fanaf91318 commited on
Commit
e580fc6
1 Parent(s): d589561

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +392 -0
trainer_state.json ADDED
@@ -0,0 +1,392 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 73.80372709831596,
3
+ "best_model_checkpoint": "./whisper-distil-v3/checkpoint-1000",
4
+ "epoch": 0.35561877667140823,
5
+ "eval_steps": 1000,
6
+ "global_step": 1000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.007112375533428165,
13
+ "grad_norm": 5.77730655670166,
14
+ "learning_rate": 3.4000000000000005e-06,
15
+ "loss": 8.7115,
16
+ "step": 20
17
+ },
18
+ {
19
+ "epoch": 0.01422475106685633,
20
+ "grad_norm": 5.685363292694092,
21
+ "learning_rate": 7.4e-06,
22
+ "loss": 8.4512,
23
+ "step": 40
24
+ },
25
+ {
26
+ "epoch": 0.021337126600284494,
27
+ "grad_norm": 6.336855888366699,
28
+ "learning_rate": 1.1400000000000001e-05,
29
+ "loss": 8.118,
30
+ "step": 60
31
+ },
32
+ {
33
+ "epoch": 0.02844950213371266,
34
+ "grad_norm": 5.820839881896973,
35
+ "learning_rate": 1.54e-05,
36
+ "loss": 7.5291,
37
+ "step": 80
38
+ },
39
+ {
40
+ "epoch": 0.03556187766714083,
41
+ "grad_norm": 6.599063873291016,
42
+ "learning_rate": 1.94e-05,
43
+ "loss": 6.4103,
44
+ "step": 100
45
+ },
46
+ {
47
+ "epoch": 0.04267425320056899,
48
+ "grad_norm": 3.500046968460083,
49
+ "learning_rate": 2.3400000000000003e-05,
50
+ "loss": 5.3304,
51
+ "step": 120
52
+ },
53
+ {
54
+ "epoch": 0.049786628733997154,
55
+ "grad_norm": 3.011868476867676,
56
+ "learning_rate": 2.7400000000000002e-05,
57
+ "loss": 4.3763,
58
+ "step": 140
59
+ },
60
+ {
61
+ "epoch": 0.05689900426742532,
62
+ "grad_norm": 2.6885221004486084,
63
+ "learning_rate": 3.1400000000000004e-05,
64
+ "loss": 3.7984,
65
+ "step": 160
66
+ },
67
+ {
68
+ "epoch": 0.06401137980085349,
69
+ "grad_norm": 2.7655692100524902,
70
+ "learning_rate": 3.54e-05,
71
+ "loss": 3.1392,
72
+ "step": 180
73
+ },
74
+ {
75
+ "epoch": 0.07112375533428165,
76
+ "grad_norm": 2.893932580947876,
77
+ "learning_rate": 3.94e-05,
78
+ "loss": 2.9927,
79
+ "step": 200
80
+ },
81
+ {
82
+ "epoch": 0.07823613086770982,
83
+ "grad_norm": 2.6793711185455322,
84
+ "learning_rate": 4.3400000000000005e-05,
85
+ "loss": 2.719,
86
+ "step": 220
87
+ },
88
+ {
89
+ "epoch": 0.08534850640113797,
90
+ "grad_norm": 2.8600358963012695,
91
+ "learning_rate": 4.74e-05,
92
+ "loss": 2.6649,
93
+ "step": 240
94
+ },
95
+ {
96
+ "epoch": 0.09246088193456614,
97
+ "grad_norm": 3.3596479892730713,
98
+ "learning_rate": 5.14e-05,
99
+ "loss": 2.6195,
100
+ "step": 260
101
+ },
102
+ {
103
+ "epoch": 0.09957325746799431,
104
+ "grad_norm": 3.0968446731567383,
105
+ "learning_rate": 5.5400000000000005e-05,
106
+ "loss": 2.2805,
107
+ "step": 280
108
+ },
109
+ {
110
+ "epoch": 0.10668563300142248,
111
+ "grad_norm": 4.1708903312683105,
112
+ "learning_rate": 5.94e-05,
113
+ "loss": 2.1195,
114
+ "step": 300
115
+ },
116
+ {
117
+ "epoch": 0.11379800853485064,
118
+ "grad_norm": 3.3060481548309326,
119
+ "learning_rate": 6.340000000000001e-05,
120
+ "loss": 2.2916,
121
+ "step": 320
122
+ },
123
+ {
124
+ "epoch": 0.12091038406827881,
125
+ "grad_norm": 3.289092540740967,
126
+ "learning_rate": 6.740000000000001e-05,
127
+ "loss": 2.1663,
128
+ "step": 340
129
+ },
130
+ {
131
+ "epoch": 0.12802275960170698,
132
+ "grad_norm": 3.3083698749542236,
133
+ "learning_rate": 7.14e-05,
134
+ "loss": 2.1373,
135
+ "step": 360
136
+ },
137
+ {
138
+ "epoch": 0.13513513513513514,
139
+ "grad_norm": 3.4208426475524902,
140
+ "learning_rate": 7.54e-05,
141
+ "loss": 2.0894,
142
+ "step": 380
143
+ },
144
+ {
145
+ "epoch": 0.1422475106685633,
146
+ "grad_norm": 3.856208562850952,
147
+ "learning_rate": 7.94e-05,
148
+ "loss": 1.9931,
149
+ "step": 400
150
+ },
151
+ {
152
+ "epoch": 0.14935988620199148,
153
+ "grad_norm": 3.556122303009033,
154
+ "learning_rate": 8.34e-05,
155
+ "loss": 1.9146,
156
+ "step": 420
157
+ },
158
+ {
159
+ "epoch": 0.15647226173541964,
160
+ "grad_norm": 3.152322769165039,
161
+ "learning_rate": 8.740000000000001e-05,
162
+ "loss": 1.9123,
163
+ "step": 440
164
+ },
165
+ {
166
+ "epoch": 0.16358463726884778,
167
+ "grad_norm": 3.6083829402923584,
168
+ "learning_rate": 9.140000000000001e-05,
169
+ "loss": 1.8787,
170
+ "step": 460
171
+ },
172
+ {
173
+ "epoch": 0.17069701280227595,
174
+ "grad_norm": 3.5843188762664795,
175
+ "learning_rate": 9.54e-05,
176
+ "loss": 1.8949,
177
+ "step": 480
178
+ },
179
+ {
180
+ "epoch": 0.17780938833570412,
181
+ "grad_norm": 4.271010398864746,
182
+ "learning_rate": 9.94e-05,
183
+ "loss": 1.8362,
184
+ "step": 500
185
+ },
186
+ {
187
+ "epoch": 0.18492176386913228,
188
+ "grad_norm": 3.9229135513305664,
189
+ "learning_rate": 0.0001,
190
+ "loss": 1.81,
191
+ "step": 520
192
+ },
193
+ {
194
+ "epoch": 0.19203413940256045,
195
+ "grad_norm": 3.9755284786224365,
196
+ "learning_rate": 0.0001,
197
+ "loss": 1.7339,
198
+ "step": 540
199
+ },
200
+ {
201
+ "epoch": 0.19914651493598862,
202
+ "grad_norm": 4.1353912353515625,
203
+ "learning_rate": 0.0001,
204
+ "loss": 1.5915,
205
+ "step": 560
206
+ },
207
+ {
208
+ "epoch": 0.20625889046941678,
209
+ "grad_norm": 3.9695911407470703,
210
+ "learning_rate": 0.0001,
211
+ "loss": 1.6895,
212
+ "step": 580
213
+ },
214
+ {
215
+ "epoch": 0.21337126600284495,
216
+ "grad_norm": 5.034478664398193,
217
+ "learning_rate": 0.0001,
218
+ "loss": 1.6259,
219
+ "step": 600
220
+ },
221
+ {
222
+ "epoch": 0.22048364153627312,
223
+ "grad_norm": 3.5285422801971436,
224
+ "learning_rate": 0.0001,
225
+ "loss": 1.7478,
226
+ "step": 620
227
+ },
228
+ {
229
+ "epoch": 0.22759601706970128,
230
+ "grad_norm": 2.889526605606079,
231
+ "learning_rate": 0.0001,
232
+ "loss": 1.6589,
233
+ "step": 640
234
+ },
235
+ {
236
+ "epoch": 0.23470839260312945,
237
+ "grad_norm": 3.114544153213501,
238
+ "learning_rate": 0.0001,
239
+ "loss": 1.5978,
240
+ "step": 660
241
+ },
242
+ {
243
+ "epoch": 0.24182076813655762,
244
+ "grad_norm": 3.4191863536834717,
245
+ "learning_rate": 0.0001,
246
+ "loss": 1.7244,
247
+ "step": 680
248
+ },
249
+ {
250
+ "epoch": 0.24893314366998578,
251
+ "grad_norm": 3.815669298171997,
252
+ "learning_rate": 0.0001,
253
+ "loss": 1.6637,
254
+ "step": 700
255
+ },
256
+ {
257
+ "epoch": 0.25604551920341395,
258
+ "grad_norm": 3.644179105758667,
259
+ "learning_rate": 0.0001,
260
+ "loss": 1.5496,
261
+ "step": 720
262
+ },
263
+ {
264
+ "epoch": 0.2631578947368421,
265
+ "grad_norm": 2.65865159034729,
266
+ "learning_rate": 0.0001,
267
+ "loss": 1.5384,
268
+ "step": 740
269
+ },
270
+ {
271
+ "epoch": 0.2702702702702703,
272
+ "grad_norm": 3.979856491088867,
273
+ "learning_rate": 0.0001,
274
+ "loss": 1.5651,
275
+ "step": 760
276
+ },
277
+ {
278
+ "epoch": 0.2773826458036984,
279
+ "grad_norm": 4.229691028594971,
280
+ "learning_rate": 0.0001,
281
+ "loss": 1.5942,
282
+ "step": 780
283
+ },
284
+ {
285
+ "epoch": 0.2844950213371266,
286
+ "grad_norm": 3.7266907691955566,
287
+ "learning_rate": 0.0001,
288
+ "loss": 1.5734,
289
+ "step": 800
290
+ },
291
+ {
292
+ "epoch": 0.29160739687055476,
293
+ "grad_norm": 2.8548645973205566,
294
+ "learning_rate": 0.0001,
295
+ "loss": 1.4683,
296
+ "step": 820
297
+ },
298
+ {
299
+ "epoch": 0.29871977240398295,
300
+ "grad_norm": 3.398973226547241,
301
+ "learning_rate": 0.0001,
302
+ "loss": 1.5065,
303
+ "step": 840
304
+ },
305
+ {
306
+ "epoch": 0.3058321479374111,
307
+ "grad_norm": 3.1035850048065186,
308
+ "learning_rate": 0.0001,
309
+ "loss": 1.5188,
310
+ "step": 860
311
+ },
312
+ {
313
+ "epoch": 0.3129445234708393,
314
+ "grad_norm": 3.0334548950195312,
315
+ "learning_rate": 0.0001,
316
+ "loss": 1.5774,
317
+ "step": 880
318
+ },
319
+ {
320
+ "epoch": 0.3200568990042674,
321
+ "grad_norm": 3.0562374591827393,
322
+ "learning_rate": 0.0001,
323
+ "loss": 1.4533,
324
+ "step": 900
325
+ },
326
+ {
327
+ "epoch": 0.32716927453769556,
328
+ "grad_norm": 2.881594657897949,
329
+ "learning_rate": 0.0001,
330
+ "loss": 1.4867,
331
+ "step": 920
332
+ },
333
+ {
334
+ "epoch": 0.33428165007112376,
335
+ "grad_norm": 3.8474974632263184,
336
+ "learning_rate": 0.0001,
337
+ "loss": 1.4431,
338
+ "step": 940
339
+ },
340
+ {
341
+ "epoch": 0.3413940256045519,
342
+ "grad_norm": 3.3145081996917725,
343
+ "learning_rate": 0.0001,
344
+ "loss": 1.588,
345
+ "step": 960
346
+ },
347
+ {
348
+ "epoch": 0.3485064011379801,
349
+ "grad_norm": 3.1782026290893555,
350
+ "learning_rate": 0.0001,
351
+ "loss": 1.396,
352
+ "step": 980
353
+ },
354
+ {
355
+ "epoch": 0.35561877667140823,
356
+ "grad_norm": 2.8277013301849365,
357
+ "learning_rate": 0.0001,
358
+ "loss": 1.3896,
359
+ "step": 1000
360
+ },
361
+ {
362
+ "epoch": 0.35561877667140823,
363
+ "eval_loss": 1.4457485675811768,
364
+ "eval_runtime": 4741.188,
365
+ "eval_samples_per_second": 2.109,
366
+ "eval_steps_per_second": 0.066,
367
+ "eval_wer": 73.80372709831596,
368
+ "step": 1000
369
+ }
370
+ ],
371
+ "logging_steps": 20,
372
+ "max_steps": 8436,
373
+ "num_input_tokens_seen": 0,
374
+ "num_train_epochs": 3,
375
+ "save_steps": 1000,
376
+ "stateful_callbacks": {
377
+ "TrainerControl": {
378
+ "args": {
379
+ "should_epoch_stop": false,
380
+ "should_evaluate": false,
381
+ "should_log": false,
382
+ "should_save": true,
383
+ "should_training_stop": false
384
+ },
385
+ "attributes": {}
386
+ }
387
+ },
388
+ "total_flos": 0.0,
389
+ "train_batch_size": 8,
390
+ "trial_name": null,
391
+ "trial_params": null
392
+ }