pszemraj commited on
Commit
7ccb758
1 Parent(s): 635278d

End of training

Browse files
Files changed (4) hide show
  1. all_results.json +14 -0
  2. eval_results.json +9 -0
  3. train_results.json +8 -0
  4. trainer_state.json +1095 -0
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.99,
3
+ "eval_loss": 1.3089168071746826,
4
+ "eval_runtime": 6.4529,
5
+ "eval_samples": 444,
6
+ "eval_samples_per_second": 68.806,
7
+ "eval_steps_per_second": 17.201,
8
+ "perplexity": 3.702161384890143,
9
+ "train_loss": 1.7149053530259566,
10
+ "train_runtime": 770.4396,
11
+ "train_samples": 8516,
12
+ "train_samples_per_second": 110.534,
13
+ "train_steps_per_second": 0.428
14
+ }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.99,
3
+ "eval_loss": 1.3089168071746826,
4
+ "eval_runtime": 6.4529,
5
+ "eval_samples": 444,
6
+ "eval_samples_per_second": 68.806,
7
+ "eval_steps_per_second": 17.201,
8
+ "perplexity": 3.702161384890143
9
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.99,
3
+ "train_loss": 1.7149053530259566,
4
+ "train_runtime": 770.4396,
5
+ "train_samples": 8516,
6
+ "train_samples_per_second": 110.534,
7
+ "train_steps_per_second": 0.428
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1095 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 9.992481203007518,
5
+ "global_step": 330,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.06,
12
+ "learning_rate": 0.00011764705882352942,
13
+ "loss": 5.1019,
14
+ "step": 2
15
+ },
16
+ {
17
+ "epoch": 0.12,
18
+ "learning_rate": 0.00023529411764705883,
19
+ "loss": 4.8202,
20
+ "step": 4
21
+ },
22
+ {
23
+ "epoch": 0.18,
24
+ "learning_rate": 0.00035294117647058826,
25
+ "loss": 4.4144,
26
+ "step": 6
27
+ },
28
+ {
29
+ "epoch": 0.24,
30
+ "learning_rate": 0.00047058823529411766,
31
+ "loss": 4.0763,
32
+ "step": 8
33
+ },
34
+ {
35
+ "epoch": 0.3,
36
+ "learning_rate": 0.0005882352941176471,
37
+ "loss": 3.7584,
38
+ "step": 10
39
+ },
40
+ {
41
+ "epoch": 0.36,
42
+ "learning_rate": 0.0007058823529411765,
43
+ "loss": 3.583,
44
+ "step": 12
45
+ },
46
+ {
47
+ "epoch": 0.42,
48
+ "learning_rate": 0.0008235294117647058,
49
+ "loss": 3.4246,
50
+ "step": 14
51
+ },
52
+ {
53
+ "epoch": 0.48,
54
+ "learning_rate": 0.0009411764705882353,
55
+ "loss": 3.301,
56
+ "step": 16
57
+ },
58
+ {
59
+ "epoch": 0.54,
60
+ "learning_rate": 0.0009999748146823375,
61
+ "loss": 3.225,
62
+ "step": 18
63
+ },
64
+ {
65
+ "epoch": 0.6,
66
+ "learning_rate": 0.0009997733473639876,
67
+ "loss": 3.1705,
68
+ "step": 20
69
+ },
70
+ {
71
+ "epoch": 0.66,
72
+ "learning_rate": 0.0009993704939095377,
73
+ "loss": 3.0495,
74
+ "step": 22
75
+ },
76
+ {
77
+ "epoch": 0.72,
78
+ "learning_rate": 0.0009987664166507748,
79
+ "loss": 2.9806,
80
+ "step": 24
81
+ },
82
+ {
83
+ "epoch": 0.78,
84
+ "learning_rate": 0.0009979613590036108,
85
+ "loss": 2.9235,
86
+ "step": 26
87
+ },
88
+ {
89
+ "epoch": 0.84,
90
+ "learning_rate": 0.0009969556453699965,
91
+ "loss": 2.8419,
92
+ "step": 28
93
+ },
94
+ {
95
+ "epoch": 0.9,
96
+ "learning_rate": 0.0009957496810072027,
97
+ "loss": 2.7978,
98
+ "step": 30
99
+ },
100
+ {
101
+ "epoch": 0.96,
102
+ "learning_rate": 0.0009943439518645192,
103
+ "loss": 2.7061,
104
+ "step": 32
105
+ },
106
+ {
107
+ "epoch": 0.99,
108
+ "eval_loss": 2.585865020751953,
109
+ "eval_runtime": 6.5169,
110
+ "eval_samples_per_second": 68.13,
111
+ "eval_steps_per_second": 17.033,
112
+ "step": 33
113
+ },
114
+ {
115
+ "epoch": 1.03,
116
+ "learning_rate": 0.0009927390243874398,
117
+ "loss": 2.9771,
118
+ "step": 34
119
+ },
120
+ {
121
+ "epoch": 1.09,
122
+ "learning_rate": 0.0009909355452894098,
123
+ "loss": 2.5679,
124
+ "step": 36
125
+ },
126
+ {
127
+ "epoch": 1.15,
128
+ "learning_rate": 0.0009889342412912295,
129
+ "loss": 2.5551,
130
+ "step": 38
131
+ },
132
+ {
133
+ "epoch": 1.21,
134
+ "learning_rate": 0.0009867359188282193,
135
+ "loss": 2.4842,
136
+ "step": 40
137
+ },
138
+ {
139
+ "epoch": 1.27,
140
+ "learning_rate": 0.0009843414637252614,
141
+ "loss": 2.4689,
142
+ "step": 42
143
+ },
144
+ {
145
+ "epoch": 1.33,
146
+ "learning_rate": 0.0009817518408398536,
147
+ "loss": 2.4216,
148
+ "step": 44
149
+ },
150
+ {
151
+ "epoch": 1.39,
152
+ "learning_rate": 0.000978968093673314,
153
+ "loss": 2.4006,
154
+ "step": 46
155
+ },
156
+ {
157
+ "epoch": 1.45,
158
+ "learning_rate": 0.0009759913439502981,
159
+ "loss": 2.342,
160
+ "step": 48
161
+ },
162
+ {
163
+ "epoch": 1.51,
164
+ "learning_rate": 0.0009728227911667932,
165
+ "loss": 2.3004,
166
+ "step": 50
167
+ },
168
+ {
169
+ "epoch": 1.57,
170
+ "learning_rate": 0.0009694637121067764,
171
+ "loss": 2.2644,
172
+ "step": 52
173
+ },
174
+ {
175
+ "epoch": 1.63,
176
+ "learning_rate": 0.0009659154603277282,
177
+ "loss": 2.2406,
178
+ "step": 54
179
+ },
180
+ {
181
+ "epoch": 1.69,
182
+ "learning_rate": 0.0009621794656152091,
183
+ "loss": 2.2076,
184
+ "step": 56
185
+ },
186
+ {
187
+ "epoch": 1.75,
188
+ "learning_rate": 0.0009582572334067213,
189
+ "loss": 2.1834,
190
+ "step": 58
191
+ },
192
+ {
193
+ "epoch": 1.81,
194
+ "learning_rate": 0.0009541503441850843,
195
+ "loss": 2.1652,
196
+ "step": 60
197
+ },
198
+ {
199
+ "epoch": 1.87,
200
+ "learning_rate": 0.0009498604528415731,
201
+ "loss": 2.1353,
202
+ "step": 62
203
+ },
204
+ {
205
+ "epoch": 1.93,
206
+ "learning_rate": 0.0009453892880090695,
207
+ "loss": 2.1394,
208
+ "step": 64
209
+ },
210
+ {
211
+ "epoch": 1.99,
212
+ "learning_rate": 0.000940738651365503,
213
+ "loss": 2.08,
214
+ "step": 66
215
+ },
216
+ {
217
+ "epoch": 1.99,
218
+ "eval_loss": 1.996474027633667,
219
+ "eval_runtime": 6.333,
220
+ "eval_samples_per_second": 70.109,
221
+ "eval_steps_per_second": 17.527,
222
+ "step": 66
223
+ },
224
+ {
225
+ "epoch": 2.06,
226
+ "learning_rate": 0.000935910416907854,
227
+ "loss": 2.2925,
228
+ "step": 68
229
+ },
230
+ {
231
+ "epoch": 2.12,
232
+ "learning_rate": 0.0009309065301970192,
233
+ "loss": 2.0167,
234
+ "step": 70
235
+ },
236
+ {
237
+ "epoch": 2.18,
238
+ "learning_rate": 0.0009257290075738364,
239
+ "loss": 1.9594,
240
+ "step": 72
241
+ },
242
+ {
243
+ "epoch": 2.24,
244
+ "learning_rate": 0.0009203799353465918,
245
+ "loss": 1.9508,
246
+ "step": 74
247
+ },
248
+ {
249
+ "epoch": 2.3,
250
+ "learning_rate": 0.0009148614689503306,
251
+ "loss": 1.9579,
252
+ "step": 76
253
+ },
254
+ {
255
+ "epoch": 2.36,
256
+ "learning_rate": 0.0009091758320783139,
257
+ "loss": 1.9166,
258
+ "step": 78
259
+ },
260
+ {
261
+ "epoch": 2.42,
262
+ "learning_rate": 0.0009033253157859713,
263
+ "loss": 1.8802,
264
+ "step": 80
265
+ },
266
+ {
267
+ "epoch": 2.48,
268
+ "learning_rate": 0.0008973122775677078,
269
+ "loss": 1.8642,
270
+ "step": 82
271
+ },
272
+ {
273
+ "epoch": 2.54,
274
+ "learning_rate": 0.0008911391404069408,
275
+ "loss": 1.8552,
276
+ "step": 84
277
+ },
278
+ {
279
+ "epoch": 2.6,
280
+ "learning_rate": 0.0008848083917997462,
281
+ "loss": 1.8637,
282
+ "step": 86
283
+ },
284
+ {
285
+ "epoch": 2.66,
286
+ "learning_rate": 0.0008783225827525098,
287
+ "loss": 1.852,
288
+ "step": 88
289
+ },
290
+ {
291
+ "epoch": 2.72,
292
+ "learning_rate": 0.0008716843267539868,
293
+ "loss": 1.7914,
294
+ "step": 90
295
+ },
296
+ {
297
+ "epoch": 2.78,
298
+ "learning_rate": 0.0008648962987221837,
299
+ "loss": 1.8048,
300
+ "step": 92
301
+ },
302
+ {
303
+ "epoch": 2.84,
304
+ "learning_rate": 0.0008579612339264867,
305
+ "loss": 1.7966,
306
+ "step": 94
307
+ },
308
+ {
309
+ "epoch": 2.9,
310
+ "learning_rate": 0.0008508819268854713,
311
+ "loss": 1.7871,
312
+ "step": 96
313
+ },
314
+ {
315
+ "epoch": 2.96,
316
+ "learning_rate": 0.0008436612302408376,
317
+ "loss": 1.7623,
318
+ "step": 98
319
+ },
320
+ {
321
+ "epoch": 2.99,
322
+ "eval_loss": 1.724814772605896,
323
+ "eval_runtime": 7.0329,
324
+ "eval_samples_per_second": 63.132,
325
+ "eval_steps_per_second": 15.783,
326
+ "step": 99
327
+ },
328
+ {
329
+ "epoch": 3.03,
330
+ "learning_rate": 0.0008363020536079239,
331
+ "loss": 1.9929,
332
+ "step": 100
333
+ },
334
+ {
335
+ "epoch": 3.09,
336
+ "learning_rate": 0.0008288073624032633,
337
+ "loss": 1.7159,
338
+ "step": 102
339
+ },
340
+ {
341
+ "epoch": 3.15,
342
+ "learning_rate": 0.0008211801766496537,
343
+ "loss": 1.6946,
344
+ "step": 104
345
+ },
346
+ {
347
+ "epoch": 3.21,
348
+ "learning_rate": 0.000813423569759226,
349
+ "loss": 1.6397,
350
+ "step": 106
351
+ },
352
+ {
353
+ "epoch": 3.27,
354
+ "learning_rate": 0.0008055406672949956,
355
+ "loss": 1.669,
356
+ "step": 108
357
+ },
358
+ {
359
+ "epoch": 3.33,
360
+ "learning_rate": 0.0007975346457114034,
361
+ "loss": 1.6531,
362
+ "step": 110
363
+ },
364
+ {
365
+ "epoch": 3.39,
366
+ "learning_rate": 0.0007894087310743467,
367
+ "loss": 1.6478,
368
+ "step": 112
369
+ },
370
+ {
371
+ "epoch": 3.45,
372
+ "learning_rate": 0.0007811661977612201,
373
+ "loss": 1.6231,
374
+ "step": 114
375
+ },
376
+ {
377
+ "epoch": 3.51,
378
+ "learning_rate": 0.0007728103671414887,
379
+ "loss": 1.6478,
380
+ "step": 116
381
+ },
382
+ {
383
+ "epoch": 3.57,
384
+ "learning_rate": 0.0007643446062383273,
385
+ "loss": 1.6287,
386
+ "step": 118
387
+ },
388
+ {
389
+ "epoch": 3.63,
390
+ "learning_rate": 0.0007557723263718596,
391
+ "loss": 1.5939,
392
+ "step": 120
393
+ },
394
+ {
395
+ "epoch": 3.69,
396
+ "learning_rate": 0.0007470969817845518,
397
+ "loss": 1.6309,
398
+ "step": 122
399
+ },
400
+ {
401
+ "epoch": 3.75,
402
+ "learning_rate": 0.000738322068249308,
403
+ "loss": 1.5665,
404
+ "step": 124
405
+ },
406
+ {
407
+ "epoch": 3.81,
408
+ "learning_rate": 0.0007294511216608307,
409
+ "loss": 1.5953,
410
+ "step": 126
411
+ },
412
+ {
413
+ "epoch": 3.87,
414
+ "learning_rate": 0.0007204877166108151,
415
+ "loss": 1.5987,
416
+ "step": 128
417
+ },
418
+ {
419
+ "epoch": 3.93,
420
+ "learning_rate": 0.0007114354649475498,
421
+ "loss": 1.5961,
422
+ "step": 130
423
+ },
424
+ {
425
+ "epoch": 3.99,
426
+ "learning_rate": 0.0007022980143205046,
427
+ "loss": 1.5408,
428
+ "step": 132
429
+ },
430
+ {
431
+ "epoch": 3.99,
432
+ "eval_loss": 1.5449421405792236,
433
+ "eval_runtime": 7.1026,
434
+ "eval_samples_per_second": 62.512,
435
+ "eval_steps_per_second": 15.628,
436
+ "step": 132
437
+ },
438
+ {
439
+ "epoch": 4.06,
440
+ "learning_rate": 0.0006930790467104916,
441
+ "loss": 1.7394,
442
+ "step": 134
443
+ },
444
+ {
445
+ "epoch": 4.12,
446
+ "learning_rate": 0.0006837822769459941,
447
+ "loss": 1.5015,
448
+ "step": 136
449
+ },
450
+ {
451
+ "epoch": 4.18,
452
+ "learning_rate": 0.000674411451206257,
453
+ "loss": 1.4962,
454
+ "step": 138
455
+ },
456
+ {
457
+ "epoch": 4.24,
458
+ "learning_rate": 0.0006649703455117458,
459
+ "loss": 1.496,
460
+ "step": 140
461
+ },
462
+ {
463
+ "epoch": 4.3,
464
+ "learning_rate": 0.0006554627642025807,
465
+ "loss": 1.4703,
466
+ "step": 142
467
+ },
468
+ {
469
+ "epoch": 4.36,
470
+ "learning_rate": 0.0006458925384055585,
471
+ "loss": 1.474,
472
+ "step": 144
473
+ },
474
+ {
475
+ "epoch": 4.42,
476
+ "learning_rate": 0.0006362635244903819,
477
+ "loss": 1.4663,
478
+ "step": 146
479
+ },
480
+ {
481
+ "epoch": 4.48,
482
+ "learning_rate": 0.0006265796025157153,
483
+ "loss": 1.4556,
484
+ "step": 148
485
+ },
486
+ {
487
+ "epoch": 4.54,
488
+ "learning_rate": 0.0006168446746656973,
489
+ "loss": 1.4779,
490
+ "step": 150
491
+ },
492
+ {
493
+ "epoch": 4.6,
494
+ "learning_rate": 0.0006070626636775348,
495
+ "loss": 1.4687,
496
+ "step": 152
497
+ },
498
+ {
499
+ "epoch": 4.66,
500
+ "learning_rate": 0.0005972375112608181,
501
+ "loss": 1.4614,
502
+ "step": 154
503
+ },
504
+ {
505
+ "epoch": 4.72,
506
+ "learning_rate": 0.000587373176509189,
507
+ "loss": 1.4615,
508
+ "step": 156
509
+ },
510
+ {
511
+ "epoch": 4.78,
512
+ "learning_rate": 0.0005774736343050039,
513
+ "loss": 1.4479,
514
+ "step": 158
515
+ },
516
+ {
517
+ "epoch": 4.84,
518
+ "learning_rate": 0.0005675428737176367,
519
+ "loss": 1.427,
520
+ "step": 160
521
+ },
522
+ {
523
+ "epoch": 4.9,
524
+ "learning_rate": 0.000557584896396062,
525
+ "loss": 1.4327,
526
+ "step": 162
527
+ },
528
+ {
529
+ "epoch": 4.96,
530
+ "learning_rate": 0.0005476037149563726,
531
+ "loss": 1.4147,
532
+ "step": 164
533
+ },
534
+ {
535
+ "epoch": 4.99,
536
+ "eval_loss": 1.4437452554702759,
537
+ "eval_runtime": 6.9311,
538
+ "eval_samples_per_second": 64.059,
539
+ "eval_steps_per_second": 16.015,
540
+ "step": 165
541
+ },
542
+ {
543
+ "epoch": 5.03,
544
+ "learning_rate": 0.0005376033513648743,
545
+ "loss": 1.5806,
546
+ "step": 166
547
+ },
548
+ {
549
+ "epoch": 5.09,
550
+ "learning_rate": 0.0005275878353174165,
551
+ "loss": 1.3567,
552
+ "step": 168
553
+ },
554
+ {
555
+ "epoch": 5.15,
556
+ "learning_rate": 0.0005175612026156045,
557
+ "loss": 1.3639,
558
+ "step": 170
559
+ },
560
+ {
561
+ "epoch": 5.21,
562
+ "learning_rate": 0.0005075274935405553,
563
+ "loss": 1.3578,
564
+ "step": 172
565
+ },
566
+ {
567
+ "epoch": 5.27,
568
+ "learning_rate": 0.0004974907512248451,
569
+ "loss": 1.3787,
570
+ "step": 174
571
+ },
572
+ {
573
+ "epoch": 5.33,
574
+ "learning_rate": 0.0004874550200233085,
575
+ "loss": 1.3406,
576
+ "step": 176
577
+ },
578
+ {
579
+ "epoch": 5.39,
580
+ "learning_rate": 0.0004774243438833481,
581
+ "loss": 1.368,
582
+ "step": 178
583
+ },
584
+ {
585
+ "epoch": 5.45,
586
+ "learning_rate": 0.00046740276471540364,
587
+ "loss": 1.3549,
588
+ "step": 180
589
+ },
590
+ {
591
+ "epoch": 5.51,
592
+ "learning_rate": 0.00045739432076424515,
593
+ "loss": 1.3655,
594
+ "step": 182
595
+ },
596
+ {
597
+ "epoch": 5.57,
598
+ "learning_rate": 0.00044740304498174226,
599
+ "loss": 1.35,
600
+ "step": 184
601
+ },
602
+ {
603
+ "epoch": 5.63,
604
+ "learning_rate": 0.0004374329634017669,
605
+ "loss": 1.3604,
606
+ "step": 186
607
+ },
608
+ {
609
+ "epoch": 5.69,
610
+ "learning_rate": 0.00042748809351788165,
611
+ "loss": 1.3692,
612
+ "step": 188
613
+ },
614
+ {
615
+ "epoch": 5.75,
616
+ "learning_rate": 0.0004175724426644724,
617
+ "loss": 1.3231,
618
+ "step": 190
619
+ },
620
+ {
621
+ "epoch": 5.81,
622
+ "learning_rate": 0.00040769000640197205,
623
+ "loss": 1.3361,
624
+ "step": 192
625
+ },
626
+ {
627
+ "epoch": 5.87,
628
+ "learning_rate": 0.00039784476690683085,
629
+ "loss": 1.3391,
630
+ "step": 194
631
+ },
632
+ {
633
+ "epoch": 5.93,
634
+ "learning_rate": 0.0003880406913668777,
635
+ "loss": 1.3259,
636
+ "step": 196
637
+ },
638
+ {
639
+ "epoch": 5.99,
640
+ "learning_rate": 0.0003782817303827226,
641
+ "loss": 1.3593,
642
+ "step": 198
643
+ },
644
+ {
645
+ "epoch": 5.99,
646
+ "eval_loss": 1.3767662048339844,
647
+ "eval_runtime": 6.7366,
648
+ "eval_samples_per_second": 65.908,
649
+ "eval_steps_per_second": 16.477,
650
+ "step": 198
651
+ },
652
+ {
653
+ "epoch": 6.06,
654
+ "learning_rate": 0.0003685718163758427,
655
+ "loss": 1.4657,
656
+ "step": 200
657
+ },
658
+ {
659
+ "epoch": 6.12,
660
+ "learning_rate": 0.0003589148620039941,
661
+ "loss": 1.2816,
662
+ "step": 202
663
+ },
664
+ {
665
+ "epoch": 6.18,
666
+ "learning_rate": 0.00034931475858458635,
667
+ "loss": 1.2989,
668
+ "step": 204
669
+ },
670
+ {
671
+ "epoch": 6.24,
672
+ "learning_rate": 0.0003397753745266571,
673
+ "loss": 1.3021,
674
+ "step": 206
675
+ },
676
+ {
677
+ "epoch": 6.3,
678
+ "learning_rate": 0.0003303005537720778,
679
+ "loss": 1.2478,
680
+ "step": 208
681
+ },
682
+ {
683
+ "epoch": 6.36,
684
+ "learning_rate": 0.00032089411424661863,
685
+ "loss": 1.2827,
686
+ "step": 210
687
+ },
688
+ {
689
+ "epoch": 6.42,
690
+ "learning_rate": 0.0003115598463214956,
691
+ "loss": 1.2458,
692
+ "step": 212
693
+ },
694
+ {
695
+ "epoch": 6.48,
696
+ "learning_rate": 0.0003023015112860228,
697
+ "loss": 1.2954,
698
+ "step": 214
699
+ },
700
+ {
701
+ "epoch": 6.54,
702
+ "learning_rate": 0.00029312283983198097,
703
+ "loss": 1.2782,
704
+ "step": 216
705
+ },
706
+ {
707
+ "epoch": 6.6,
708
+ "learning_rate": 0.0002840275305503186,
709
+ "loss": 1.2653,
710
+ "step": 218
711
+ },
712
+ {
713
+ "epoch": 6.66,
714
+ "learning_rate": 0.00027501924844078535,
715
+ "loss": 1.2701,
716
+ "step": 220
717
+ },
718
+ {
719
+ "epoch": 6.72,
720
+ "learning_rate": 0.0002661016234351018,
721
+ "loss": 1.2862,
722
+ "step": 222
723
+ },
724
+ {
725
+ "epoch": 6.78,
726
+ "learning_rate": 0.00025727824893426166,
727
+ "loss": 1.277,
728
+ "step": 224
729
+ },
730
+ {
731
+ "epoch": 6.84,
732
+ "learning_rate": 0.00024855268036055346,
733
+ "loss": 1.2791,
734
+ "step": 226
735
+ },
736
+ {
737
+ "epoch": 6.9,
738
+ "learning_rate": 0.00023992843372488355,
739
+ "loss": 1.266,
740
+ "step": 228
741
+ },
742
+ {
743
+ "epoch": 6.96,
744
+ "learning_rate": 0.00023140898420998424,
745
+ "loss": 1.2703,
746
+ "step": 230
747
+ },
748
+ {
749
+ "epoch": 6.99,
750
+ "eval_loss": 1.3362174034118652,
751
+ "eval_runtime": 6.9977,
752
+ "eval_samples_per_second": 63.449,
753
+ "eval_steps_per_second": 15.862,
754
+ "step": 231
755
+ },
756
+ {
757
+ "epoch": 7.03,
758
+ "learning_rate": 0.0002229977647700707,
759
+ "loss": 1.4282,
760
+ "step": 232
761
+ },
762
+ {
763
+ "epoch": 7.09,
764
+ "learning_rate": 0.00021469816474751563,
765
+ "loss": 1.2356,
766
+ "step": 234
767
+ },
768
+ {
769
+ "epoch": 7.15,
770
+ "learning_rate": 0.00020651352850709653,
771
+ "loss": 1.247,
772
+ "step": 236
773
+ },
774
+ {
775
+ "epoch": 7.21,
776
+ "learning_rate": 0.00019844715408836789,
777
+ "loss": 1.2564,
778
+ "step": 238
779
+ },
780
+ {
781
+ "epoch": 7.27,
782
+ "learning_rate": 0.00019050229187669949,
783
+ "loss": 1.2187,
784
+ "step": 240
785
+ },
786
+ {
787
+ "epoch": 7.33,
788
+ "learning_rate": 0.00018268214329351796,
789
+ "loss": 1.2388,
790
+ "step": 242
791
+ },
792
+ {
793
+ "epoch": 7.39,
794
+ "learning_rate": 0.00017498985950627793,
795
+ "loss": 1.2368,
796
+ "step": 244
797
+ },
798
+ {
799
+ "epoch": 7.45,
800
+ "learning_rate": 0.00016742854015868347,
801
+ "loss": 1.212,
802
+ "step": 246
803
+ },
804
+ {
805
+ "epoch": 7.51,
806
+ "learning_rate": 0.00016000123212167155,
807
+ "loss": 1.2377,
808
+ "step": 248
809
+ },
810
+ {
811
+ "epoch": 7.57,
812
+ "learning_rate": 0.00015271092826566108,
813
+ "loss": 1.2146,
814
+ "step": 250
815
+ },
816
+ {
817
+ "epoch": 7.63,
818
+ "learning_rate": 0.0001455605662545592,
819
+ "loss": 1.2209,
820
+ "step": 252
821
+ },
822
+ {
823
+ "epoch": 7.69,
824
+ "learning_rate": 0.00013855302736201687,
825
+ "loss": 1.2319,
826
+ "step": 254
827
+ },
828
+ {
829
+ "epoch": 7.75,
830
+ "learning_rate": 0.00013169113531040461,
831
+ "loss": 1.2271,
832
+ "step": 256
833
+ },
834
+ {
835
+ "epoch": 7.81,
836
+ "learning_rate": 0.00012497765513297976,
837
+ "loss": 1.2021,
838
+ "step": 258
839
+ },
840
+ {
841
+ "epoch": 7.87,
842
+ "learning_rate": 0.00011841529205970281,
843
+ "loss": 1.2264,
844
+ "step": 260
845
+ },
846
+ {
847
+ "epoch": 7.93,
848
+ "learning_rate": 0.00011200669042715162,
849
+ "loss": 1.2228,
850
+ "step": 262
851
+ },
852
+ {
853
+ "epoch": 7.99,
854
+ "learning_rate": 0.00010575443261297229,
855
+ "loss": 1.2528,
856
+ "step": 264
857
+ },
858
+ {
859
+ "epoch": 7.99,
860
+ "eval_loss": 1.3175491094589233,
861
+ "eval_runtime": 6.7551,
862
+ "eval_samples_per_second": 65.729,
863
+ "eval_steps_per_second": 16.432,
864
+ "step": 264
865
+ },
866
+ {
867
+ "epoch": 8.06,
868
+ "learning_rate": 9.96610379952989e-05,
869
+ "loss": 1.3448,
870
+ "step": 266
871
+ },
872
+ {
873
+ "epoch": 8.12,
874
+ "learning_rate": 9.37289619375562e-05,
875
+ "loss": 1.1886,
876
+ "step": 268
877
+ },
878
+ {
879
+ "epoch": 8.18,
880
+ "learning_rate": 8.7960594799059e-05,
881
+ "loss": 1.2062,
882
+ "step": 270
883
+ },
884
+ {
885
+ "epoch": 8.24,
886
+ "learning_rate": 8.235826097180565e-05,
887
+ "loss": 1.2207,
888
+ "step": 272
889
+ },
890
+ {
891
+ "epoch": 8.3,
892
+ "learning_rate": 7.692421794385312e-05,
893
+ "loss": 1.2095,
894
+ "step": 274
895
+ },
896
+ {
897
+ "epoch": 8.36,
898
+ "learning_rate": 7.166065538964955e-05,
899
+ "loss": 1.1986,
900
+ "step": 276
901
+ },
902
+ {
903
+ "epoch": 8.42,
904
+ "learning_rate": 6.656969428769566e-05,
905
+ "loss": 1.1962,
906
+ "step": 278
907
+ },
908
+ {
909
+ "epoch": 8.48,
910
+ "learning_rate": 6.165338606588517e-05,
911
+ "loss": 1.2164,
912
+ "step": 280
913
+ },
914
+ {
915
+ "epoch": 8.54,
916
+ "learning_rate": 5.6913711774872144e-05,
917
+ "loss": 1.1904,
918
+ "step": 282
919
+ },
920
+ {
921
+ "epoch": 8.6,
922
+ "learning_rate": 5.235258128979675e-05,
923
+ "loss": 1.2172,
924
+ "step": 284
925
+ },
926
+ {
927
+ "epoch": 8.66,
928
+ "learning_rate": 4.797183254069176e-05,
929
+ "loss": 1.2345,
930
+ "step": 286
931
+ },
932
+ {
933
+ "epoch": 8.72,
934
+ "learning_rate": 4.3773230771879005e-05,
935
+ "loss": 1.1994,
936
+ "step": 288
937
+ },
938
+ {
939
+ "epoch": 8.78,
940
+ "learning_rate": 3.975846783065662e-05,
941
+ "loss": 1.1963,
942
+ "step": 290
943
+ },
944
+ {
945
+ "epoch": 8.84,
946
+ "learning_rate": 3.5929161485559694e-05,
947
+ "loss": 1.1995,
948
+ "step": 292
949
+ },
950
+ {
951
+ "epoch": 8.9,
952
+ "learning_rate": 3.2286854774472905e-05,
953
+ "loss": 1.1779,
954
+ "step": 294
955
+ },
956
+ {
957
+ "epoch": 8.96,
958
+ "learning_rate": 2.883301538285582e-05,
959
+ "loss": 1.1981,
960
+ "step": 296
961
+ },
962
+ {
963
+ "epoch": 8.99,
964
+ "eval_loss": 1.3090853691101074,
965
+ "eval_runtime": 6.8294,
966
+ "eval_samples_per_second": 65.013,
967
+ "eval_steps_per_second": 16.253,
968
+ "step": 297
969
+ },
970
+ {
971
+ "epoch": 9.03,
972
+ "learning_rate": 2.5569035052332156e-05,
973
+ "loss": 1.3461,
974
+ "step": 298
975
+ },
976
+ {
977
+ "epoch": 9.09,
978
+ "learning_rate": 2.2496229019879632e-05,
979
+ "loss": 1.1812,
980
+ "step": 300
981
+ },
982
+ {
983
+ "epoch": 9.15,
984
+ "learning_rate": 1.9615835487849675e-05,
985
+ "loss": 1.177,
986
+ "step": 302
987
+ },
988
+ {
989
+ "epoch": 9.21,
990
+ "learning_rate": 1.6929015125027312e-05,
991
+ "loss": 1.1856,
992
+ "step": 304
993
+ },
994
+ {
995
+ "epoch": 9.27,
996
+ "learning_rate": 1.443685059893396e-05,
997
+ "loss": 1.1984,
998
+ "step": 306
999
+ },
1000
+ {
1001
+ "epoch": 9.33,
1002
+ "learning_rate": 1.2140346139561276e-05,
1003
+ "loss": 1.1711,
1004
+ "step": 308
1005
+ },
1006
+ {
1007
+ "epoch": 9.39,
1008
+ "learning_rate": 1.0040427134711649e-05,
1009
+ "loss": 1.1905,
1010
+ "step": 310
1011
+ },
1012
+ {
1013
+ "epoch": 9.45,
1014
+ "learning_rate": 8.137939757108525e-06,
1015
+ "loss": 1.1924,
1016
+ "step": 312
1017
+ },
1018
+ {
1019
+ "epoch": 9.51,
1020
+ "learning_rate": 6.433650623427378e-06,
1021
+ "loss": 1.2033,
1022
+ "step": 314
1023
+ },
1024
+ {
1025
+ "epoch": 9.57,
1026
+ "learning_rate": 4.928246485383147e-06,
1027
+ "loss": 1.2159,
1028
+ "step": 316
1029
+ },
1030
+ {
1031
+ "epoch": 9.63,
1032
+ "learning_rate": 3.6223339530006004e-06,
1033
+ "loss": 1.1831,
1034
+ "step": 318
1035
+ },
1036
+ {
1037
+ "epoch": 9.69,
1038
+ "learning_rate": 2.516439250177749e-06,
1039
+ "loss": 1.2011,
1040
+ "step": 320
1041
+ },
1042
+ {
1043
+ "epoch": 9.75,
1044
+ "learning_rate": 1.611008002641412e-06,
1045
+ "loss": 1.1991,
1046
+ "step": 322
1047
+ },
1048
+ {
1049
+ "epoch": 9.81,
1050
+ "learning_rate": 9.064050583800221e-07,
1051
+ "loss": 1.1786,
1052
+ "step": 324
1053
+ },
1054
+ {
1055
+ "epoch": 9.87,
1056
+ "learning_rate": 4.029143406262259e-07,
1057
+ "loss": 1.1945,
1058
+ "step": 326
1059
+ },
1060
+ {
1061
+ "epoch": 9.93,
1062
+ "learning_rate": 1.0073873344895734e-07,
1063
+ "loss": 1.1972,
1064
+ "step": 328
1065
+ },
1066
+ {
1067
+ "epoch": 9.99,
1068
+ "learning_rate": 0.0,
1069
+ "loss": 1.2117,
1070
+ "step": 330
1071
+ },
1072
+ {
1073
+ "epoch": 9.99,
1074
+ "eval_loss": 1.3089168071746826,
1075
+ "eval_runtime": 6.8863,
1076
+ "eval_samples_per_second": 64.476,
1077
+ "eval_steps_per_second": 16.119,
1078
+ "step": 330
1079
+ },
1080
+ {
1081
+ "epoch": 9.99,
1082
+ "step": 330,
1083
+ "total_flos": 1.1112428209176576e+16,
1084
+ "train_loss": 1.7149053530259566,
1085
+ "train_runtime": 770.4396,
1086
+ "train_samples_per_second": 110.534,
1087
+ "train_steps_per_second": 0.428
1088
+ }
1089
+ ],
1090
+ "max_steps": 330,
1091
+ "num_train_epochs": 10,
1092
+ "total_flos": 1.1112428209176576e+16,
1093
+ "trial_name": null,
1094
+ "trial_params": null
1095
+ }