jimypbr commited on
Commit
2c3bc59
1 Parent(s): c6723ff

Training in progress, step 500

Browse files
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.0,
3
+ "train_loss": 1.088201597354519,
4
+ "train_runtime": 719.9384,
5
+ "train_samples": 88550,
6
+ "train_samples_per_second": 245.993,
7
+ "train_steps_per_second": 1.92
8
+ }
config.json CHANGED
@@ -26,7 +26,7 @@
26
  "position_biased_input": false,
27
  "relative_attention": true,
28
  "torch_dtype": "float16",
29
- "transformers_version": "4.18.0.dev0",
30
  "type_vocab_size": 0,
31
  "vocab_size": 50265
32
  }
 
26
  "position_biased_input": false,
27
  "relative_attention": true,
28
  "torch_dtype": "float16",
29
+ "transformers_version": "4.19.0.dev0",
30
  "type_vocab_size": 0,
31
  "vocab_size": 50265
32
  }
ipu_config.json CHANGED
@@ -28,6 +28,6 @@
28
  "replicated_tensor_sharding": true,
29
  "replication_factor": 4,
30
  "seed": 1984,
31
- "transformers_version": "4.18.0.dev0",
32
  "use_popdist": false
33
  }
 
28
  "replicated_tensor_sharding": true,
29
  "replication_factor": 4,
30
  "seed": 1984,
31
+ "transformers_version": "4.19.0.dev0",
32
  "use_popdist": false
33
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:26181949daf0a3b96f33978b8a607d3a4f8bf5308e8fb64db8072291df9a61bd
3
  size 277274547
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e1f09f6119a29f383a3d1ea74bf0c26235173cda966dd57182282e10454a2b6
3
  size 277274547
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.0,
3
+ "train_loss": 1.088201597354519,
4
+ "train_runtime": 719.9384,
5
+ "train_samples": 88550,
6
+ "train_samples_per_second": 245.993,
7
+ "train_steps_per_second": 1.92
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,853 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.0,
5
+ "global_step": 1382,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 1.7341040462427746e-06,
13
+ "loss": 6.1875,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.03,
18
+ "learning_rate": 3.468208092485549e-06,
19
+ "loss": 5.6434,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.04,
24
+ "learning_rate": 5.202312138728324e-06,
25
+ "loss": 5.1445,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.06,
30
+ "learning_rate": 6.936416184971098e-06,
31
+ "loss": 4.9082,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.07,
36
+ "learning_rate": 8.670520231213873e-06,
37
+ "loss": 4.6742,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.09,
42
+ "learning_rate": 1.0404624277456647e-05,
43
+ "loss": 4.0891,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.1,
48
+ "learning_rate": 1.2138728323699422e-05,
49
+ "loss": 3.7152,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.12,
54
+ "learning_rate": 1.3872832369942197e-05,
55
+ "loss": 3.3334,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.13,
60
+ "learning_rate": 1.560693641618497e-05,
61
+ "loss": 2.2468,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.14,
66
+ "learning_rate": 1.7341040462427746e-05,
67
+ "loss": 1.7937,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.16,
72
+ "learning_rate": 1.9075144508670522e-05,
73
+ "loss": 1.8321,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.17,
78
+ "learning_rate": 2.0809248554913295e-05,
79
+ "loss": 1.4568,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.19,
84
+ "learning_rate": 2.2543352601156068e-05,
85
+ "loss": 1.3031,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.2,
90
+ "learning_rate": 2.4277456647398844e-05,
91
+ "loss": 1.0809,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.22,
96
+ "learning_rate": 2.601156069364162e-05,
97
+ "loss": 1.4456,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.23,
102
+ "learning_rate": 2.7745664739884393e-05,
103
+ "loss": 1.5199,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.25,
108
+ "learning_rate": 2.9479768786127166e-05,
109
+ "loss": 0.9764,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.26,
114
+ "learning_rate": 3.121387283236994e-05,
115
+ "loss": 1.2064,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.27,
120
+ "learning_rate": 3.294797687861272e-05,
121
+ "loss": 1.0915,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.29,
126
+ "learning_rate": 3.468208092485549e-05,
127
+ "loss": 1.3032,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.3,
132
+ "learning_rate": 3.6416184971098265e-05,
133
+ "loss": 1.0709,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.32,
138
+ "learning_rate": 3.8150289017341044e-05,
139
+ "loss": 1.6387,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.33,
144
+ "learning_rate": 3.988439306358381e-05,
145
+ "loss": 1.2317,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.35,
150
+ "learning_rate": 4.161849710982659e-05,
151
+ "loss": 1.0637,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.36,
156
+ "learning_rate": 4.335260115606937e-05,
157
+ "loss": 1.0287,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.38,
162
+ "learning_rate": 4.5086705202312136e-05,
163
+ "loss": 1.074,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.39,
168
+ "learning_rate": 4.6820809248554915e-05,
169
+ "loss": 0.868,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.41,
174
+ "learning_rate": 4.855491329479769e-05,
175
+ "loss": 0.8482,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.42,
180
+ "learning_rate": 5.028901734104046e-05,
181
+ "loss": 1.2787,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.43,
186
+ "learning_rate": 5.202312138728324e-05,
187
+ "loss": 1.1221,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.45,
192
+ "learning_rate": 5.3757225433526014e-05,
193
+ "loss": 0.9836,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.46,
198
+ "learning_rate": 5.5491329479768787e-05,
199
+ "loss": 1.5154,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.48,
204
+ "learning_rate": 5.7225433526011566e-05,
205
+ "loss": 0.9228,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.49,
210
+ "learning_rate": 5.895953757225433e-05,
211
+ "loss": 0.5593,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.51,
216
+ "learning_rate": 5.976833976833977e-05,
217
+ "loss": 0.8908,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.52,
222
+ "learning_rate": 5.9189189189189195e-05,
223
+ "loss": 0.9526,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.54,
228
+ "learning_rate": 5.861003861003862e-05,
229
+ "loss": 1.2775,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.55,
234
+ "learning_rate": 5.8030888030888026e-05,
235
+ "loss": 1.1524,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.56,
240
+ "learning_rate": 5.745173745173745e-05,
241
+ "loss": 1.2414,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.58,
246
+ "learning_rate": 5.687258687258687e-05,
247
+ "loss": 1.1137,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.59,
252
+ "learning_rate": 5.629343629343629e-05,
253
+ "loss": 0.9212,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.61,
258
+ "learning_rate": 5.5714285714285715e-05,
259
+ "loss": 1.0503,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 0.62,
264
+ "learning_rate": 5.513513513513514e-05,
265
+ "loss": 0.9738,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 0.64,
270
+ "learning_rate": 5.455598455598455e-05,
271
+ "loss": 1.2703,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 0.65,
276
+ "learning_rate": 5.3976833976833975e-05,
277
+ "loss": 1.086,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 0.67,
282
+ "learning_rate": 5.33976833976834e-05,
283
+ "loss": 0.8879,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 0.68,
288
+ "learning_rate": 5.281853281853282e-05,
289
+ "loss": 0.8404,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 0.69,
294
+ "learning_rate": 5.223938223938224e-05,
295
+ "loss": 0.775,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 0.71,
300
+ "learning_rate": 5.166023166023166e-05,
301
+ "loss": 1.2602,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 0.72,
306
+ "learning_rate": 5.108108108108108e-05,
307
+ "loss": 1.2931,
308
+ "step": 500
309
+ },
310
+ {
311
+ "epoch": 0.74,
312
+ "learning_rate": 5.05019305019305e-05,
313
+ "loss": 1.1383,
314
+ "step": 510
315
+ },
316
+ {
317
+ "epoch": 0.75,
318
+ "learning_rate": 4.9922779922779924e-05,
319
+ "loss": 0.7263,
320
+ "step": 520
321
+ },
322
+ {
323
+ "epoch": 0.77,
324
+ "learning_rate": 4.9343629343629346e-05,
325
+ "loss": 0.5444,
326
+ "step": 530
327
+ },
328
+ {
329
+ "epoch": 0.78,
330
+ "learning_rate": 4.876447876447877e-05,
331
+ "loss": 0.8266,
332
+ "step": 540
333
+ },
334
+ {
335
+ "epoch": 0.8,
336
+ "learning_rate": 4.8185328185328184e-05,
337
+ "loss": 1.0101,
338
+ "step": 550
339
+ },
340
+ {
341
+ "epoch": 0.81,
342
+ "learning_rate": 4.7606177606177606e-05,
343
+ "loss": 0.8983,
344
+ "step": 560
345
+ },
346
+ {
347
+ "epoch": 0.82,
348
+ "learning_rate": 4.702702702702703e-05,
349
+ "loss": 0.5349,
350
+ "step": 570
351
+ },
352
+ {
353
+ "epoch": 0.84,
354
+ "learning_rate": 4.644787644787645e-05,
355
+ "loss": 0.7786,
356
+ "step": 580
357
+ },
358
+ {
359
+ "epoch": 0.85,
360
+ "learning_rate": 4.586872586872587e-05,
361
+ "loss": 0.8275,
362
+ "step": 590
363
+ },
364
+ {
365
+ "epoch": 0.87,
366
+ "learning_rate": 4.5289575289575295e-05,
367
+ "loss": 0.8844,
368
+ "step": 600
369
+ },
370
+ {
371
+ "epoch": 0.88,
372
+ "learning_rate": 4.471042471042471e-05,
373
+ "loss": 1.1885,
374
+ "step": 610
375
+ },
376
+ {
377
+ "epoch": 0.9,
378
+ "learning_rate": 4.413127413127413e-05,
379
+ "loss": 0.8453,
380
+ "step": 620
381
+ },
382
+ {
383
+ "epoch": 0.91,
384
+ "learning_rate": 4.3552123552123555e-05,
385
+ "loss": 0.7368,
386
+ "step": 630
387
+ },
388
+ {
389
+ "epoch": 0.93,
390
+ "learning_rate": 4.297297297297298e-05,
391
+ "loss": 1.0366,
392
+ "step": 640
393
+ },
394
+ {
395
+ "epoch": 0.94,
396
+ "learning_rate": 4.23938223938224e-05,
397
+ "loss": 0.4579,
398
+ "step": 650
399
+ },
400
+ {
401
+ "epoch": 0.96,
402
+ "learning_rate": 4.181467181467182e-05,
403
+ "loss": 0.873,
404
+ "step": 660
405
+ },
406
+ {
407
+ "epoch": 0.97,
408
+ "learning_rate": 4.123552123552124e-05,
409
+ "loss": 1.0861,
410
+ "step": 670
411
+ },
412
+ {
413
+ "epoch": 0.98,
414
+ "learning_rate": 4.065637065637066e-05,
415
+ "loss": 0.7572,
416
+ "step": 680
417
+ },
418
+ {
419
+ "epoch": 1.0,
420
+ "learning_rate": 4.0077220077220075e-05,
421
+ "loss": 0.7174,
422
+ "step": 690
423
+ },
424
+ {
425
+ "epoch": 1.01,
426
+ "learning_rate": 3.94980694980695e-05,
427
+ "loss": 0.4577,
428
+ "step": 700
429
+ },
430
+ {
431
+ "epoch": 1.03,
432
+ "learning_rate": 3.891891891891892e-05,
433
+ "loss": 0.6622,
434
+ "step": 710
435
+ },
436
+ {
437
+ "epoch": 1.04,
438
+ "learning_rate": 3.833976833976834e-05,
439
+ "loss": 0.3438,
440
+ "step": 720
441
+ },
442
+ {
443
+ "epoch": 1.06,
444
+ "learning_rate": 3.776061776061776e-05,
445
+ "loss": 0.7274,
446
+ "step": 730
447
+ },
448
+ {
449
+ "epoch": 1.07,
450
+ "learning_rate": 3.718146718146718e-05,
451
+ "loss": 0.6925,
452
+ "step": 740
453
+ },
454
+ {
455
+ "epoch": 1.09,
456
+ "learning_rate": 3.66023166023166e-05,
457
+ "loss": 0.6156,
458
+ "step": 750
459
+ },
460
+ {
461
+ "epoch": 1.1,
462
+ "learning_rate": 3.6023166023166024e-05,
463
+ "loss": 0.8509,
464
+ "step": 760
465
+ },
466
+ {
467
+ "epoch": 1.11,
468
+ "learning_rate": 3.5444015444015446e-05,
469
+ "loss": 0.5898,
470
+ "step": 770
471
+ },
472
+ {
473
+ "epoch": 1.13,
474
+ "learning_rate": 3.486486486486486e-05,
475
+ "loss": 0.7896,
476
+ "step": 780
477
+ },
478
+ {
479
+ "epoch": 1.14,
480
+ "learning_rate": 3.4285714285714284e-05,
481
+ "loss": 0.6095,
482
+ "step": 790
483
+ },
484
+ {
485
+ "epoch": 1.16,
486
+ "learning_rate": 3.3706563706563706e-05,
487
+ "loss": 0.7593,
488
+ "step": 800
489
+ },
490
+ {
491
+ "epoch": 1.17,
492
+ "learning_rate": 3.312741312741313e-05,
493
+ "loss": 0.7934,
494
+ "step": 810
495
+ },
496
+ {
497
+ "epoch": 1.19,
498
+ "learning_rate": 3.254826254826255e-05,
499
+ "loss": 0.6189,
500
+ "step": 820
501
+ },
502
+ {
503
+ "epoch": 1.2,
504
+ "learning_rate": 3.196911196911197e-05,
505
+ "loss": 0.8726,
506
+ "step": 830
507
+ },
508
+ {
509
+ "epoch": 1.22,
510
+ "learning_rate": 3.138996138996139e-05,
511
+ "loss": 1.0715,
512
+ "step": 840
513
+ },
514
+ {
515
+ "epoch": 1.23,
516
+ "learning_rate": 3.081081081081081e-05,
517
+ "loss": 0.8481,
518
+ "step": 850
519
+ },
520
+ {
521
+ "epoch": 1.24,
522
+ "learning_rate": 3.0231660231660233e-05,
523
+ "loss": 0.906,
524
+ "step": 860
525
+ },
526
+ {
527
+ "epoch": 1.26,
528
+ "learning_rate": 2.9652509652509655e-05,
529
+ "loss": 0.727,
530
+ "step": 870
531
+ },
532
+ {
533
+ "epoch": 1.27,
534
+ "learning_rate": 2.9073359073359074e-05,
535
+ "loss": 1.1817,
536
+ "step": 880
537
+ },
538
+ {
539
+ "epoch": 1.29,
540
+ "learning_rate": 2.8494208494208496e-05,
541
+ "loss": 0.8072,
542
+ "step": 890
543
+ },
544
+ {
545
+ "epoch": 1.3,
546
+ "learning_rate": 2.7915057915057918e-05,
547
+ "loss": 0.9308,
548
+ "step": 900
549
+ },
550
+ {
551
+ "epoch": 1.32,
552
+ "learning_rate": 2.7335907335907337e-05,
553
+ "loss": 0.2952,
554
+ "step": 910
555
+ },
556
+ {
557
+ "epoch": 1.33,
558
+ "learning_rate": 2.6756756756756756e-05,
559
+ "loss": 0.7024,
560
+ "step": 920
561
+ },
562
+ {
563
+ "epoch": 1.35,
564
+ "learning_rate": 2.6177606177606178e-05,
565
+ "loss": 0.6619,
566
+ "step": 930
567
+ },
568
+ {
569
+ "epoch": 1.36,
570
+ "learning_rate": 2.5598455598455597e-05,
571
+ "loss": 0.5304,
572
+ "step": 940
573
+ },
574
+ {
575
+ "epoch": 1.37,
576
+ "learning_rate": 2.501930501930502e-05,
577
+ "loss": 0.4559,
578
+ "step": 950
579
+ },
580
+ {
581
+ "epoch": 1.39,
582
+ "learning_rate": 2.444015444015444e-05,
583
+ "loss": 0.7422,
584
+ "step": 960
585
+ },
586
+ {
587
+ "epoch": 1.4,
588
+ "learning_rate": 2.386100386100386e-05,
589
+ "loss": 0.7257,
590
+ "step": 970
591
+ },
592
+ {
593
+ "epoch": 1.42,
594
+ "learning_rate": 2.3281853281853282e-05,
595
+ "loss": 0.6573,
596
+ "step": 980
597
+ },
598
+ {
599
+ "epoch": 1.43,
600
+ "learning_rate": 2.2702702702702705e-05,
601
+ "loss": 0.7282,
602
+ "step": 990
603
+ },
604
+ {
605
+ "epoch": 1.45,
606
+ "learning_rate": 2.2123552123552123e-05,
607
+ "loss": 0.6065,
608
+ "step": 1000
609
+ },
610
+ {
611
+ "epoch": 1.46,
612
+ "learning_rate": 2.1544401544401546e-05,
613
+ "loss": 0.5585,
614
+ "step": 1010
615
+ },
616
+ {
617
+ "epoch": 1.48,
618
+ "learning_rate": 2.0965250965250968e-05,
619
+ "loss": 0.8342,
620
+ "step": 1020
621
+ },
622
+ {
623
+ "epoch": 1.49,
624
+ "learning_rate": 2.0386100386100387e-05,
625
+ "loss": 0.7599,
626
+ "step": 1030
627
+ },
628
+ {
629
+ "epoch": 1.51,
630
+ "learning_rate": 1.980694980694981e-05,
631
+ "loss": 0.7969,
632
+ "step": 1040
633
+ },
634
+ {
635
+ "epoch": 1.52,
636
+ "learning_rate": 1.922779922779923e-05,
637
+ "loss": 0.5474,
638
+ "step": 1050
639
+ },
640
+ {
641
+ "epoch": 1.53,
642
+ "learning_rate": 1.864864864864865e-05,
643
+ "loss": 0.8421,
644
+ "step": 1060
645
+ },
646
+ {
647
+ "epoch": 1.55,
648
+ "learning_rate": 1.806949806949807e-05,
649
+ "loss": 0.565,
650
+ "step": 1070
651
+ },
652
+ {
653
+ "epoch": 1.56,
654
+ "learning_rate": 1.7490347490347488e-05,
655
+ "loss": 0.3672,
656
+ "step": 1080
657
+ },
658
+ {
659
+ "epoch": 1.58,
660
+ "learning_rate": 1.691119691119691e-05,
661
+ "loss": 0.505,
662
+ "step": 1090
663
+ },
664
+ {
665
+ "epoch": 1.59,
666
+ "learning_rate": 1.6332046332046332e-05,
667
+ "loss": 0.5057,
668
+ "step": 1100
669
+ },
670
+ {
671
+ "epoch": 1.61,
672
+ "learning_rate": 1.575289575289575e-05,
673
+ "loss": 0.7104,
674
+ "step": 1110
675
+ },
676
+ {
677
+ "epoch": 1.62,
678
+ "learning_rate": 1.5173745173745173e-05,
679
+ "loss": 0.6253,
680
+ "step": 1120
681
+ },
682
+ {
683
+ "epoch": 1.64,
684
+ "learning_rate": 1.4594594594594596e-05,
685
+ "loss": 0.9309,
686
+ "step": 1130
687
+ },
688
+ {
689
+ "epoch": 1.65,
690
+ "learning_rate": 1.4015444015444016e-05,
691
+ "loss": 0.7422,
692
+ "step": 1140
693
+ },
694
+ {
695
+ "epoch": 1.66,
696
+ "learning_rate": 1.3436293436293437e-05,
697
+ "loss": 0.6471,
698
+ "step": 1150
699
+ },
700
+ {
701
+ "epoch": 1.68,
702
+ "learning_rate": 1.2857142857142857e-05,
703
+ "loss": 0.7338,
704
+ "step": 1160
705
+ },
706
+ {
707
+ "epoch": 1.69,
708
+ "learning_rate": 1.227799227799228e-05,
709
+ "loss": 0.874,
710
+ "step": 1170
711
+ },
712
+ {
713
+ "epoch": 1.71,
714
+ "learning_rate": 1.1698841698841698e-05,
715
+ "loss": 0.7624,
716
+ "step": 1180
717
+ },
718
+ {
719
+ "epoch": 1.72,
720
+ "learning_rate": 1.1119691119691119e-05,
721
+ "loss": 0.7777,
722
+ "step": 1190
723
+ },
724
+ {
725
+ "epoch": 1.74,
726
+ "learning_rate": 1.0540540540540541e-05,
727
+ "loss": 0.8091,
728
+ "step": 1200
729
+ },
730
+ {
731
+ "epoch": 1.75,
732
+ "learning_rate": 9.961389961389962e-06,
733
+ "loss": 0.9177,
734
+ "step": 1210
735
+ },
736
+ {
737
+ "epoch": 1.77,
738
+ "learning_rate": 9.382239382239382e-06,
739
+ "loss": 1.0223,
740
+ "step": 1220
741
+ },
742
+ {
743
+ "epoch": 1.78,
744
+ "learning_rate": 8.803088803088804e-06,
745
+ "loss": 0.2531,
746
+ "step": 1230
747
+ },
748
+ {
749
+ "epoch": 1.79,
750
+ "learning_rate": 8.223938223938225e-06,
751
+ "loss": 0.6171,
752
+ "step": 1240
753
+ },
754
+ {
755
+ "epoch": 1.81,
756
+ "learning_rate": 7.644787644787645e-06,
757
+ "loss": 0.6996,
758
+ "step": 1250
759
+ },
760
+ {
761
+ "epoch": 1.82,
762
+ "learning_rate": 7.065637065637066e-06,
763
+ "loss": 0.5205,
764
+ "step": 1260
765
+ },
766
+ {
767
+ "epoch": 1.84,
768
+ "learning_rate": 6.486486486486487e-06,
769
+ "loss": 0.65,
770
+ "step": 1270
771
+ },
772
+ {
773
+ "epoch": 1.85,
774
+ "learning_rate": 5.907335907335907e-06,
775
+ "loss": 0.5435,
776
+ "step": 1280
777
+ },
778
+ {
779
+ "epoch": 1.87,
780
+ "learning_rate": 5.3281853281853285e-06,
781
+ "loss": 0.5957,
782
+ "step": 1290
783
+ },
784
+ {
785
+ "epoch": 1.88,
786
+ "learning_rate": 4.749034749034749e-06,
787
+ "loss": 0.5854,
788
+ "step": 1300
789
+ },
790
+ {
791
+ "epoch": 1.9,
792
+ "learning_rate": 4.1698841698841696e-06,
793
+ "loss": 0.779,
794
+ "step": 1310
795
+ },
796
+ {
797
+ "epoch": 1.91,
798
+ "learning_rate": 3.5907335907335905e-06,
799
+ "loss": 0.4551,
800
+ "step": 1320
801
+ },
802
+ {
803
+ "epoch": 1.92,
804
+ "learning_rate": 3.0115830115830115e-06,
805
+ "loss": 0.7719,
806
+ "step": 1330
807
+ },
808
+ {
809
+ "epoch": 1.94,
810
+ "learning_rate": 2.4324324324324325e-06,
811
+ "loss": 0.5182,
812
+ "step": 1340
813
+ },
814
+ {
815
+ "epoch": 1.95,
816
+ "learning_rate": 1.8532818532818534e-06,
817
+ "loss": 0.6974,
818
+ "step": 1350
819
+ },
820
+ {
821
+ "epoch": 1.97,
822
+ "learning_rate": 1.2741312741312742e-06,
823
+ "loss": 0.5345,
824
+ "step": 1360
825
+ },
826
+ {
827
+ "epoch": 1.98,
828
+ "learning_rate": 6.949806949806949e-07,
829
+ "loss": 0.6196,
830
+ "step": 1370
831
+ },
832
+ {
833
+ "epoch": 2.0,
834
+ "learning_rate": 1.1583011583011584e-07,
835
+ "loss": 0.6318,
836
+ "step": 1380
837
+ },
838
+ {
839
+ "epoch": 2.0,
840
+ "step": 1382,
841
+ "total_flos": 5.175834887457866e+18,
842
+ "train_loss": 1.088201597354519,
843
+ "train_runtime": 719.9384,
844
+ "train_samples_per_second": 245.993,
845
+ "train_steps_per_second": 1.92
846
+ }
847
+ ],
848
+ "max_steps": 1382,
849
+ "num_train_epochs": 2,
850
+ "total_flos": 5.175834887457866e+18,
851
+ "trial_name": null,
852
+ "trial_params": null
853
+ }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fc837e7a00a18e38df1f54977c690bcfafbbd010d97f787ceaf49b8bf70387c1
3
  size 2671
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06353b604a041d059d0084a629fe746f6812de687acde9d950add2d342ff0558
3
  size 2671