gary109 commited on
Commit
3a67842
1 Parent(s): a1c9dbf

End of training

Browse files
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 100.0,
3
+ "eval_loss": 2.332979202270508,
4
+ "eval_runtime": 0.5143,
5
+ "eval_samples": 2,
6
+ "eval_samples_per_second": 3.889,
7
+ "eval_steps_per_second": 1.944,
8
+ "eval_wer": 1.0,
9
+ "train_loss": 9.722876396179199,
10
+ "train_runtime": 860.4357,
11
+ "train_samples": 16,
12
+ "train_samples_per_second": 1.86,
13
+ "train_steps_per_second": 0.116
14
+ }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 100.0,
3
+ "eval_loss": 2.332979202270508,
4
+ "eval_runtime": 0.5143,
5
+ "eval_samples": 2,
6
+ "eval_samples_per_second": 3.889,
7
+ "eval_steps_per_second": 1.944,
8
+ "eval_wer": 1.0
9
+ }
runs/Nov25_14-52-26_fe9cd2661e6d/events.out.tfevents.1669389300.fe9cd2661e6d.2338.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4adfcd38b51eb7a9674d36f2b9e780cd9c6a4abcfb78991e4d74ebcd06001771
3
+ size 352
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 100.0,
3
+ "train_loss": 9.722876396179199,
4
+ "train_runtime": 860.4357,
5
+ "train_samples": 16,
6
+ "train_samples_per_second": 1.86,
7
+ "train_steps_per_second": 0.116
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,985 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 2.332979202270508,
3
+ "best_model_checkpoint": "ai-light-dance_drums_ft_pretrain_wav2vec2-base-new_onset-rbma13-2_7k/checkpoint-40",
4
+ "epoch": 100.0,
5
+ "global_step": 100,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 1.0,
12
+ "eval_loss": 68.1357650756836,
13
+ "eval_runtime": 0.3976,
14
+ "eval_samples_per_second": 5.03,
15
+ "eval_steps_per_second": 2.515,
16
+ "eval_wer": 1.0,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 2.0,
21
+ "eval_loss": 68.1357650756836,
22
+ "eval_runtime": 0.4117,
23
+ "eval_samples_per_second": 4.858,
24
+ "eval_steps_per_second": 2.429,
25
+ "eval_wer": 1.0,
26
+ "step": 2
27
+ },
28
+ {
29
+ "epoch": 3.0,
30
+ "eval_loss": 68.1357650756836,
31
+ "eval_runtime": 0.4341,
32
+ "eval_samples_per_second": 4.607,
33
+ "eval_steps_per_second": 2.304,
34
+ "eval_wer": 1.0,
35
+ "step": 3
36
+ },
37
+ {
38
+ "epoch": 4.0,
39
+ "eval_loss": 68.02447509765625,
40
+ "eval_runtime": 0.4095,
41
+ "eval_samples_per_second": 4.884,
42
+ "eval_steps_per_second": 2.442,
43
+ "eval_wer": 1.0,
44
+ "step": 4
45
+ },
46
+ {
47
+ "epoch": 5.0,
48
+ "eval_loss": 67.78739166259766,
49
+ "eval_runtime": 0.4118,
50
+ "eval_samples_per_second": 4.857,
51
+ "eval_steps_per_second": 2.428,
52
+ "eval_wer": 1.0,
53
+ "step": 5
54
+ },
55
+ {
56
+ "epoch": 6.0,
57
+ "eval_loss": 67.45346069335938,
58
+ "eval_runtime": 0.4145,
59
+ "eval_samples_per_second": 4.825,
60
+ "eval_steps_per_second": 2.413,
61
+ "eval_wer": 1.0,
62
+ "step": 6
63
+ },
64
+ {
65
+ "epoch": 7.0,
66
+ "eval_loss": 67.01423645019531,
67
+ "eval_runtime": 0.4121,
68
+ "eval_samples_per_second": 4.853,
69
+ "eval_steps_per_second": 2.427,
70
+ "eval_wer": 1.0,
71
+ "step": 7
72
+ },
73
+ {
74
+ "epoch": 8.0,
75
+ "eval_loss": 67.01423645019531,
76
+ "eval_runtime": 0.4095,
77
+ "eval_samples_per_second": 4.884,
78
+ "eval_steps_per_second": 2.442,
79
+ "eval_wer": 1.0,
80
+ "step": 8
81
+ },
82
+ {
83
+ "epoch": 9.0,
84
+ "eval_loss": 66.4334716796875,
85
+ "eval_runtime": 0.4143,
86
+ "eval_samples_per_second": 4.828,
87
+ "eval_steps_per_second": 2.414,
88
+ "eval_wer": 1.0,
89
+ "step": 9
90
+ },
91
+ {
92
+ "epoch": 10.0,
93
+ "learning_rate": 7e-05,
94
+ "loss": 38.4011,
95
+ "step": 10
96
+ },
97
+ {
98
+ "epoch": 10.0,
99
+ "eval_loss": 65.70997619628906,
100
+ "eval_runtime": 0.4061,
101
+ "eval_samples_per_second": 4.925,
102
+ "eval_steps_per_second": 2.463,
103
+ "eval_wer": 1.0,
104
+ "step": 10
105
+ },
106
+ {
107
+ "epoch": 11.0,
108
+ "eval_loss": 64.820556640625,
109
+ "eval_runtime": 0.4105,
110
+ "eval_samples_per_second": 4.872,
111
+ "eval_steps_per_second": 2.436,
112
+ "eval_wer": 1.0,
113
+ "step": 11
114
+ },
115
+ {
116
+ "epoch": 12.0,
117
+ "eval_loss": 63.82392120361328,
118
+ "eval_runtime": 0.4177,
119
+ "eval_samples_per_second": 4.788,
120
+ "eval_steps_per_second": 2.394,
121
+ "eval_wer": 1.0,
122
+ "step": 12
123
+ },
124
+ {
125
+ "epoch": 13.0,
126
+ "eval_loss": 62.648887634277344,
127
+ "eval_runtime": 0.41,
128
+ "eval_samples_per_second": 4.878,
129
+ "eval_steps_per_second": 2.439,
130
+ "eval_wer": 1.0,
131
+ "step": 13
132
+ },
133
+ {
134
+ "epoch": 14.0,
135
+ "eval_loss": 61.307098388671875,
136
+ "eval_runtime": 0.4232,
137
+ "eval_samples_per_second": 4.726,
138
+ "eval_steps_per_second": 2.363,
139
+ "eval_wer": 1.0,
140
+ "step": 14
141
+ },
142
+ {
143
+ "epoch": 15.0,
144
+ "eval_loss": 59.74273681640625,
145
+ "eval_runtime": 0.4053,
146
+ "eval_samples_per_second": 4.934,
147
+ "eval_steps_per_second": 2.467,
148
+ "eval_wer": 1.0,
149
+ "step": 15
150
+ },
151
+ {
152
+ "epoch": 16.0,
153
+ "eval_loss": 58.02557373046875,
154
+ "eval_runtime": 0.4173,
155
+ "eval_samples_per_second": 4.793,
156
+ "eval_steps_per_second": 2.397,
157
+ "eval_wer": 0.98,
158
+ "step": 16
159
+ },
160
+ {
161
+ "epoch": 17.0,
162
+ "eval_loss": 56.03272247314453,
163
+ "eval_runtime": 0.4178,
164
+ "eval_samples_per_second": 4.787,
165
+ "eval_steps_per_second": 2.394,
166
+ "eval_wer": 1.0,
167
+ "step": 17
168
+ },
169
+ {
170
+ "epoch": 18.0,
171
+ "eval_loss": 53.77244567871094,
172
+ "eval_runtime": 0.4131,
173
+ "eval_samples_per_second": 4.841,
174
+ "eval_steps_per_second": 2.421,
175
+ "eval_wer": 1.0,
176
+ "step": 18
177
+ },
178
+ {
179
+ "epoch": 19.0,
180
+ "eval_loss": 51.255592346191406,
181
+ "eval_runtime": 0.4212,
182
+ "eval_samples_per_second": 4.748,
183
+ "eval_steps_per_second": 2.374,
184
+ "eval_wer": 1.0,
185
+ "step": 19
186
+ },
187
+ {
188
+ "epoch": 20.0,
189
+ "learning_rate": 0.00016999999999999999,
190
+ "loss": 33.2554,
191
+ "step": 20
192
+ },
193
+ {
194
+ "epoch": 20.0,
195
+ "eval_loss": 48.49556350708008,
196
+ "eval_runtime": 0.4219,
197
+ "eval_samples_per_second": 4.741,
198
+ "eval_steps_per_second": 2.37,
199
+ "eval_wer": 1.0,
200
+ "step": 20
201
+ },
202
+ {
203
+ "epoch": 21.0,
204
+ "eval_loss": 45.40376281738281,
205
+ "eval_runtime": 0.4164,
206
+ "eval_samples_per_second": 4.803,
207
+ "eval_steps_per_second": 2.401,
208
+ "eval_wer": 1.0,
209
+ "step": 21
210
+ },
211
+ {
212
+ "epoch": 22.0,
213
+ "eval_loss": 41.99801254272461,
214
+ "eval_runtime": 0.4115,
215
+ "eval_samples_per_second": 4.86,
216
+ "eval_steps_per_second": 2.43,
217
+ "eval_wer": 1.0,
218
+ "step": 22
219
+ },
220
+ {
221
+ "epoch": 23.0,
222
+ "eval_loss": 41.99801254272461,
223
+ "eval_runtime": 0.4148,
224
+ "eval_samples_per_second": 4.822,
225
+ "eval_steps_per_second": 2.411,
226
+ "eval_wer": 1.0,
227
+ "step": 23
228
+ },
229
+ {
230
+ "epoch": 24.0,
231
+ "eval_loss": 38.22810745239258,
232
+ "eval_runtime": 0.4152,
233
+ "eval_samples_per_second": 4.817,
234
+ "eval_steps_per_second": 2.409,
235
+ "eval_wer": 1.0,
236
+ "step": 24
237
+ },
238
+ {
239
+ "epoch": 25.0,
240
+ "eval_loss": 34.15771484375,
241
+ "eval_runtime": 0.4229,
242
+ "eval_samples_per_second": 4.73,
243
+ "eval_steps_per_second": 2.365,
244
+ "eval_wer": 1.0,
245
+ "step": 25
246
+ },
247
+ {
248
+ "epoch": 26.0,
249
+ "eval_loss": 29.798538208007812,
250
+ "eval_runtime": 0.4129,
251
+ "eval_samples_per_second": 4.843,
252
+ "eval_steps_per_second": 2.422,
253
+ "eval_wer": 1.0,
254
+ "step": 26
255
+ },
256
+ {
257
+ "epoch": 27.0,
258
+ "eval_loss": 25.11461067199707,
259
+ "eval_runtime": 0.441,
260
+ "eval_samples_per_second": 4.535,
261
+ "eval_steps_per_second": 2.268,
262
+ "eval_wer": 1.0,
263
+ "step": 27
264
+ },
265
+ {
266
+ "epoch": 28.0,
267
+ "eval_loss": 20.228670120239258,
268
+ "eval_runtime": 0.4246,
269
+ "eval_samples_per_second": 4.71,
270
+ "eval_steps_per_second": 2.355,
271
+ "eval_wer": 1.0,
272
+ "step": 28
273
+ },
274
+ {
275
+ "epoch": 29.0,
276
+ "eval_loss": 15.340555191040039,
277
+ "eval_runtime": 0.4159,
278
+ "eval_samples_per_second": 4.809,
279
+ "eval_steps_per_second": 2.405,
280
+ "eval_wer": 1.0,
281
+ "step": 29
282
+ },
283
+ {
284
+ "epoch": 30.0,
285
+ "learning_rate": 0.00026,
286
+ "loss": 15.1206,
287
+ "step": 30
288
+ },
289
+ {
290
+ "epoch": 30.0,
291
+ "eval_loss": 10.769251823425293,
292
+ "eval_runtime": 0.4274,
293
+ "eval_samples_per_second": 4.68,
294
+ "eval_steps_per_second": 2.34,
295
+ "eval_wer": 1.0,
296
+ "step": 30
297
+ },
298
+ {
299
+ "epoch": 31.0,
300
+ "eval_loss": 6.8997721672058105,
301
+ "eval_runtime": 0.4239,
302
+ "eval_samples_per_second": 4.718,
303
+ "eval_steps_per_second": 2.359,
304
+ "eval_wer": 1.0,
305
+ "step": 31
306
+ },
307
+ {
308
+ "epoch": 32.0,
309
+ "eval_loss": 4.590747356414795,
310
+ "eval_runtime": 0.423,
311
+ "eval_samples_per_second": 4.728,
312
+ "eval_steps_per_second": 2.364,
313
+ "eval_wer": 1.0,
314
+ "step": 32
315
+ },
316
+ {
317
+ "epoch": 33.0,
318
+ "eval_loss": 3.359570026397705,
319
+ "eval_runtime": 0.4292,
320
+ "eval_samples_per_second": 4.66,
321
+ "eval_steps_per_second": 2.33,
322
+ "eval_wer": 1.0,
323
+ "step": 33
324
+ },
325
+ {
326
+ "epoch": 34.0,
327
+ "eval_loss": 2.7710704803466797,
328
+ "eval_runtime": 0.4158,
329
+ "eval_samples_per_second": 4.81,
330
+ "eval_steps_per_second": 2.405,
331
+ "eval_wer": 1.0,
332
+ "step": 34
333
+ },
334
+ {
335
+ "epoch": 35.0,
336
+ "eval_loss": 2.5961642265319824,
337
+ "eval_runtime": 0.4262,
338
+ "eval_samples_per_second": 4.692,
339
+ "eval_steps_per_second": 2.346,
340
+ "eval_wer": 1.0,
341
+ "step": 35
342
+ },
343
+ {
344
+ "epoch": 36.0,
345
+ "eval_loss": 2.900197982788086,
346
+ "eval_runtime": 0.4192,
347
+ "eval_samples_per_second": 4.771,
348
+ "eval_steps_per_second": 2.386,
349
+ "eval_wer": 1.0,
350
+ "step": 36
351
+ },
352
+ {
353
+ "epoch": 37.0,
354
+ "eval_loss": 3.0060811042785645,
355
+ "eval_runtime": 0.4182,
356
+ "eval_samples_per_second": 4.783,
357
+ "eval_steps_per_second": 2.391,
358
+ "eval_wer": 1.0,
359
+ "step": 37
360
+ },
361
+ {
362
+ "epoch": 38.0,
363
+ "eval_loss": 2.817464828491211,
364
+ "eval_runtime": 0.4241,
365
+ "eval_samples_per_second": 4.716,
366
+ "eval_steps_per_second": 2.358,
367
+ "eval_wer": 1.0,
368
+ "step": 38
369
+ },
370
+ {
371
+ "epoch": 39.0,
372
+ "eval_loss": 2.4512133598327637,
373
+ "eval_runtime": 0.4212,
374
+ "eval_samples_per_second": 4.748,
375
+ "eval_steps_per_second": 2.374,
376
+ "eval_wer": 1.0,
377
+ "step": 39
378
+ },
379
+ {
380
+ "epoch": 40.0,
381
+ "learning_rate": 0.00027428571428571427,
382
+ "loss": 2.4298,
383
+ "step": 40
384
+ },
385
+ {
386
+ "epoch": 40.0,
387
+ "eval_loss": 2.332979202270508,
388
+ "eval_runtime": 0.4231,
389
+ "eval_samples_per_second": 4.727,
390
+ "eval_steps_per_second": 2.363,
391
+ "eval_wer": 1.0,
392
+ "step": 40
393
+ },
394
+ {
395
+ "epoch": 41.0,
396
+ "eval_loss": 2.3766188621520996,
397
+ "eval_runtime": 0.4235,
398
+ "eval_samples_per_second": 4.723,
399
+ "eval_steps_per_second": 2.361,
400
+ "eval_wer": 1.0,
401
+ "step": 41
402
+ },
403
+ {
404
+ "epoch": 42.0,
405
+ "eval_loss": 2.5625646114349365,
406
+ "eval_runtime": 0.4219,
407
+ "eval_samples_per_second": 4.741,
408
+ "eval_steps_per_second": 2.37,
409
+ "eval_wer": 1.0,
410
+ "step": 42
411
+ },
412
+ {
413
+ "epoch": 43.0,
414
+ "eval_loss": 2.9632387161254883,
415
+ "eval_runtime": 0.4235,
416
+ "eval_samples_per_second": 4.722,
417
+ "eval_steps_per_second": 2.361,
418
+ "eval_wer": 1.0,
419
+ "step": 43
420
+ },
421
+ {
422
+ "epoch": 44.0,
423
+ "eval_loss": 3.279555082321167,
424
+ "eval_runtime": 0.4175,
425
+ "eval_samples_per_second": 4.79,
426
+ "eval_steps_per_second": 2.395,
427
+ "eval_wer": 1.0,
428
+ "step": 44
429
+ },
430
+ {
431
+ "epoch": 45.0,
432
+ "eval_loss": 3.4015283584594727,
433
+ "eval_runtime": 0.4178,
434
+ "eval_samples_per_second": 4.787,
435
+ "eval_steps_per_second": 2.393,
436
+ "eval_wer": 1.0,
437
+ "step": 45
438
+ },
439
+ {
440
+ "epoch": 46.0,
441
+ "eval_loss": 3.2808432579040527,
442
+ "eval_runtime": 0.4209,
443
+ "eval_samples_per_second": 4.752,
444
+ "eval_steps_per_second": 2.376,
445
+ "eval_wer": 1.0,
446
+ "step": 46
447
+ },
448
+ {
449
+ "epoch": 47.0,
450
+ "eval_loss": 3.2373273372650146,
451
+ "eval_runtime": 0.4225,
452
+ "eval_samples_per_second": 4.734,
453
+ "eval_steps_per_second": 2.367,
454
+ "eval_wer": 1.0,
455
+ "step": 47
456
+ },
457
+ {
458
+ "epoch": 48.0,
459
+ "eval_loss": 3.2461540699005127,
460
+ "eval_runtime": 0.4121,
461
+ "eval_samples_per_second": 4.853,
462
+ "eval_steps_per_second": 2.427,
463
+ "eval_wer": 1.0,
464
+ "step": 48
465
+ },
466
+ {
467
+ "epoch": 49.0,
468
+ "eval_loss": 3.6168460845947266,
469
+ "eval_runtime": 0.436,
470
+ "eval_samples_per_second": 4.587,
471
+ "eval_steps_per_second": 2.293,
472
+ "eval_wer": 1.0,
473
+ "step": 49
474
+ },
475
+ {
476
+ "epoch": 50.0,
477
+ "learning_rate": 0.00023142857142857142,
478
+ "loss": 1.6143,
479
+ "step": 50
480
+ },
481
+ {
482
+ "epoch": 50.0,
483
+ "eval_loss": 3.6624670028686523,
484
+ "eval_runtime": 0.4363,
485
+ "eval_samples_per_second": 4.584,
486
+ "eval_steps_per_second": 2.292,
487
+ "eval_wer": 1.0,
488
+ "step": 50
489
+ },
490
+ {
491
+ "epoch": 51.0,
492
+ "eval_loss": 3.759349822998047,
493
+ "eval_runtime": 0.4284,
494
+ "eval_samples_per_second": 4.668,
495
+ "eval_steps_per_second": 2.334,
496
+ "eval_wer": 1.0,
497
+ "step": 51
498
+ },
499
+ {
500
+ "epoch": 52.0,
501
+ "eval_loss": 3.9326891899108887,
502
+ "eval_runtime": 0.4199,
503
+ "eval_samples_per_second": 4.763,
504
+ "eval_steps_per_second": 2.382,
505
+ "eval_wer": 1.0,
506
+ "step": 52
507
+ },
508
+ {
509
+ "epoch": 53.0,
510
+ "eval_loss": 3.718548536300659,
511
+ "eval_runtime": 0.4129,
512
+ "eval_samples_per_second": 4.843,
513
+ "eval_steps_per_second": 2.422,
514
+ "eval_wer": 1.0,
515
+ "step": 53
516
+ },
517
+ {
518
+ "epoch": 54.0,
519
+ "eval_loss": 3.90997052192688,
520
+ "eval_runtime": 0.4206,
521
+ "eval_samples_per_second": 4.755,
522
+ "eval_steps_per_second": 2.377,
523
+ "eval_wer": 1.0,
524
+ "step": 54
525
+ },
526
+ {
527
+ "epoch": 55.0,
528
+ "eval_loss": 4.312277317047119,
529
+ "eval_runtime": 0.4164,
530
+ "eval_samples_per_second": 4.803,
531
+ "eval_steps_per_second": 2.401,
532
+ "eval_wer": 1.0,
533
+ "step": 55
534
+ },
535
+ {
536
+ "epoch": 56.0,
537
+ "eval_loss": 4.2904229164123535,
538
+ "eval_runtime": 0.4192,
539
+ "eval_samples_per_second": 4.771,
540
+ "eval_steps_per_second": 2.386,
541
+ "eval_wer": 1.0,
542
+ "step": 56
543
+ },
544
+ {
545
+ "epoch": 57.0,
546
+ "eval_loss": 3.951892852783203,
547
+ "eval_runtime": 0.4198,
548
+ "eval_samples_per_second": 4.765,
549
+ "eval_steps_per_second": 2.382,
550
+ "eval_wer": 1.0,
551
+ "step": 57
552
+ },
553
+ {
554
+ "epoch": 58.0,
555
+ "eval_loss": 3.45182728767395,
556
+ "eval_runtime": 0.4295,
557
+ "eval_samples_per_second": 4.657,
558
+ "eval_steps_per_second": 2.328,
559
+ "eval_wer": 1.0,
560
+ "step": 58
561
+ },
562
+ {
563
+ "epoch": 59.0,
564
+ "eval_loss": 3.0196707248687744,
565
+ "eval_runtime": 0.4211,
566
+ "eval_samples_per_second": 4.75,
567
+ "eval_steps_per_second": 2.375,
568
+ "eval_wer": 1.0,
569
+ "step": 59
570
+ },
571
+ {
572
+ "epoch": 60.0,
573
+ "learning_rate": 0.00018857142857142854,
574
+ "loss": 1.4054,
575
+ "step": 60
576
+ },
577
+ {
578
+ "epoch": 60.0,
579
+ "eval_loss": 2.886315107345581,
580
+ "eval_runtime": 0.4413,
581
+ "eval_samples_per_second": 4.533,
582
+ "eval_steps_per_second": 2.266,
583
+ "eval_wer": 1.0,
584
+ "step": 60
585
+ },
586
+ {
587
+ "epoch": 61.0,
588
+ "eval_loss": 2.97536563873291,
589
+ "eval_runtime": 0.4417,
590
+ "eval_samples_per_second": 4.528,
591
+ "eval_steps_per_second": 2.264,
592
+ "eval_wer": 1.0,
593
+ "step": 61
594
+ },
595
+ {
596
+ "epoch": 62.0,
597
+ "eval_loss": 3.299807548522949,
598
+ "eval_runtime": 0.426,
599
+ "eval_samples_per_second": 4.695,
600
+ "eval_steps_per_second": 2.347,
601
+ "eval_wer": 1.0,
602
+ "step": 62
603
+ },
604
+ {
605
+ "epoch": 63.0,
606
+ "eval_loss": 3.871462821960449,
607
+ "eval_runtime": 0.4236,
608
+ "eval_samples_per_second": 4.721,
609
+ "eval_steps_per_second": 2.36,
610
+ "eval_wer": 1.0,
611
+ "step": 63
612
+ },
613
+ {
614
+ "epoch": 64.0,
615
+ "eval_loss": 4.189750671386719,
616
+ "eval_runtime": 0.4223,
617
+ "eval_samples_per_second": 4.736,
618
+ "eval_steps_per_second": 2.368,
619
+ "eval_wer": 1.0,
620
+ "step": 64
621
+ },
622
+ {
623
+ "epoch": 65.0,
624
+ "eval_loss": 4.18125057220459,
625
+ "eval_runtime": 0.4138,
626
+ "eval_samples_per_second": 4.833,
627
+ "eval_steps_per_second": 2.417,
628
+ "eval_wer": 1.0,
629
+ "step": 65
630
+ },
631
+ {
632
+ "epoch": 66.0,
633
+ "eval_loss": 3.9024930000305176,
634
+ "eval_runtime": 0.4391,
635
+ "eval_samples_per_second": 4.555,
636
+ "eval_steps_per_second": 2.277,
637
+ "eval_wer": 1.0,
638
+ "step": 66
639
+ },
640
+ {
641
+ "epoch": 67.0,
642
+ "eval_loss": 3.4319417476654053,
643
+ "eval_runtime": 0.4273,
644
+ "eval_samples_per_second": 4.681,
645
+ "eval_steps_per_second": 2.34,
646
+ "eval_wer": 1.0,
647
+ "step": 67
648
+ },
649
+ {
650
+ "epoch": 68.0,
651
+ "eval_loss": 3.275496482849121,
652
+ "eval_runtime": 0.4173,
653
+ "eval_samples_per_second": 4.793,
654
+ "eval_steps_per_second": 2.396,
655
+ "eval_wer": 1.0,
656
+ "step": 68
657
+ },
658
+ {
659
+ "epoch": 69.0,
660
+ "eval_loss": 3.334933042526245,
661
+ "eval_runtime": 0.4221,
662
+ "eval_samples_per_second": 4.738,
663
+ "eval_steps_per_second": 2.369,
664
+ "eval_wer": 1.0,
665
+ "step": 69
666
+ },
667
+ {
668
+ "epoch": 70.0,
669
+ "learning_rate": 0.0001457142857142857,
670
+ "loss": 1.3121,
671
+ "step": 70
672
+ },
673
+ {
674
+ "epoch": 70.0,
675
+ "eval_loss": 3.548488140106201,
676
+ "eval_runtime": 0.4291,
677
+ "eval_samples_per_second": 4.661,
678
+ "eval_steps_per_second": 2.331,
679
+ "eval_wer": 1.0,
680
+ "step": 70
681
+ },
682
+ {
683
+ "epoch": 71.0,
684
+ "eval_loss": 3.9019012451171875,
685
+ "eval_runtime": 0.4171,
686
+ "eval_samples_per_second": 4.795,
687
+ "eval_steps_per_second": 2.398,
688
+ "eval_wer": 1.0,
689
+ "step": 71
690
+ },
691
+ {
692
+ "epoch": 72.0,
693
+ "eval_loss": 4.081923961639404,
694
+ "eval_runtime": 0.4176,
695
+ "eval_samples_per_second": 4.789,
696
+ "eval_steps_per_second": 2.395,
697
+ "eval_wer": 1.0,
698
+ "step": 72
699
+ },
700
+ {
701
+ "epoch": 73.0,
702
+ "eval_loss": 3.9955055713653564,
703
+ "eval_runtime": 0.4176,
704
+ "eval_samples_per_second": 4.789,
705
+ "eval_steps_per_second": 2.395,
706
+ "eval_wer": 1.0,
707
+ "step": 73
708
+ },
709
+ {
710
+ "epoch": 74.0,
711
+ "eval_loss": 3.70878529548645,
712
+ "eval_runtime": 0.423,
713
+ "eval_samples_per_second": 4.729,
714
+ "eval_steps_per_second": 2.364,
715
+ "eval_wer": 1.0,
716
+ "step": 74
717
+ },
718
+ {
719
+ "epoch": 75.0,
720
+ "eval_loss": 3.295703887939453,
721
+ "eval_runtime": 0.4128,
722
+ "eval_samples_per_second": 4.846,
723
+ "eval_steps_per_second": 2.423,
724
+ "eval_wer": 1.0,
725
+ "step": 75
726
+ },
727
+ {
728
+ "epoch": 76.0,
729
+ "eval_loss": 3.114126682281494,
730
+ "eval_runtime": 0.4146,
731
+ "eval_samples_per_second": 4.824,
732
+ "eval_steps_per_second": 2.412,
733
+ "eval_wer": 1.0,
734
+ "step": 76
735
+ },
736
+ {
737
+ "epoch": 77.0,
738
+ "eval_loss": 3.085181713104248,
739
+ "eval_runtime": 0.4266,
740
+ "eval_samples_per_second": 4.688,
741
+ "eval_steps_per_second": 2.344,
742
+ "eval_wer": 1.0,
743
+ "step": 77
744
+ },
745
+ {
746
+ "epoch": 78.0,
747
+ "eval_loss": 3.187084436416626,
748
+ "eval_runtime": 0.4233,
749
+ "eval_samples_per_second": 4.725,
750
+ "eval_steps_per_second": 2.362,
751
+ "eval_wer": 1.0,
752
+ "step": 78
753
+ },
754
+ {
755
+ "epoch": 79.0,
756
+ "eval_loss": 3.4126548767089844,
757
+ "eval_runtime": 0.406,
758
+ "eval_samples_per_second": 4.926,
759
+ "eval_steps_per_second": 2.463,
760
+ "eval_wer": 1.0,
761
+ "step": 79
762
+ },
763
+ {
764
+ "epoch": 80.0,
765
+ "learning_rate": 0.00010285714285714284,
766
+ "loss": 1.2576,
767
+ "step": 80
768
+ },
769
+ {
770
+ "epoch": 80.0,
771
+ "eval_loss": 3.6913061141967773,
772
+ "eval_runtime": 0.4088,
773
+ "eval_samples_per_second": 4.892,
774
+ "eval_steps_per_second": 2.446,
775
+ "eval_wer": 1.0,
776
+ "step": 80
777
+ },
778
+ {
779
+ "epoch": 81.0,
780
+ "eval_loss": 3.8285651206970215,
781
+ "eval_runtime": 0.4079,
782
+ "eval_samples_per_second": 4.904,
783
+ "eval_steps_per_second": 2.452,
784
+ "eval_wer": 1.0,
785
+ "step": 81
786
+ },
787
+ {
788
+ "epoch": 82.0,
789
+ "eval_loss": 3.815654993057251,
790
+ "eval_runtime": 0.4087,
791
+ "eval_samples_per_second": 4.894,
792
+ "eval_steps_per_second": 2.447,
793
+ "eval_wer": 1.0,
794
+ "step": 82
795
+ },
796
+ {
797
+ "epoch": 83.0,
798
+ "eval_loss": 3.6813836097717285,
799
+ "eval_runtime": 0.4035,
800
+ "eval_samples_per_second": 4.957,
801
+ "eval_steps_per_second": 2.478,
802
+ "eval_wer": 1.0,
803
+ "step": 83
804
+ },
805
+ {
806
+ "epoch": 84.0,
807
+ "eval_loss": 3.44958758354187,
808
+ "eval_runtime": 0.4079,
809
+ "eval_samples_per_second": 4.903,
810
+ "eval_steps_per_second": 2.452,
811
+ "eval_wer": 1.0,
812
+ "step": 84
813
+ },
814
+ {
815
+ "epoch": 85.0,
816
+ "eval_loss": 3.284428119659424,
817
+ "eval_runtime": 0.4066,
818
+ "eval_samples_per_second": 4.919,
819
+ "eval_steps_per_second": 2.46,
820
+ "eval_wer": 1.0,
821
+ "step": 85
822
+ },
823
+ {
824
+ "epoch": 86.0,
825
+ "eval_loss": 3.2253918647766113,
826
+ "eval_runtime": 0.405,
827
+ "eval_samples_per_second": 4.938,
828
+ "eval_steps_per_second": 2.469,
829
+ "eval_wer": 1.0,
830
+ "step": 86
831
+ },
832
+ {
833
+ "epoch": 87.0,
834
+ "eval_loss": 3.268294334411621,
835
+ "eval_runtime": 0.4089,
836
+ "eval_samples_per_second": 4.892,
837
+ "eval_steps_per_second": 2.446,
838
+ "eval_wer": 1.0,
839
+ "step": 87
840
+ },
841
+ {
842
+ "epoch": 88.0,
843
+ "eval_loss": 3.3791449069976807,
844
+ "eval_runtime": 0.4084,
845
+ "eval_samples_per_second": 4.897,
846
+ "eval_steps_per_second": 2.449,
847
+ "eval_wer": 1.0,
848
+ "step": 88
849
+ },
850
+ {
851
+ "epoch": 89.0,
852
+ "eval_loss": 3.550123453140259,
853
+ "eval_runtime": 0.4056,
854
+ "eval_samples_per_second": 4.931,
855
+ "eval_steps_per_second": 2.466,
856
+ "eval_wer": 1.0,
857
+ "step": 89
858
+ },
859
+ {
860
+ "epoch": 90.0,
861
+ "learning_rate": 5.9999999999999995e-05,
862
+ "loss": 1.2373,
863
+ "step": 90
864
+ },
865
+ {
866
+ "epoch": 90.0,
867
+ "eval_loss": 3.662168025970459,
868
+ "eval_runtime": 0.4176,
869
+ "eval_samples_per_second": 4.789,
870
+ "eval_steps_per_second": 2.395,
871
+ "eval_wer": 1.0,
872
+ "step": 90
873
+ },
874
+ {
875
+ "epoch": 91.0,
876
+ "eval_loss": 3.7207443714141846,
877
+ "eval_runtime": 0.4173,
878
+ "eval_samples_per_second": 4.793,
879
+ "eval_steps_per_second": 2.396,
880
+ "eval_wer": 1.0,
881
+ "step": 91
882
+ },
883
+ {
884
+ "epoch": 92.0,
885
+ "eval_loss": 3.6961331367492676,
886
+ "eval_runtime": 0.4129,
887
+ "eval_samples_per_second": 4.844,
888
+ "eval_steps_per_second": 2.422,
889
+ "eval_wer": 1.0,
890
+ "step": 92
891
+ },
892
+ {
893
+ "epoch": 93.0,
894
+ "eval_loss": 3.6098761558532715,
895
+ "eval_runtime": 0.4061,
896
+ "eval_samples_per_second": 4.925,
897
+ "eval_steps_per_second": 2.463,
898
+ "eval_wer": 1.0,
899
+ "step": 93
900
+ },
901
+ {
902
+ "epoch": 94.0,
903
+ "eval_loss": 3.533604860305786,
904
+ "eval_runtime": 0.4224,
905
+ "eval_samples_per_second": 4.735,
906
+ "eval_steps_per_second": 2.368,
907
+ "eval_wer": 1.0,
908
+ "step": 94
909
+ },
910
+ {
911
+ "epoch": 95.0,
912
+ "eval_loss": 3.4341912269592285,
913
+ "eval_runtime": 0.4255,
914
+ "eval_samples_per_second": 4.701,
915
+ "eval_steps_per_second": 2.35,
916
+ "eval_wer": 1.0,
917
+ "step": 95
918
+ },
919
+ {
920
+ "epoch": 96.0,
921
+ "eval_loss": 3.316969871520996,
922
+ "eval_runtime": 0.4089,
923
+ "eval_samples_per_second": 4.891,
924
+ "eval_steps_per_second": 2.445,
925
+ "eval_wer": 1.0,
926
+ "step": 96
927
+ },
928
+ {
929
+ "epoch": 97.0,
930
+ "eval_loss": 3.2624034881591797,
931
+ "eval_runtime": 0.4046,
932
+ "eval_samples_per_second": 4.943,
933
+ "eval_steps_per_second": 2.471,
934
+ "eval_wer": 1.0,
935
+ "step": 97
936
+ },
937
+ {
938
+ "epoch": 98.0,
939
+ "eval_loss": 3.2437386512756348,
940
+ "eval_runtime": 0.4006,
941
+ "eval_samples_per_second": 4.993,
942
+ "eval_steps_per_second": 2.497,
943
+ "eval_wer": 1.0,
944
+ "step": 98
945
+ },
946
+ {
947
+ "epoch": 99.0,
948
+ "eval_loss": 3.259126663208008,
949
+ "eval_runtime": 0.4066,
950
+ "eval_samples_per_second": 4.919,
951
+ "eval_steps_per_second": 2.46,
952
+ "eval_wer": 1.0,
953
+ "step": 99
954
+ },
955
+ {
956
+ "epoch": 100.0,
957
+ "learning_rate": 1.7142857142857142e-05,
958
+ "loss": 1.1952,
959
+ "step": 100
960
+ },
961
+ {
962
+ "epoch": 100.0,
963
+ "eval_loss": 3.292677402496338,
964
+ "eval_runtime": 0.4167,
965
+ "eval_samples_per_second": 4.799,
966
+ "eval_steps_per_second": 2.4,
967
+ "eval_wer": 1.0,
968
+ "step": 100
969
+ },
970
+ {
971
+ "epoch": 100.0,
972
+ "step": 100,
973
+ "total_flos": 2.094330789153408e+17,
974
+ "train_loss": 9.722876396179199,
975
+ "train_runtime": 860.4357,
976
+ "train_samples_per_second": 1.86,
977
+ "train_steps_per_second": 0.116
978
+ }
979
+ ],
980
+ "max_steps": 100,
981
+ "num_train_epochs": 100,
982
+ "total_flos": 2.094330789153408e+17,
983
+ "trial_name": null,
984
+ "trial_params": null
985
+ }