gary109 commited on
Commit
2c292b5
1 Parent(s): 7d6c336

End of training

Browse files
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_loss": 0.42307427525520325,
4
+ "eval_runtime": 223.2666,
5
+ "eval_samples": 1965,
6
+ "eval_samples_per_second": 8.801,
7
+ "eval_steps_per_second": 1.102,
8
+ "eval_wer": 0.1596944119277925,
9
+ "train_loss": 0.13009323823279229,
10
+ "train_runtime": 33247.4378,
11
+ "train_samples": 17685,
12
+ "train_samples_per_second": 5.319,
13
+ "train_steps_per_second": 0.042
14
+ }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_loss": 0.42307427525520325,
4
+ "eval_runtime": 223.2666,
5
+ "eval_samples": 1965,
6
+ "eval_samples_per_second": 8.801,
7
+ "eval_steps_per_second": 1.102,
8
+ "eval_wer": 0.1596944119277925
9
+ }
runs/Jun21_09-10-53_98a0f29bb8e1/events.out.tfevents.1655841123.98a0f29bb8e1.5897.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4bda05e05798290bd2cef685d83b8d2faa38d5044d1a3fd371d0df7ec10250d
3
+ size 358
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "train_loss": 0.13009323823279229,
4
+ "train_runtime": 33247.4378,
5
+ "train_samples": 17685,
6
+ "train_samples_per_second": 5.319,
7
+ "train_steps_per_second": 0.042
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,943 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.42307427525520325,
3
+ "best_model_checkpoint": "ai-light-dance_singing_ft_wav2vec2-large-xlsr-53-5gram-v4/checkpoint-552",
4
+ "epoch": 9.99864314789688,
5
+ "global_step": 1380,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.07,
12
+ "learning_rate": 8e-08,
13
+ "loss": 0.1429,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.14,
18
+ "learning_rate": 1.6e-07,
19
+ "loss": 0.1481,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.22,
24
+ "learning_rate": 2.4e-07,
25
+ "loss": 0.1462,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.29,
30
+ "learning_rate": 3.2e-07,
31
+ "loss": 0.1509,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.36,
36
+ "learning_rate": 4e-07,
37
+ "loss": 0.1472,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.43,
42
+ "learning_rate": 4.8e-07,
43
+ "loss": 0.1351,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.51,
48
+ "learning_rate": 5.6e-07,
49
+ "loss": 0.1424,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.58,
54
+ "learning_rate": 6.4e-07,
55
+ "loss": 0.1417,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.65,
60
+ "learning_rate": 7.2e-07,
61
+ "loss": 0.1396,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.72,
66
+ "learning_rate": 8e-07,
67
+ "loss": 0.1409,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.8,
72
+ "learning_rate": 8.799999999999999e-07,
73
+ "loss": 0.1424,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.87,
78
+ "learning_rate": 9.6e-07,
79
+ "loss": 0.1196,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.94,
84
+ "learning_rate": 1.04e-06,
85
+ "loss": 0.1335,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 1.0,
90
+ "eval_loss": 0.4255823493003845,
91
+ "eval_runtime": 221.8873,
92
+ "eval_samples_per_second": 8.856,
93
+ "eval_steps_per_second": 1.109,
94
+ "eval_wer": 0.16052615365658307,
95
+ "step": 138
96
+ },
97
+ {
98
+ "epoch": 1.01,
99
+ "learning_rate": 1.12e-06,
100
+ "loss": 0.1412,
101
+ "step": 140
102
+ },
103
+ {
104
+ "epoch": 1.09,
105
+ "learning_rate": 1.2e-06,
106
+ "loss": 0.1357,
107
+ "step": 150
108
+ },
109
+ {
110
+ "epoch": 1.16,
111
+ "learning_rate": 1.28e-06,
112
+ "loss": 0.1437,
113
+ "step": 160
114
+ },
115
+ {
116
+ "epoch": 1.23,
117
+ "learning_rate": 1.3600000000000001e-06,
118
+ "loss": 0.134,
119
+ "step": 170
120
+ },
121
+ {
122
+ "epoch": 1.3,
123
+ "learning_rate": 1.44e-06,
124
+ "loss": 0.1426,
125
+ "step": 180
126
+ },
127
+ {
128
+ "epoch": 1.38,
129
+ "learning_rate": 1.5199999999999998e-06,
130
+ "loss": 0.1353,
131
+ "step": 190
132
+ },
133
+ {
134
+ "epoch": 1.45,
135
+ "learning_rate": 1.6e-06,
136
+ "loss": 0.1406,
137
+ "step": 200
138
+ },
139
+ {
140
+ "epoch": 1.52,
141
+ "learning_rate": 1.6799999999999998e-06,
142
+ "loss": 0.1403,
143
+ "step": 210
144
+ },
145
+ {
146
+ "epoch": 1.59,
147
+ "learning_rate": 1.752e-06,
148
+ "loss": 0.136,
149
+ "step": 220
150
+ },
151
+ {
152
+ "epoch": 1.67,
153
+ "learning_rate": 1.832e-06,
154
+ "loss": 0.1392,
155
+ "step": 230
156
+ },
157
+ {
158
+ "epoch": 1.74,
159
+ "learning_rate": 1.9119999999999997e-06,
160
+ "loss": 0.1462,
161
+ "step": 240
162
+ },
163
+ {
164
+ "epoch": 1.81,
165
+ "learning_rate": 1.9919999999999997e-06,
166
+ "loss": 0.1266,
167
+ "step": 250
168
+ },
169
+ {
170
+ "epoch": 1.88,
171
+ "learning_rate": 2.072e-06,
172
+ "loss": 0.1296,
173
+ "step": 260
174
+ },
175
+ {
176
+ "epoch": 1.96,
177
+ "learning_rate": 2.152e-06,
178
+ "loss": 0.1288,
179
+ "step": 270
180
+ },
181
+ {
182
+ "epoch": 2.0,
183
+ "eval_loss": 0.42343059182167053,
184
+ "eval_runtime": 220.812,
185
+ "eval_samples_per_second": 8.899,
186
+ "eval_steps_per_second": 1.114,
187
+ "eval_wer": 0.1602181011644384,
188
+ "step": 276
189
+ },
190
+ {
191
+ "epoch": 2.03,
192
+ "learning_rate": 2.232e-06,
193
+ "loss": 0.1483,
194
+ "step": 280
195
+ },
196
+ {
197
+ "epoch": 2.1,
198
+ "learning_rate": 2.3119999999999996e-06,
199
+ "loss": 0.1344,
200
+ "step": 290
201
+ },
202
+ {
203
+ "epoch": 2.17,
204
+ "learning_rate": 2.3919999999999997e-06,
205
+ "loss": 0.1253,
206
+ "step": 300
207
+ },
208
+ {
209
+ "epoch": 2.25,
210
+ "learning_rate": 2.472e-06,
211
+ "loss": 0.1559,
212
+ "step": 310
213
+ },
214
+ {
215
+ "epoch": 2.32,
216
+ "learning_rate": 2.552e-06,
217
+ "loss": 0.1326,
218
+ "step": 320
219
+ },
220
+ {
221
+ "epoch": 2.39,
222
+ "learning_rate": 2.632e-06,
223
+ "loss": 0.1303,
224
+ "step": 330
225
+ },
226
+ {
227
+ "epoch": 2.46,
228
+ "learning_rate": 2.712e-06,
229
+ "loss": 0.131,
230
+ "step": 340
231
+ },
232
+ {
233
+ "epoch": 2.54,
234
+ "learning_rate": 2.7919999999999997e-06,
235
+ "loss": 0.1474,
236
+ "step": 350
237
+ },
238
+ {
239
+ "epoch": 2.61,
240
+ "learning_rate": 2.8719999999999998e-06,
241
+ "loss": 0.1302,
242
+ "step": 360
243
+ },
244
+ {
245
+ "epoch": 2.68,
246
+ "learning_rate": 2.952e-06,
247
+ "loss": 0.1304,
248
+ "step": 370
249
+ },
250
+ {
251
+ "epoch": 2.75,
252
+ "learning_rate": 3.032e-06,
253
+ "loss": 0.1452,
254
+ "step": 380
255
+ },
256
+ {
257
+ "epoch": 2.82,
258
+ "learning_rate": 3.112e-06,
259
+ "loss": 0.1344,
260
+ "step": 390
261
+ },
262
+ {
263
+ "epoch": 2.9,
264
+ "learning_rate": 3.192e-06,
265
+ "loss": 0.1307,
266
+ "step": 400
267
+ },
268
+ {
269
+ "epoch": 2.97,
270
+ "learning_rate": 3.2719999999999998e-06,
271
+ "loss": 0.1278,
272
+ "step": 410
273
+ },
274
+ {
275
+ "epoch": 3.0,
276
+ "eval_loss": 0.4242633581161499,
277
+ "eval_runtime": 220.7296,
278
+ "eval_samples_per_second": 8.902,
279
+ "eval_steps_per_second": 1.114,
280
+ "eval_wer": 0.15966360667857804,
281
+ "step": 414
282
+ },
283
+ {
284
+ "epoch": 3.04,
285
+ "learning_rate": 3.352e-06,
286
+ "loss": 0.1423,
287
+ "step": 420
288
+ },
289
+ {
290
+ "epoch": 3.12,
291
+ "learning_rate": 3.432e-06,
292
+ "loss": 0.1306,
293
+ "step": 430
294
+ },
295
+ {
296
+ "epoch": 3.19,
297
+ "learning_rate": 3.512e-06,
298
+ "loss": 0.1348,
299
+ "step": 440
300
+ },
301
+ {
302
+ "epoch": 3.26,
303
+ "learning_rate": 3.592e-06,
304
+ "loss": 0.1445,
305
+ "step": 450
306
+ },
307
+ {
308
+ "epoch": 3.33,
309
+ "learning_rate": 3.672e-06,
310
+ "loss": 0.1418,
311
+ "step": 460
312
+ },
313
+ {
314
+ "epoch": 3.41,
315
+ "learning_rate": 3.7519999999999994e-06,
316
+ "loss": 0.1356,
317
+ "step": 470
318
+ },
319
+ {
320
+ "epoch": 3.48,
321
+ "learning_rate": 3.832e-06,
322
+ "loss": 0.1291,
323
+ "step": 480
324
+ },
325
+ {
326
+ "epoch": 3.55,
327
+ "learning_rate": 3.912e-06,
328
+ "loss": 0.1318,
329
+ "step": 490
330
+ },
331
+ {
332
+ "epoch": 3.62,
333
+ "learning_rate": 3.992e-06,
334
+ "loss": 0.1342,
335
+ "step": 500
336
+ },
337
+ {
338
+ "epoch": 3.69,
339
+ "learning_rate": 3.9590909090909085e-06,
340
+ "loss": 0.1184,
341
+ "step": 510
342
+ },
343
+ {
344
+ "epoch": 3.77,
345
+ "learning_rate": 3.913636363636363e-06,
346
+ "loss": 0.1338,
347
+ "step": 520
348
+ },
349
+ {
350
+ "epoch": 3.84,
351
+ "learning_rate": 3.868181818181818e-06,
352
+ "loss": 0.1372,
353
+ "step": 530
354
+ },
355
+ {
356
+ "epoch": 3.91,
357
+ "learning_rate": 3.822727272727272e-06,
358
+ "loss": 0.127,
359
+ "step": 540
360
+ },
361
+ {
362
+ "epoch": 3.98,
363
+ "learning_rate": 3.7772727272727273e-06,
364
+ "loss": 0.1345,
365
+ "step": 550
366
+ },
367
+ {
368
+ "epoch": 4.0,
369
+ "eval_loss": 0.42307427525520325,
370
+ "eval_runtime": 221.8583,
371
+ "eval_samples_per_second": 8.857,
372
+ "eval_steps_per_second": 1.109,
373
+ "eval_wer": 0.1596944119277925,
374
+ "step": 552
375
+ },
376
+ {
377
+ "epoch": 4.06,
378
+ "learning_rate": 3.731818181818182e-06,
379
+ "loss": 0.1328,
380
+ "step": 560
381
+ },
382
+ {
383
+ "epoch": 4.13,
384
+ "learning_rate": 3.6863636363636365e-06,
385
+ "loss": 0.1305,
386
+ "step": 570
387
+ },
388
+ {
389
+ "epoch": 4.2,
390
+ "learning_rate": 3.640909090909091e-06,
391
+ "loss": 0.1344,
392
+ "step": 580
393
+ },
394
+ {
395
+ "epoch": 4.27,
396
+ "learning_rate": 3.595454545454545e-06,
397
+ "loss": 0.1362,
398
+ "step": 590
399
+ },
400
+ {
401
+ "epoch": 4.35,
402
+ "learning_rate": 3.5499999999999995e-06,
403
+ "loss": 0.1335,
404
+ "step": 600
405
+ },
406
+ {
407
+ "epoch": 4.42,
408
+ "learning_rate": 3.504545454545454e-06,
409
+ "loss": 0.1309,
410
+ "step": 610
411
+ },
412
+ {
413
+ "epoch": 4.49,
414
+ "learning_rate": 3.4590909090909087e-06,
415
+ "loss": 0.129,
416
+ "step": 620
417
+ },
418
+ {
419
+ "epoch": 4.56,
420
+ "learning_rate": 3.4136363636363633e-06,
421
+ "loss": 0.1321,
422
+ "step": 630
423
+ },
424
+ {
425
+ "epoch": 4.64,
426
+ "learning_rate": 3.368181818181818e-06,
427
+ "loss": 0.1276,
428
+ "step": 640
429
+ },
430
+ {
431
+ "epoch": 4.71,
432
+ "learning_rate": 3.3227272727272724e-06,
433
+ "loss": 0.1304,
434
+ "step": 650
435
+ },
436
+ {
437
+ "epoch": 4.78,
438
+ "learning_rate": 3.277272727272727e-06,
439
+ "loss": 0.1308,
440
+ "step": 660
441
+ },
442
+ {
443
+ "epoch": 4.85,
444
+ "learning_rate": 3.2318181818181816e-06,
445
+ "loss": 0.1257,
446
+ "step": 670
447
+ },
448
+ {
449
+ "epoch": 4.93,
450
+ "learning_rate": 3.1863636363636362e-06,
451
+ "loss": 0.1299,
452
+ "step": 680
453
+ },
454
+ {
455
+ "epoch": 5.0,
456
+ "learning_rate": 3.140909090909091e-06,
457
+ "loss": 0.1344,
458
+ "step": 690
459
+ },
460
+ {
461
+ "epoch": 5.0,
462
+ "eval_loss": 0.4246298372745514,
463
+ "eval_runtime": 221.0199,
464
+ "eval_samples_per_second": 8.891,
465
+ "eval_steps_per_second": 1.113,
466
+ "eval_wer": 0.15966360667857804,
467
+ "step": 690
468
+ },
469
+ {
470
+ "epoch": 5.07,
471
+ "learning_rate": 3.095454545454545e-06,
472
+ "loss": 0.1366,
473
+ "step": 700
474
+ },
475
+ {
476
+ "epoch": 5.14,
477
+ "learning_rate": 3.0499999999999996e-06,
478
+ "loss": 0.1345,
479
+ "step": 710
480
+ },
481
+ {
482
+ "epoch": 5.22,
483
+ "learning_rate": 3.004545454545454e-06,
484
+ "loss": 0.1286,
485
+ "step": 720
486
+ },
487
+ {
488
+ "epoch": 5.29,
489
+ "learning_rate": 2.959090909090909e-06,
490
+ "loss": 0.1336,
491
+ "step": 730
492
+ },
493
+ {
494
+ "epoch": 5.36,
495
+ "learning_rate": 2.9136363636363634e-06,
496
+ "loss": 0.1267,
497
+ "step": 740
498
+ },
499
+ {
500
+ "epoch": 5.43,
501
+ "learning_rate": 2.868181818181818e-06,
502
+ "loss": 0.1202,
503
+ "step": 750
504
+ },
505
+ {
506
+ "epoch": 5.51,
507
+ "learning_rate": 2.8227272727272726e-06,
508
+ "loss": 0.1252,
509
+ "step": 760
510
+ },
511
+ {
512
+ "epoch": 5.58,
513
+ "learning_rate": 2.777272727272727e-06,
514
+ "loss": 0.1257,
515
+ "step": 770
516
+ },
517
+ {
518
+ "epoch": 5.65,
519
+ "learning_rate": 2.7318181818181818e-06,
520
+ "loss": 0.1151,
521
+ "step": 780
522
+ },
523
+ {
524
+ "epoch": 5.72,
525
+ "learning_rate": 2.6863636363636364e-06,
526
+ "loss": 0.1254,
527
+ "step": 790
528
+ },
529
+ {
530
+ "epoch": 5.8,
531
+ "learning_rate": 2.640909090909091e-06,
532
+ "loss": 0.1289,
533
+ "step": 800
534
+ },
535
+ {
536
+ "epoch": 5.87,
537
+ "learning_rate": 2.595454545454545e-06,
538
+ "loss": 0.1289,
539
+ "step": 810
540
+ },
541
+ {
542
+ "epoch": 5.94,
543
+ "learning_rate": 2.5499999999999997e-06,
544
+ "loss": 0.1237,
545
+ "step": 820
546
+ },
547
+ {
548
+ "epoch": 6.0,
549
+ "eval_loss": 0.4279264509677887,
550
+ "eval_runtime": 221.1731,
551
+ "eval_samples_per_second": 8.884,
552
+ "eval_steps_per_second": 1.112,
553
+ "eval_wer": 0.15954038568172016,
554
+ "step": 828
555
+ },
556
+ {
557
+ "epoch": 6.01,
558
+ "learning_rate": 2.5045454545454543e-06,
559
+ "loss": 0.138,
560
+ "step": 830
561
+ },
562
+ {
563
+ "epoch": 6.09,
564
+ "learning_rate": 2.459090909090909e-06,
565
+ "loss": 0.1273,
566
+ "step": 840
567
+ },
568
+ {
569
+ "epoch": 6.16,
570
+ "learning_rate": 2.4136363636363635e-06,
571
+ "loss": 0.1264,
572
+ "step": 850
573
+ },
574
+ {
575
+ "epoch": 6.23,
576
+ "learning_rate": 2.368181818181818e-06,
577
+ "loss": 0.1172,
578
+ "step": 860
579
+ },
580
+ {
581
+ "epoch": 6.3,
582
+ "learning_rate": 2.3227272727272727e-06,
583
+ "loss": 0.1273,
584
+ "step": 870
585
+ },
586
+ {
587
+ "epoch": 6.38,
588
+ "learning_rate": 2.2772727272727273e-06,
589
+ "loss": 0.1186,
590
+ "step": 880
591
+ },
592
+ {
593
+ "epoch": 6.45,
594
+ "learning_rate": 2.231818181818182e-06,
595
+ "loss": 0.1229,
596
+ "step": 890
597
+ },
598
+ {
599
+ "epoch": 6.52,
600
+ "learning_rate": 2.1863636363636365e-06,
601
+ "loss": 0.1341,
602
+ "step": 900
603
+ },
604
+ {
605
+ "epoch": 6.59,
606
+ "learning_rate": 2.140909090909091e-06,
607
+ "loss": 0.125,
608
+ "step": 910
609
+ },
610
+ {
611
+ "epoch": 6.67,
612
+ "learning_rate": 2.0954545454545453e-06,
613
+ "loss": 0.124,
614
+ "step": 920
615
+ },
616
+ {
617
+ "epoch": 6.74,
618
+ "learning_rate": 2.05e-06,
619
+ "loss": 0.1303,
620
+ "step": 930
621
+ },
622
+ {
623
+ "epoch": 6.81,
624
+ "learning_rate": 2.0045454545454544e-06,
625
+ "loss": 0.1371,
626
+ "step": 940
627
+ },
628
+ {
629
+ "epoch": 6.88,
630
+ "learning_rate": 1.959090909090909e-06,
631
+ "loss": 0.115,
632
+ "step": 950
633
+ },
634
+ {
635
+ "epoch": 6.96,
636
+ "learning_rate": 1.9136363636363636e-06,
637
+ "loss": 0.1109,
638
+ "step": 960
639
+ },
640
+ {
641
+ "epoch": 7.0,
642
+ "eval_loss": 0.435354620218277,
643
+ "eval_runtime": 221.7806,
644
+ "eval_samples_per_second": 8.86,
645
+ "eval_steps_per_second": 1.109,
646
+ "eval_wer": 0.1573224077382786,
647
+ "step": 966
648
+ },
649
+ {
650
+ "epoch": 7.03,
651
+ "learning_rate": 1.868181818181818e-06,
652
+ "loss": 0.1374,
653
+ "step": 970
654
+ },
655
+ {
656
+ "epoch": 7.1,
657
+ "learning_rate": 1.8227272727272726e-06,
658
+ "loss": 0.1225,
659
+ "step": 980
660
+ },
661
+ {
662
+ "epoch": 7.17,
663
+ "learning_rate": 1.7772727272727272e-06,
664
+ "loss": 0.1217,
665
+ "step": 990
666
+ },
667
+ {
668
+ "epoch": 7.25,
669
+ "learning_rate": 1.7318181818181818e-06,
670
+ "loss": 0.1215,
671
+ "step": 1000
672
+ },
673
+ {
674
+ "epoch": 7.32,
675
+ "learning_rate": 1.6863636363636362e-06,
676
+ "loss": 0.1235,
677
+ "step": 1010
678
+ },
679
+ {
680
+ "epoch": 7.39,
681
+ "learning_rate": 1.6409090909090908e-06,
682
+ "loss": 0.1183,
683
+ "step": 1020
684
+ },
685
+ {
686
+ "epoch": 7.46,
687
+ "learning_rate": 1.5954545454545454e-06,
688
+ "loss": 0.1248,
689
+ "step": 1030
690
+ },
691
+ {
692
+ "epoch": 7.54,
693
+ "learning_rate": 1.55e-06,
694
+ "loss": 0.1274,
695
+ "step": 1040
696
+ },
697
+ {
698
+ "epoch": 7.61,
699
+ "learning_rate": 1.5045454545454546e-06,
700
+ "loss": 0.1287,
701
+ "step": 1050
702
+ },
703
+ {
704
+ "epoch": 7.68,
705
+ "learning_rate": 1.459090909090909e-06,
706
+ "loss": 0.127,
707
+ "step": 1060
708
+ },
709
+ {
710
+ "epoch": 7.75,
711
+ "learning_rate": 1.4136363636363636e-06,
712
+ "loss": 0.1281,
713
+ "step": 1070
714
+ },
715
+ {
716
+ "epoch": 7.82,
717
+ "learning_rate": 1.3681818181818182e-06,
718
+ "loss": 0.1223,
719
+ "step": 1080
720
+ },
721
+ {
722
+ "epoch": 7.9,
723
+ "learning_rate": 1.3227272727272727e-06,
724
+ "loss": 0.1163,
725
+ "step": 1090
726
+ },
727
+ {
728
+ "epoch": 7.97,
729
+ "learning_rate": 1.2772727272727273e-06,
730
+ "loss": 0.1247,
731
+ "step": 1100
732
+ },
733
+ {
734
+ "epoch": 8.0,
735
+ "eval_loss": 0.43181082606315613,
736
+ "eval_runtime": 222.5795,
737
+ "eval_samples_per_second": 8.828,
738
+ "eval_steps_per_second": 1.105,
739
+ "eval_wer": 0.15698354999691946,
740
+ "step": 1104
741
+ },
742
+ {
743
+ "epoch": 8.04,
744
+ "learning_rate": 1.231818181818182e-06,
745
+ "loss": 0.1235,
746
+ "step": 1110
747
+ },
748
+ {
749
+ "epoch": 8.12,
750
+ "learning_rate": 1.1863636363636363e-06,
751
+ "loss": 0.1212,
752
+ "step": 1120
753
+ },
754
+ {
755
+ "epoch": 8.19,
756
+ "learning_rate": 1.140909090909091e-06,
757
+ "loss": 0.1265,
758
+ "step": 1130
759
+ },
760
+ {
761
+ "epoch": 8.26,
762
+ "learning_rate": 1.0954545454545453e-06,
763
+ "loss": 0.1243,
764
+ "step": 1140
765
+ },
766
+ {
767
+ "epoch": 8.33,
768
+ "learning_rate": 1.05e-06,
769
+ "loss": 0.1187,
770
+ "step": 1150
771
+ },
772
+ {
773
+ "epoch": 8.41,
774
+ "learning_rate": 1.0045454545454545e-06,
775
+ "loss": 0.1182,
776
+ "step": 1160
777
+ },
778
+ {
779
+ "epoch": 8.48,
780
+ "learning_rate": 9.59090909090909e-07,
781
+ "loss": 0.1266,
782
+ "step": 1170
783
+ },
784
+ {
785
+ "epoch": 8.55,
786
+ "learning_rate": 9.136363636363637e-07,
787
+ "loss": 0.1288,
788
+ "step": 1180
789
+ },
790
+ {
791
+ "epoch": 8.62,
792
+ "learning_rate": 8.681818181818182e-07,
793
+ "loss": 0.1149,
794
+ "step": 1190
795
+ },
796
+ {
797
+ "epoch": 8.69,
798
+ "learning_rate": 8.227272727272727e-07,
799
+ "loss": 0.122,
800
+ "step": 1200
801
+ },
802
+ {
803
+ "epoch": 8.77,
804
+ "learning_rate": 7.772727272727272e-07,
805
+ "loss": 0.1195,
806
+ "step": 1210
807
+ },
808
+ {
809
+ "epoch": 8.84,
810
+ "learning_rate": 7.318181818181818e-07,
811
+ "loss": 0.1144,
812
+ "step": 1220
813
+ },
814
+ {
815
+ "epoch": 8.91,
816
+ "learning_rate": 6.863636363636363e-07,
817
+ "loss": 0.1185,
818
+ "step": 1230
819
+ },
820
+ {
821
+ "epoch": 8.98,
822
+ "learning_rate": 6.409090909090908e-07,
823
+ "loss": 0.1372,
824
+ "step": 1240
825
+ },
826
+ {
827
+ "epoch": 9.0,
828
+ "eval_loss": 0.4340789318084717,
829
+ "eval_runtime": 224.1668,
830
+ "eval_samples_per_second": 8.766,
831
+ "eval_steps_per_second": 1.097,
832
+ "eval_wer": 0.15726079723984968,
833
+ "step": 1242
834
+ },
835
+ {
836
+ "epoch": 9.06,
837
+ "learning_rate": 5.954545454545454e-07,
838
+ "loss": 0.1214,
839
+ "step": 1250
840
+ },
841
+ {
842
+ "epoch": 9.13,
843
+ "learning_rate": 5.5e-07,
844
+ "loss": 0.1147,
845
+ "step": 1260
846
+ },
847
+ {
848
+ "epoch": 9.2,
849
+ "learning_rate": 5.045454545454545e-07,
850
+ "loss": 0.1161,
851
+ "step": 1270
852
+ },
853
+ {
854
+ "epoch": 9.27,
855
+ "learning_rate": 4.5909090909090906e-07,
856
+ "loss": 0.1334,
857
+ "step": 1280
858
+ },
859
+ {
860
+ "epoch": 9.35,
861
+ "learning_rate": 4.136363636363636e-07,
862
+ "loss": 0.1165,
863
+ "step": 1290
864
+ },
865
+ {
866
+ "epoch": 9.42,
867
+ "learning_rate": 3.6818181818181815e-07,
868
+ "loss": 0.1171,
869
+ "step": 1300
870
+ },
871
+ {
872
+ "epoch": 9.49,
873
+ "learning_rate": 3.2272727272727274e-07,
874
+ "loss": 0.1327,
875
+ "step": 1310
876
+ },
877
+ {
878
+ "epoch": 9.56,
879
+ "learning_rate": 2.772727272727273e-07,
880
+ "loss": 0.1171,
881
+ "step": 1320
882
+ },
883
+ {
884
+ "epoch": 9.64,
885
+ "learning_rate": 2.318181818181818e-07,
886
+ "loss": 0.1204,
887
+ "step": 1330
888
+ },
889
+ {
890
+ "epoch": 9.71,
891
+ "learning_rate": 1.8636363636363637e-07,
892
+ "loss": 0.1298,
893
+ "step": 1340
894
+ },
895
+ {
896
+ "epoch": 9.78,
897
+ "learning_rate": 1.4090909090909089e-07,
898
+ "loss": 0.1305,
899
+ "step": 1350
900
+ },
901
+ {
902
+ "epoch": 9.85,
903
+ "learning_rate": 9.545454545454546e-08,
904
+ "loss": 0.1294,
905
+ "step": 1360
906
+ },
907
+ {
908
+ "epoch": 9.93,
909
+ "learning_rate": 5e-08,
910
+ "loss": 0.115,
911
+ "step": 1370
912
+ },
913
+ {
914
+ "epoch": 10.0,
915
+ "learning_rate": 4.545454545454545e-09,
916
+ "loss": 0.1256,
917
+ "step": 1380
918
+ },
919
+ {
920
+ "epoch": 10.0,
921
+ "eval_loss": 0.4328223168849945,
922
+ "eval_runtime": 221.2478,
923
+ "eval_samples_per_second": 8.881,
924
+ "eval_steps_per_second": 1.112,
925
+ "eval_wer": 0.1575072392335654,
926
+ "step": 1380
927
+ },
928
+ {
929
+ "epoch": 10.0,
930
+ "step": 1380,
931
+ "total_flos": 5.359319261116583e+19,
932
+ "train_loss": 0.13009323823279229,
933
+ "train_runtime": 33247.4378,
934
+ "train_samples_per_second": 5.319,
935
+ "train_steps_per_second": 0.042
936
+ }
937
+ ],
938
+ "max_steps": 1380,
939
+ "num_train_epochs": 10,
940
+ "total_flos": 5.359319261116583e+19,
941
+ "trial_name": null,
942
+ "trial_params": null
943
+ }