ "learning_rate": 4.3893924135615974e-05,
265
+ "loss": 2.8922,
266
+ "step": 3900
267
+ },
268
+ {
269
+ "epoch": 1.51,
270
+ "learning_rate": 4.3423967774420946e-05,
271
+ "loss": 2.8591,
272
+ "step": 4000
273
+ },
274
+ {
275
+ "epoch": 1.51,
276
+ "eval_loss": 4.031496524810791,
277
+ "eval_runtime": 4.2566,
278
+ "eval_samples_per_second": 100.784,
279
+ "eval_steps_per_second": 6.343,
280
+ "step": 4000
281
+ },
282
+ {
283
+ "epoch": 1.55,
284
+ "learning_rate": 4.295401141322592e-05,
285
+ "loss": 2.8911,
286
+ "step": 4100
287
+ },
288
+ {
289
+ "epoch": 1.59,
290
+ "learning_rate": 4.248405505203088e-05,
291
+ "loss": 2.8629,
292
+ "step": 4200
293
+ },
294
+ {
295
+ "epoch": 1.62,
296
+ "learning_rate": 4.201409869083585e-05,
297
+ "loss": 2.8655,
298
+ "step": 4300
299
+ },
300
+ {
301
+ "epoch": 1.66,
302
+ "learning_rate": 4.154414232964082e-05,
303
+ "loss": 2.9125,
304
+ "step": 4400
305
+ },
306
+ {
307
+ "epoch": 1.7,
308
+ "learning_rate": 4.107418596844578e-05,
309
+ "loss": 2.9118,
310
+ "step": 4500
311
+ },
312
+ {
313
+ "epoch": 1.74,
314
+ "learning_rate": 4.0604229607250753e-05,
315
+ "loss": 2.8677,
316
+ "step": 4600
317
+ },
318
+ {
319
+ "epoch": 1.77,
320
+ "learning_rate": 4.0134273246055725e-05,
321
+ "loss": 2.8446,
322
+ "step": 4700
323
+ },
324
+ {
325
+ "epoch": 1.81,
326
+ "learning_rate": 3.9664316884860696e-05,
327
+ "loss": 2.8516,
328
+ "step": 4800
329
+ },
330
+ {
331
+ "epoch": 1.85,
332
+ "learning_rate": 3.919436052366566e-05,
333
+ "loss": 2.8291,
334
+ "step": 4900
335
+ },
336
+ {
337
+ "epoch": 1.89,
338
+ "learning_rate": 3.8724404162470625e-05,
339
+ "loss": 2.8866,
340
+ "step": 5000
341
+ },
342
+ {
343
+ "epoch": 1.89,
344
+ "eval_loss": 4.007871150970459,
345
+ "eval_runtime": 4.253,
346
+ "eval_samples_per_second": 100.869,
347
+ "eval_steps_per_second": 6.348,
348
+ "step": 5000
349
+ },
350
+ {
351
+ "epoch": 1.93,
352
+ "learning_rate": 3.82544478012756e-05,
353
+ "loss": 2.8411,
354
+ "step": 5100
355
+ },
356
+ {
357
+ "epoch": 1.96,
358
+ "learning_rate": 3.778449144008056e-05,
359
+ "loss": 2.8922,
360
+ "step": 5200
361
+ },
362
+ {
363
+ "epoch": 2.0,
364
+ "learning_rate": 3.7314535078885526e-05,
365
+ "loss": 2.8015,
366
+ "step": 5300
367
+ },
368
+ {
369
+ "epoch": 2.04,
370
+ "learning_rate": 3.6844578717690504e-05,
371
+ "loss": 2.591,
372
+ "step": 5400
373
+ },
374
+ {
375
+ "epoch": 2.08,
376
+ "learning_rate": 3.637462235649547e-05,
377
+ "loss": 2.6147,
378
+ "step": 5500
379
+ },
380
+ {
381
+ "epoch": 2.11,
382
+ "learning_rate": 3.590466599530044e-05,
383
+ "loss": 2.6515,
384
+ "step": 5600
385
+ },
386
+ {
387
+ "epoch": 2.15,
388
+ "learning_rate": 3.5434709634105405e-05,
389
+ "loss": 2.6206,
390
+ "step": 5700
391
+ },
392
+ {
393
+ "epoch": 2.19,
394
+ "learning_rate": 3.496475327291037e-05,
395
+ "loss": 2.5989,
396
+ "step": 5800
397
+ },
398
+ {
399
+ "epoch": 2.23,
400
+ "learning_rate": 3.449479691171534e-05,
401
+ "loss": 2.6168,
402
+ "step": 5900
403
+ },
404
+ {
405
+ "epoch": 2.27,
406
+ "learning_rate": 3.4024840550520305e-05,
407
+ "loss": 2.6146,
408
+ "step": 6000
409
+ },
410
+ {
411
+ "epoch": 2.27,
412
+ "eval_loss": 4.015293598175049,
413
+ "eval_runtime": 4.2602,
414
+ "eval_samples_per_second": 100.7,
415
+ "eval_steps_per_second": 6.338,
416
+ "step": 6000
417
+ },
418
+ {
419
+ "epoch": 2.3,
420
+ "learning_rate": 3.3554884189325276e-05,
421
+ "loss": 2.6718,
422
+ "step": 6100
423
+ },
424
+ {
425
+ "epoch": 2.34,
426
+ "learning_rate": 3.308492782813025e-05,
427
+ "loss": 2.6144,
428
+ "step": 6200
429
+ },
430
+ {
431
+ "epoch": 2.38,
432
+ "learning_rate": 3.261497146693521e-05,
433
+ "loss": 2.6326,
434
+ "step": 6300
435
+ },
436
+ {
437
+ "epoch": 2.42,
438
+ "learning_rate": 3.2145015105740184e-05,
439
+ "loss": 2.6431,
440
+ "step": 6400
441
+ },
442
+ {
443
+ "epoch": 2.45,
444
+ "learning_rate": 3.167505874454515e-05,
445
+ "loss": 2.6734,
446
+ "step": 6500
447
+ },
448
+ {
449
+ "epoch": 2.49,
450
+ "learning_rate": 3.120510238335011e-05,
451
+ "loss": 2.6928,
452
+ "step": 6600
453
+ },
454
+ {
455
+ "epoch": 2.53,
456
+ "learning_rate": 3.0735146022155084e-05,
457
+ "loss": 2.642,
458
+ "step": 6700
459
+ },
460
+ {
461
+ "epoch": 2.57,
462
+ "learning_rate": 3.0265189660960056e-05,
463
+ "loss": 2.6116,
464
+ "step": 6800
465
+ },
466
+ {
467
+ "epoch": 2.61,
468
+ "learning_rate": 2.9795233299765024e-05,
469
+ "loss": 2.663,
470
+ "step": 6900
471
+ },
472
+ {
473
+ "epoch": 2.64,
474
+ "learning_rate": 2.9325276938569988e-05,
475
+ "loss": 2.6576,
476
+ "step": 7000
477
+ },
478
+ {
479
+ "epoch": 2.64,
480
+ "eval_loss": 4.002060890197754,
481
+ "eval_runtime": 4.2577,
482
+ "eval_samples_per_second": 100.759,
483
+ "eval_steps_per_second": 6.341,
484
+ "step": 7000
485
+ },
486
+ {
487
+ "epoch": 2.68,
488
+ "learning_rate": 2.8855320577374956e-05,
489
+ "loss": 2.6327,
490
+ "step": 7100
491
+ },
492
+ {
493
+ "epoch": 2.72,
494
+ "learning_rate": 2.8385364216179924e-05,
495
+ "loss": 2.6684,
496
+ "step": 7200
497
+ },
498
+ {
499
+ "epoch": 2.76,
500
+ "learning_rate": 2.7915407854984896e-05,
501
+ "loss": 2.6826,
502
+ "step": 7300
503
+ },
504
+ {
505
+ "epoch": 2.79,
506
+ "learning_rate": 2.744545149378986e-05,
507
+ "loss": 2.6216,
508
+ "step": 7400
509
+ },
510
+ {
511
+ "epoch": 2.83,
512
+ "learning_rate": 2.6975495132594828e-05,
513
+ "loss": 2.62,
514
+ "step": 7500
515
+ },
516
+ {
517
+ "epoch": 2.87,
518
+ "learning_rate": 2.65055387713998e-05,
519
+ "loss": 2.6401,
520
+ "step": 7600
521
+ },
522
+ {
523
+ "epoch": 2.91,
524
+ "learning_rate": 2.6035582410204767e-05,
525
+ "loss": 2.6276,
526
+ "step": 7700
527
+ },
528
+ {
529
+ "epoch": 2.95,
530
+ "learning_rate": 2.5565626049009732e-05,
531
+ "loss": 2.6809,
532
+ "step": 7800
533
+ },
534
+ {
535
+ "epoch": 2.98,
536
+ "learning_rate": 2.5095669687814703e-05,
537
+ "loss": 2.6544,
538
+ "step": 7900
539
+ },
540
+ {
541
+ "epoch": 3.02,
542
+ "learning_rate": 2.462571332661967e-05,
543
+ "loss": 2.5361,
544
+ "step": 8000
545
+ },
546
+ {
547
+ "epoch": 3.02,
548
+ "eval_loss": 4.010578632354736,
549
+ "eval_runtime": 4.2557,
550
+ "eval_samples_per_second": 100.806,
551
+ "eval_steps_per_second": 6.344,
552
+ "step": 8000
553
+ },
554
+ {
555
+ "epoch": 3.06,
556
+ "learning_rate": 2.416045652903659e-05,
557
+ "loss": 2.4391,
558
+ "step": 8100
559
+ },
560
+ {
561
+ "epoch": 3.1,
562
+ "learning_rate": 2.3690500167841557e-05,
563
+ "loss": 2.4457,
564
+ "step": 8200
565
+ },
566
+ {
567
+ "epoch": 3.13,
568
+ "learning_rate": 2.3220543806646525e-05,
569
+ "loss": 2.4267,
570
+ "step": 8300
571
+ },
572
+ {
573
+ "epoch": 3.17,
574
+ "learning_rate": 2.2750587445451493e-05,
575
+ "loss": 2.4724,
576
+ "step": 8400
577
+ },
578
+ {
579
+ "epoch": 3.21,
580
+ "learning_rate": 2.228063108425646e-05,
581
+ "loss": 2.4534,
582
+ "step": 8500
583
+ },
584
+ {
585
+ "epoch": 3.25,
586
+ "learning_rate": 2.1810674723061432e-05,
587
+ "loss": 2.446,
588
+ "step": 8600
589
+ },
590
+ {
591
+ "epoch": 3.29,
592
+ "learning_rate": 2.1340718361866397e-05,
593
+ "loss": 2.4415,
594
+ "step": 8700
595
+ },
596
+ {
597
+ "epoch": 3.32,
598
+ "learning_rate": 2.0870762000671365e-05,
599
+ "loss": 2.4938,
600
+ "step": 8800
601
+ },
602
+ {
603
+ "epoch": 3.36,
604
+ "learning_rate": 2.0400805639476333e-05,
605
+ "loss": 2.4444,
606
+ "step": 8900
607
+ },
608
+ {
609
+ "epoch": 3.4,
610
+ "learning_rate": 1.9935548841893253e-05,
611
+ "loss": 2.4552,
612
+ "step": 9000
613
+ },
614
+ {
615
+ "epoch": 3.4,
616
+ "eval_loss": 4.0137410163879395,
617
+ "eval_runtime": 4.2578,
618
+ "eval_samples_per_second": 100.757,
619
+ "eval_steps_per_second": 6.341,
620
+ "step": 9000
621
+ },
622
+ {
623
+ "epoch": 3.44,
624
+ "learning_rate": 1.946559248069822e-05,
625
+ "loss": 2.4298,
626
+ "step": 9100
627
+ },
628
+ {
629
+ "epoch": 3.47,
630
+ "learning_rate": 1.899563611950319e-05,
631
+ "loss": 2.4714,
632
+ "step": 9200
633
+ },
634
+ {
635
+ "epoch": 3.51,
636
+ "learning_rate": 1.8525679758308157e-05,
637
+ "loss": 2.4545,
638
+ "step": 9300
639
+ },
640
+ {
641
+ "epoch": 3.55,
642
+ "learning_rate": 1.8055723397113125e-05,
643
+ "loss": 2.4551,
644
+ "step": 9400
645
+ },
646
+ {
647
+ "epoch": 3.59,
648
+ "learning_rate": 1.7585767035918093e-05,
649
+ "loss": 2.4403,
650
+ "step": 9500
651
+ },
652
+ {
653
+ "epoch": 3.63,
654
+ "learning_rate": 1.711581067472306e-05,
655
+ "loss": 2.4297,
656
+ "step": 9600
657
+ },
658
+ {
659
+ "epoch": 3.66,
660
+ "learning_rate": 1.664585431352803e-05,
661
+ "loss": 2.4709,
662
+ "step": 9700
663
+ },
664
+ {
665
+ "epoch": 3.7,
666
+ "learning_rate": 1.6175897952332997e-05,
667
+ "loss": 2.4781,
668
+ "step": 9800
669
+ },
670
+ {
671
+ "epoch": 3.74,
672
+ "learning_rate": 1.5705941591137965e-05,
673
+ "loss": 2.4719,
674
+ "step": 9900
675
+ },
676
+ {
677
+ "epoch": 3.78,
678
+ "learning_rate": 1.5235985229942935e-05,
679
+ "loss": 2.4595,
680
+ "step": 10000
681
+ },
682
+ {
683
+ "epoch": 3.78,
684
+ "eval_loss": 3.9967169761657715,
685
+ "eval_runtime": 4.254,
686
+ "eval_samples_per_second": 100.845,
687
+ "eval_steps_per_second": 6.347,
688
+ "step": 10000
689
+ },
690
+ {
691
+ "epoch": 3.81,
692
+ "learning_rate": 1.4766028868747901e-05,
693
+ "loss": 2.4654,
694
+ "step": 10100
695
+ },
696
+ {
697
+ "epoch": 3.85,
698
+ "learning_rate": 1.4296072507552869e-05,
699
+ "loss": 2.4776,
700
+ "step": 10200
701
+ },
702
+ {
703
+ "epoch": 3.89,
704
+ "learning_rate": 1.3826116146357837e-05,
705
+ "loss": 2.4636,
706
+ "step": 10300
707
+ },
708
+ {
709
+ "epoch": 3.93,
710
+ "learning_rate": 1.3356159785162807e-05,
711
+ "loss": 2.45,
712
+ "step": 10400
713
+ },
714
+ {
715
+ "epoch": 3.97,
716
+ "learning_rate": 1.2886203423967773e-05,
717
+ "loss": 2.4622,
718
+ "step": 10500
719
+ },
720
+ {
721
+ "epoch": 4.0,
722
+ "learning_rate": 1.2416247062772743e-05,
723
+ "loss": 2.4645,
724
+ "step": 10600
725
+ },
726
+ {
727
+ "epoch": 4.04,
728
+ "learning_rate": 1.194629070157771e-05,
729
+ "loss": 2.3412,
730
+ "step": 10700
731
+ },
732
+ {
733
+ "epoch": 4.08,
734
+ "learning_rate": 1.1476334340382679e-05,
735
+ "loss": 2.3713,
736
+ "step": 10800
737
+ },
738
+ {
739
+ "epoch": 4.12,
740
+ "learning_rate": 1.1006377979187647e-05,
741
+ "loss": 2.3126,
742
+ "step": 10900
743
+ },
744
+ {
745
+ "epoch": 4.15,
746
+ "learning_rate": 1.0536421617992615e-05,
747
+ "loss": 2.3286,
748
+ "step": 11000
749
+ },
750
+ {
751
+ "epoch": 4.15,
752
+ "eval_loss": 4.020402431488037,
753
+ "eval_runtime": 4.2569,
754
+ "eval_samples_per_second": 100.778,
755
+ "eval_steps_per_second": 6.343,
756
+ "step": 11000
757
+ },
758
+ {
759
+ "epoch": 4.19,
760
+ "learning_rate": 1.0066465256797583e-05,
761
+ "loss": 2.3153,
762
+ "step": 11100
763
+ },
764
+ {
765
+ "epoch": 4.23,
766
+ "learning_rate": 9.6012084592145e-06,
767
+ "loss": 2.3465,
768
+ "step": 11200
769
+ },
770
+ {
771
+ "epoch": 4.27,
772
+ "learning_rate": 9.13125209801947e-06,
773
+ "loss": 2.3226,
774
+ "step": 11300
775
+ },
776
+ {
777
+ "epoch": 4.31,
778
+ "learning_rate": 8.661295736824438e-06,
779
+ "loss": 2.3332,
780
+ "step": 11400
781
+ },
782
+ {
783
+ "epoch": 4.34,
784
+ "learning_rate": 8.191339375629406e-06,
785
+ "loss": 2.2948,
786
+ "step": 11500
787
+ },
788
+ {
789
+ "epoch": 4.38,
790
+ "learning_rate": 7.721383014434374e-06,
791
+ "loss": 2.2821,
792
+ "step": 11600
793
+ },
794
+ {
795
+ "epoch": 4.42,
796
+ "learning_rate": 7.251426653239342e-06,
797
+ "loss": 2.3112,
798
+ "step": 11700
799
+ },
800
+ {
801
+ "epoch": 4.46,
802
+ "learning_rate": 6.7814702920443104e-06,
803
+ "loss": 2.2696,
804
+ "step": 11800
805
+ },
806
+ {
807
+ "epoch": 4.49,
808
+ "learning_rate": 6.311513930849278e-06,
809
+ "loss": 2.3236,
810
+ "step": 11900
811
+ },
812
+ {
813
+ "epoch": 4.53,
814
+ "learning_rate": 5.841557569654246e-06,
815
+ "loss": 2.2805,
816
+ "step": 12000
817
+ },
818
+ {
819
+ "epoch": 4.53,
820
+ "eval_loss": 4.015416622161865,
821
+ "eval_runtime": 4.2597,
822
+ "eval_samples_per_second": 100.711,
823
+ "eval_steps_per_second": 6.338,
824
+ "step": 12000
825
+ },
826
+ {
827
+ "epoch": 4.57,
828
+ "learning_rate": 5.371601208459214e-06,
829
+ "loss": 2.3157,
830
+ "step": 12100
831
+ },
832
+ {
833
+ "epoch": 4.61,
834
+ "learning_rate": 4.901644847264182e-06,
835
+ "loss": 2.3227,
836
+ "step": 12200
837
+ },
838
+ {
839
+ "epoch": 4.65,
840
+ "learning_rate": 4.43168848606915e-06,
841
+ "loss": 2.3303,
842
+ "step": 12300
843
+ },
844
+ {
845
+ "epoch": 4.68,
846
+ "learning_rate": 3.961732124874118e-06,
847
+ "loss": 2.3228,
848
+ "step": 12400
849
+ },
850
+ {
851
+ "epoch": 4.72,
852
+ "learning_rate": 3.491775763679087e-06,
853
+ "loss": 2.3152,
854
+ "step": 12500
855
+ },
856
+ {
857
+ "epoch": 4.76,
858
+ "learning_rate": 3.021819402484055e-06,
859
+ "loss": 2.3273,
860
+ "step": 12600
861
+ },
862
+ {
863
+ "epoch": 4.8,
864
+ "learning_rate": 2.551863041289023e-06,
865
+ "loss": 2.3569,
866
+ "step": 12700
867
+ },
868
+ {
869
+ "epoch": 4.83,
870
+ "learning_rate": 2.0819066800939914e-06,
871
+ "loss": 2.3205,
872
+ "step": 12800
873
+ },
874
+ {
875
+ "epoch": 4.87,
876
+ "learning_rate": 1.6119503188989594e-06,
877
+ "loss": 2.2739,
878
+ "step": 12900
879
+ },
880
+ {
881
+ "epoch": 4.91,
882
+ "learning_rate": 1.1419939577039274e-06,
883
+ "loss": 2.3098,
884
+ "step": 13000
885
+ },
886
+ {
887
+ "epoch": 4.91,
888
+ "eval_loss": 4.012601852416992,
889
+ "eval_runtime": 4.2579,
890
+ "eval_samples_per_second": 100.754,
891
+ "eval_steps_per_second": 6.341,
892
+ "step": 13000
893
+ },
894
+ {
895
+ "epoch": 4.95,
896
+ "learning_rate": 6.720375965088956e-07,
897
+ "loss": 2.2791,
898
+ "step": 13100
899
+ },
900
+ {
901
+ "epoch": 4.98,
902
+ "learning_rate": 2.0208123531386372e-07,
903
+ "loss": 2.2955,
904
+ "step": 13200
905
+ },
906
+ {
907
+ "epoch": 5.0,
908
+ "step": 13240,
909
+ "total_flos": 5.535079759872e+16,
910
+ "train_loss": 2.6959276412548974,
911
+ "train_runtime": 6590.4176,
912
+ "train_samples_per_second": 32.143,
913
+ "train_steps_per_second": 2.009
914
+ }
915
+ ],
916
+ "max_steps": 13240,
917
+ "num_train_epochs": 5,
918
+ "total_flos": 5.535079759872e+16,
919
+ "trial_name": null,
920
+ "trial_params": null
921
+ }