rishabhjain16 commited on
Commit
342fea5
1 Parent(s): a27459f

End of training

Browse files
README.md CHANGED
@@ -2,20 +2,33 @@
2
  license: apache-2.0
3
  base_model: openai/whisper-large-v2
4
  tags:
 
5
  - generated_from_trainer
 
 
6
  metrics:
7
  - wer
8
  model-index:
9
- - name: openai/whisper-large-v2
10
- results: []
 
 
 
 
 
 
 
 
 
 
11
  ---
12
 
13
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
  should probably proofread and complete it, then remove this comment. -->
15
 
16
- # openai/whisper-large-v2
17
 
18
- This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.7918
21
  - Wer: 34.6232
 
2
  license: apache-2.0
3
  base_model: openai/whisper-large-v2
4
  tags:
5
+ - whisper-event
6
  - generated_from_trainer
7
+ datasets:
8
+ - rishabhjain16/owr_cv_albanian
9
  metrics:
10
  - wer
11
  model-index:
12
+ - name: Whisper large V2 to CV Albanian
13
+ results:
14
+ - task:
15
+ name: Automatic Speech Recognition
16
+ type: automatic-speech-recognition
17
+ dataset:
18
+ name: rishabhjain16/owr_cv_albanian default
19
+ type: rishabhjain16/owr_cv_albanian
20
+ metrics:
21
+ - name: Wer
22
+ type: wer
23
+ value: 34.623217922606926
24
  ---
25
 
26
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
27
  should probably proofread and complete it, then remove this comment. -->
28
 
29
+ # Whisper large V2 to CV Albanian
30
 
31
+ This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the rishabhjain16/owr_cv_albanian default dataset.
32
  It achieves the following results on the evaluation set:
33
  - Loss: 0.7918
34
  - Wer: 34.6232
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 92.01,
3
+ "eval_loss": 0.7918470501899719,
4
+ "eval_runtime": 102.6317,
5
+ "eval_samples_per_second": 3.742,
6
+ "eval_steps_per_second": 0.234,
7
+ "eval_wer": 34.623217922606926,
8
+ "train_loss": 0.04092751889359206,
9
+ "train_runtime": 27870.4476,
10
+ "train_samples_per_second": 2.87,
11
+ "train_steps_per_second": 0.179
12
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 92.01,
3
+ "eval_loss": 0.7918470501899719,
4
+ "eval_runtime": 102.6317,
5
+ "eval_samples_per_second": 3.742,
6
+ "eval_steps_per_second": 0.234,
7
+ "eval_wer": 34.623217922606926
8
+ }
runs/Mar01_15-02-56_5d19937efe64/events.out.tfevents.1709333773.5d19937efe64.1529.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06fc5e4ecbcd773f5a226909892ac162bcf92a616adf75ea29c38995d75d3e83
3
+ size 358
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 92.01,
3
+ "train_loss": 0.04092751889359206,
4
+ "train_runtime": 27870.4476,
5
+ "train_samples_per_second": 2.87,
6
+ "train_steps_per_second": 0.179
7
+ }
trainer_state.json ADDED
@@ -0,0 +1,1320 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 34.623217922606926,
3
+ "best_model_checkpoint": "./checkpoint-5000",
4
+ "epoch": 92.0064,
5
+ "eval_steps": 500,
6
+ "global_step": 5000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 4.0000000000000003e-07,
14
+ "loss": 1.7607,
15
+ "step": 25
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 9.000000000000001e-07,
20
+ "loss": 1.3297,
21
+ "step": 50
22
+ },
23
+ {
24
+ "epoch": 1.0,
25
+ "learning_rate": 1.4000000000000001e-06,
26
+ "loss": 0.8829,
27
+ "step": 75
28
+ },
29
+ {
30
+ "epoch": 1.01,
31
+ "learning_rate": 1.9000000000000002e-06,
32
+ "loss": 0.7389,
33
+ "step": 100
34
+ },
35
+ {
36
+ "epoch": 2.0,
37
+ "learning_rate": 2.4000000000000003e-06,
38
+ "loss": 0.5364,
39
+ "step": 125
40
+ },
41
+ {
42
+ "epoch": 2.01,
43
+ "learning_rate": 2.88e-06,
44
+ "loss": 0.4617,
45
+ "step": 150
46
+ },
47
+ {
48
+ "epoch": 3.0,
49
+ "learning_rate": 3.3800000000000007e-06,
50
+ "loss": 0.3465,
51
+ "step": 175
52
+ },
53
+ {
54
+ "epoch": 3.01,
55
+ "learning_rate": 3.86e-06,
56
+ "loss": 0.2559,
57
+ "step": 200
58
+ },
59
+ {
60
+ "epoch": 4.0,
61
+ "learning_rate": 4.360000000000001e-06,
62
+ "loss": 0.2157,
63
+ "step": 225
64
+ },
65
+ {
66
+ "epoch": 4.01,
67
+ "learning_rate": 4.86e-06,
68
+ "loss": 0.1294,
69
+ "step": 250
70
+ },
71
+ {
72
+ "epoch": 5.0,
73
+ "learning_rate": 5.36e-06,
74
+ "loss": 0.1439,
75
+ "step": 275
76
+ },
77
+ {
78
+ "epoch": 5.01,
79
+ "learning_rate": 5.86e-06,
80
+ "loss": 0.0834,
81
+ "step": 300
82
+ },
83
+ {
84
+ "epoch": 6.0,
85
+ "learning_rate": 6.360000000000001e-06,
86
+ "loss": 0.0977,
87
+ "step": 325
88
+ },
89
+ {
90
+ "epoch": 6.01,
91
+ "learning_rate": 6.860000000000001e-06,
92
+ "loss": 0.0593,
93
+ "step": 350
94
+ },
95
+ {
96
+ "epoch": 6.01,
97
+ "learning_rate": 7.360000000000001e-06,
98
+ "loss": 0.0798,
99
+ "step": 375
100
+ },
101
+ {
102
+ "epoch": 7.0,
103
+ "learning_rate": 7.860000000000001e-06,
104
+ "loss": 0.0495,
105
+ "step": 400
106
+ },
107
+ {
108
+ "epoch": 7.01,
109
+ "learning_rate": 8.36e-06,
110
+ "loss": 0.0618,
111
+ "step": 425
112
+ },
113
+ {
114
+ "epoch": 8.0,
115
+ "learning_rate": 8.860000000000002e-06,
116
+ "loss": 0.0597,
117
+ "step": 450
118
+ },
119
+ {
120
+ "epoch": 8.01,
121
+ "learning_rate": 9.360000000000002e-06,
122
+ "loss": 0.0561,
123
+ "step": 475
124
+ },
125
+ {
126
+ "epoch": 9.0,
127
+ "learning_rate": 9.86e-06,
128
+ "loss": 0.0515,
129
+ "step": 500
130
+ },
131
+ {
132
+ "epoch": 9.0,
133
+ "eval_loss": 0.6733026504516602,
134
+ "eval_runtime": 110.5203,
135
+ "eval_samples_per_second": 3.474,
136
+ "eval_steps_per_second": 0.217,
137
+ "eval_wer": 42.484725050916495,
138
+ "step": 500
139
+ },
140
+ {
141
+ "epoch": 9.01,
142
+ "learning_rate": 9.960000000000001e-06,
143
+ "loss": 0.0518,
144
+ "step": 525
145
+ },
146
+ {
147
+ "epoch": 10.0,
148
+ "learning_rate": 9.904444444444445e-06,
149
+ "loss": 0.0528,
150
+ "step": 550
151
+ },
152
+ {
153
+ "epoch": 10.01,
154
+ "learning_rate": 9.84888888888889e-06,
155
+ "loss": 0.0451,
156
+ "step": 575
157
+ },
158
+ {
159
+ "epoch": 11.0,
160
+ "learning_rate": 9.793333333333333e-06,
161
+ "loss": 0.0403,
162
+ "step": 600
163
+ },
164
+ {
165
+ "epoch": 11.01,
166
+ "learning_rate": 9.737777777777779e-06,
167
+ "loss": 0.0382,
168
+ "step": 625
169
+ },
170
+ {
171
+ "epoch": 12.0,
172
+ "learning_rate": 9.682222222222223e-06,
173
+ "loss": 0.0432,
174
+ "step": 650
175
+ },
176
+ {
177
+ "epoch": 12.01,
178
+ "learning_rate": 9.626666666666667e-06,
179
+ "loss": 0.0251,
180
+ "step": 675
181
+ },
182
+ {
183
+ "epoch": 12.01,
184
+ "learning_rate": 9.571111111111113e-06,
185
+ "loss": 0.0238,
186
+ "step": 700
187
+ },
188
+ {
189
+ "epoch": 13.0,
190
+ "learning_rate": 9.515555555555557e-06,
191
+ "loss": 0.0256,
192
+ "step": 725
193
+ },
194
+ {
195
+ "epoch": 13.01,
196
+ "learning_rate": 9.460000000000001e-06,
197
+ "loss": 0.0285,
198
+ "step": 750
199
+ },
200
+ {
201
+ "epoch": 14.0,
202
+ "learning_rate": 9.404444444444445e-06,
203
+ "loss": 0.0224,
204
+ "step": 775
205
+ },
206
+ {
207
+ "epoch": 14.01,
208
+ "learning_rate": 9.348888888888889e-06,
209
+ "loss": 0.0171,
210
+ "step": 800
211
+ },
212
+ {
213
+ "epoch": 15.0,
214
+ "learning_rate": 9.293333333333335e-06,
215
+ "loss": 0.0144,
216
+ "step": 825
217
+ },
218
+ {
219
+ "epoch": 15.01,
220
+ "learning_rate": 9.237777777777779e-06,
221
+ "loss": 0.0224,
222
+ "step": 850
223
+ },
224
+ {
225
+ "epoch": 16.0,
226
+ "learning_rate": 9.182222222222223e-06,
227
+ "loss": 0.0189,
228
+ "step": 875
229
+ },
230
+ {
231
+ "epoch": 16.01,
232
+ "learning_rate": 9.126666666666667e-06,
233
+ "loss": 0.0128,
234
+ "step": 900
235
+ },
236
+ {
237
+ "epoch": 17.0,
238
+ "learning_rate": 9.07111111111111e-06,
239
+ "loss": 0.0114,
240
+ "step": 925
241
+ },
242
+ {
243
+ "epoch": 17.01,
244
+ "learning_rate": 9.015555555555557e-06,
245
+ "loss": 0.012,
246
+ "step": 950
247
+ },
248
+ {
249
+ "epoch": 18.0,
250
+ "learning_rate": 8.96e-06,
251
+ "loss": 0.0137,
252
+ "step": 975
253
+ },
254
+ {
255
+ "epoch": 18.01,
256
+ "learning_rate": 8.904444444444446e-06,
257
+ "loss": 0.0101,
258
+ "step": 1000
259
+ },
260
+ {
261
+ "epoch": 18.01,
262
+ "eval_loss": 0.680969774723053,
263
+ "eval_runtime": 105.2408,
264
+ "eval_samples_per_second": 3.649,
265
+ "eval_steps_per_second": 0.228,
266
+ "eval_wer": 37.596741344195515,
267
+ "step": 1000
268
+ },
269
+ {
270
+ "epoch": 18.01,
271
+ "learning_rate": 8.848888888888889e-06,
272
+ "loss": 0.0115,
273
+ "step": 1025
274
+ },
275
+ {
276
+ "epoch": 19.0,
277
+ "learning_rate": 8.793333333333334e-06,
278
+ "loss": 0.0065,
279
+ "step": 1050
280
+ },
281
+ {
282
+ "epoch": 19.01,
283
+ "learning_rate": 8.737777777777778e-06,
284
+ "loss": 0.012,
285
+ "step": 1075
286
+ },
287
+ {
288
+ "epoch": 20.0,
289
+ "learning_rate": 8.682222222222222e-06,
290
+ "loss": 0.008,
291
+ "step": 1100
292
+ },
293
+ {
294
+ "epoch": 20.01,
295
+ "learning_rate": 8.626666666666668e-06,
296
+ "loss": 0.0106,
297
+ "step": 1125
298
+ },
299
+ {
300
+ "epoch": 21.0,
301
+ "learning_rate": 8.571111111111112e-06,
302
+ "loss": 0.0084,
303
+ "step": 1150
304
+ },
305
+ {
306
+ "epoch": 21.01,
307
+ "learning_rate": 8.515555555555556e-06,
308
+ "loss": 0.0065,
309
+ "step": 1175
310
+ },
311
+ {
312
+ "epoch": 22.0,
313
+ "learning_rate": 8.46e-06,
314
+ "loss": 0.0066,
315
+ "step": 1200
316
+ },
317
+ {
318
+ "epoch": 22.01,
319
+ "learning_rate": 8.404444444444444e-06,
320
+ "loss": 0.0118,
321
+ "step": 1225
322
+ },
323
+ {
324
+ "epoch": 23.0,
325
+ "learning_rate": 8.34888888888889e-06,
326
+ "loss": 0.0077,
327
+ "step": 1250
328
+ },
329
+ {
330
+ "epoch": 23.01,
331
+ "learning_rate": 8.293333333333334e-06,
332
+ "loss": 0.0083,
333
+ "step": 1275
334
+ },
335
+ {
336
+ "epoch": 24.0,
337
+ "learning_rate": 8.237777777777778e-06,
338
+ "loss": 0.0111,
339
+ "step": 1300
340
+ },
341
+ {
342
+ "epoch": 24.01,
343
+ "learning_rate": 8.182222222222222e-06,
344
+ "loss": 0.0075,
345
+ "step": 1325
346
+ },
347
+ {
348
+ "epoch": 24.01,
349
+ "learning_rate": 8.126666666666668e-06,
350
+ "loss": 0.0083,
351
+ "step": 1350
352
+ },
353
+ {
354
+ "epoch": 25.0,
355
+ "learning_rate": 8.071111111111112e-06,
356
+ "loss": 0.0043,
357
+ "step": 1375
358
+ },
359
+ {
360
+ "epoch": 25.01,
361
+ "learning_rate": 8.015555555555556e-06,
362
+ "loss": 0.0074,
363
+ "step": 1400
364
+ },
365
+ {
366
+ "epoch": 26.0,
367
+ "learning_rate": 7.960000000000002e-06,
368
+ "loss": 0.0047,
369
+ "step": 1425
370
+ },
371
+ {
372
+ "epoch": 26.01,
373
+ "learning_rate": 7.904444444444444e-06,
374
+ "loss": 0.0057,
375
+ "step": 1450
376
+ },
377
+ {
378
+ "epoch": 27.0,
379
+ "learning_rate": 7.84888888888889e-06,
380
+ "loss": 0.0077,
381
+ "step": 1475
382
+ },
383
+ {
384
+ "epoch": 27.01,
385
+ "learning_rate": 7.793333333333334e-06,
386
+ "loss": 0.0074,
387
+ "step": 1500
388
+ },
389
+ {
390
+ "epoch": 27.01,
391
+ "eval_loss": 0.7185311317443848,
392
+ "eval_runtime": 104.8405,
393
+ "eval_samples_per_second": 3.663,
394
+ "eval_steps_per_second": 0.229,
395
+ "eval_wer": 38.085539714867615,
396
+ "step": 1500
397
+ },
398
+ {
399
+ "epoch": 28.0,
400
+ "learning_rate": 7.737777777777778e-06,
401
+ "loss": 0.0059,
402
+ "step": 1525
403
+ },
404
+ {
405
+ "epoch": 28.01,
406
+ "learning_rate": 7.682222222222224e-06,
407
+ "loss": 0.0085,
408
+ "step": 1550
409
+ },
410
+ {
411
+ "epoch": 29.0,
412
+ "learning_rate": 7.626666666666668e-06,
413
+ "loss": 0.0076,
414
+ "step": 1575
415
+ },
416
+ {
417
+ "epoch": 29.01,
418
+ "learning_rate": 7.571111111111112e-06,
419
+ "loss": 0.0019,
420
+ "step": 1600
421
+ },
422
+ {
423
+ "epoch": 30.0,
424
+ "learning_rate": 7.515555555555556e-06,
425
+ "loss": 0.0078,
426
+ "step": 1625
427
+ },
428
+ {
429
+ "epoch": 30.01,
430
+ "learning_rate": 7.4600000000000006e-06,
431
+ "loss": 0.0032,
432
+ "step": 1650
433
+ },
434
+ {
435
+ "epoch": 31.0,
436
+ "learning_rate": 7.4044444444444455e-06,
437
+ "loss": 0.0027,
438
+ "step": 1675
439
+ },
440
+ {
441
+ "epoch": 31.01,
442
+ "learning_rate": 7.3488888888888895e-06,
443
+ "loss": 0.0034,
444
+ "step": 1700
445
+ },
446
+ {
447
+ "epoch": 31.01,
448
+ "learning_rate": 7.2933333333333335e-06,
449
+ "loss": 0.0036,
450
+ "step": 1725
451
+ },
452
+ {
453
+ "epoch": 32.0,
454
+ "learning_rate": 7.237777777777778e-06,
455
+ "loss": 0.0024,
456
+ "step": 1750
457
+ },
458
+ {
459
+ "epoch": 32.01,
460
+ "learning_rate": 7.1822222222222224e-06,
461
+ "loss": 0.001,
462
+ "step": 1775
463
+ },
464
+ {
465
+ "epoch": 33.0,
466
+ "learning_rate": 7.126666666666667e-06,
467
+ "loss": 0.0007,
468
+ "step": 1800
469
+ },
470
+ {
471
+ "epoch": 33.01,
472
+ "learning_rate": 7.071111111111112e-06,
473
+ "loss": 0.0057,
474
+ "step": 1825
475
+ },
476
+ {
477
+ "epoch": 34.0,
478
+ "learning_rate": 7.015555555555556e-06,
479
+ "loss": 0.0052,
480
+ "step": 1850
481
+ },
482
+ {
483
+ "epoch": 34.01,
484
+ "learning_rate": 6.96e-06,
485
+ "loss": 0.0043,
486
+ "step": 1875
487
+ },
488
+ {
489
+ "epoch": 35.0,
490
+ "learning_rate": 6.904444444444444e-06,
491
+ "loss": 0.0021,
492
+ "step": 1900
493
+ },
494
+ {
495
+ "epoch": 35.01,
496
+ "learning_rate": 6.848888888888889e-06,
497
+ "loss": 0.0021,
498
+ "step": 1925
499
+ },
500
+ {
501
+ "epoch": 36.0,
502
+ "learning_rate": 6.793333333333334e-06,
503
+ "loss": 0.004,
504
+ "step": 1950
505
+ },
506
+ {
507
+ "epoch": 36.01,
508
+ "learning_rate": 6.737777777777779e-06,
509
+ "loss": 0.0016,
510
+ "step": 1975
511
+ },
512
+ {
513
+ "epoch": 37.0,
514
+ "learning_rate": 6.682222222222223e-06,
515
+ "loss": 0.0009,
516
+ "step": 2000
517
+ },
518
+ {
519
+ "epoch": 37.0,
520
+ "eval_loss": 0.6986598372459412,
521
+ "eval_runtime": 104.187,
522
+ "eval_samples_per_second": 3.686,
523
+ "eval_steps_per_second": 0.23,
524
+ "eval_wer": 35.5193482688391,
525
+ "step": 2000
526
+ },
527
+ {
528
+ "epoch": 37.01,
529
+ "learning_rate": 6.626666666666667e-06,
530
+ "loss": 0.0014,
531
+ "step": 2025
532
+ },
533
+ {
534
+ "epoch": 37.01,
535
+ "learning_rate": 6.571111111111111e-06,
536
+ "loss": 0.0008,
537
+ "step": 2050
538
+ },
539
+ {
540
+ "epoch": 38.0,
541
+ "learning_rate": 6.515555555555556e-06,
542
+ "loss": 0.0012,
543
+ "step": 2075
544
+ },
545
+ {
546
+ "epoch": 38.01,
547
+ "learning_rate": 6.460000000000001e-06,
548
+ "loss": 0.0007,
549
+ "step": 2100
550
+ },
551
+ {
552
+ "epoch": 39.0,
553
+ "learning_rate": 6.404444444444446e-06,
554
+ "loss": 0.0005,
555
+ "step": 2125
556
+ },
557
+ {
558
+ "epoch": 39.01,
559
+ "learning_rate": 6.348888888888889e-06,
560
+ "loss": 0.0025,
561
+ "step": 2150
562
+ },
563
+ {
564
+ "epoch": 40.0,
565
+ "learning_rate": 6.293333333333334e-06,
566
+ "loss": 0.0003,
567
+ "step": 2175
568
+ },
569
+ {
570
+ "epoch": 40.01,
571
+ "learning_rate": 6.237777777777778e-06,
572
+ "loss": 0.0004,
573
+ "step": 2200
574
+ },
575
+ {
576
+ "epoch": 41.0,
577
+ "learning_rate": 6.182222222222223e-06,
578
+ "loss": 0.0003,
579
+ "step": 2225
580
+ },
581
+ {
582
+ "epoch": 41.01,
583
+ "learning_rate": 6.126666666666668e-06,
584
+ "loss": 0.0002,
585
+ "step": 2250
586
+ },
587
+ {
588
+ "epoch": 42.0,
589
+ "learning_rate": 6.0711111111111125e-06,
590
+ "loss": 0.0002,
591
+ "step": 2275
592
+ },
593
+ {
594
+ "epoch": 42.01,
595
+ "learning_rate": 6.015555555555556e-06,
596
+ "loss": 0.0002,
597
+ "step": 2300
598
+ },
599
+ {
600
+ "epoch": 43.0,
601
+ "learning_rate": 5.9600000000000005e-06,
602
+ "loss": 0.0002,
603
+ "step": 2325
604
+ },
605
+ {
606
+ "epoch": 43.01,
607
+ "learning_rate": 5.9044444444444446e-06,
608
+ "loss": 0.0002,
609
+ "step": 2350
610
+ },
611
+ {
612
+ "epoch": 43.01,
613
+ "learning_rate": 5.8488888888888895e-06,
614
+ "loss": 0.0002,
615
+ "step": 2375
616
+ },
617
+ {
618
+ "epoch": 44.0,
619
+ "learning_rate": 5.793333333333334e-06,
620
+ "loss": 0.0002,
621
+ "step": 2400
622
+ },
623
+ {
624
+ "epoch": 44.01,
625
+ "learning_rate": 5.737777777777778e-06,
626
+ "loss": 0.0002,
627
+ "step": 2425
628
+ },
629
+ {
630
+ "epoch": 45.0,
631
+ "learning_rate": 5.682222222222222e-06,
632
+ "loss": 0.0002,
633
+ "step": 2450
634
+ },
635
+ {
636
+ "epoch": 45.01,
637
+ "learning_rate": 5.626666666666667e-06,
638
+ "loss": 0.0002,
639
+ "step": 2475
640
+ },
641
+ {
642
+ "epoch": 46.0,
643
+ "learning_rate": 5.571111111111111e-06,
644
+ "loss": 0.0002,
645
+ "step": 2500
646
+ },
647
+ {
648
+ "epoch": 46.0,
649
+ "eval_loss": 0.7392615675926208,
650
+ "eval_runtime": 104.4509,
651
+ "eval_samples_per_second": 3.676,
652
+ "eval_steps_per_second": 0.23,
653
+ "eval_wer": 35.03054989816701,
654
+ "step": 2500
655
+ },
656
+ {
657
+ "epoch": 46.01,
658
+ "learning_rate": 5.515555555555556e-06,
659
+ "loss": 0.0002,
660
+ "step": 2525
661
+ },
662
+ {
663
+ "epoch": 47.0,
664
+ "learning_rate": 5.460000000000001e-06,
665
+ "loss": 0.0002,
666
+ "step": 2550
667
+ },
668
+ {
669
+ "epoch": 47.01,
670
+ "learning_rate": 5.404444444444444e-06,
671
+ "loss": 0.0001,
672
+ "step": 2575
673
+ },
674
+ {
675
+ "epoch": 48.0,
676
+ "learning_rate": 5.348888888888889e-06,
677
+ "loss": 0.0001,
678
+ "step": 2600
679
+ },
680
+ {
681
+ "epoch": 48.01,
682
+ "learning_rate": 5.293333333333334e-06,
683
+ "loss": 0.0001,
684
+ "step": 2625
685
+ },
686
+ {
687
+ "epoch": 49.0,
688
+ "learning_rate": 5.237777777777778e-06,
689
+ "loss": 0.0001,
690
+ "step": 2650
691
+ },
692
+ {
693
+ "epoch": 49.01,
694
+ "learning_rate": 5.182222222222223e-06,
695
+ "loss": 0.0001,
696
+ "step": 2675
697
+ },
698
+ {
699
+ "epoch": 49.01,
700
+ "learning_rate": 5.126666666666668e-06,
701
+ "loss": 0.0001,
702
+ "step": 2700
703
+ },
704
+ {
705
+ "epoch": 50.01,
706
+ "learning_rate": 5.071111111111111e-06,
707
+ "loss": 0.0001,
708
+ "step": 2725
709
+ },
710
+ {
711
+ "epoch": 50.01,
712
+ "learning_rate": 5.015555555555556e-06,
713
+ "loss": 0.0001,
714
+ "step": 2750
715
+ },
716
+ {
717
+ "epoch": 51.0,
718
+ "learning_rate": 4.960000000000001e-06,
719
+ "loss": 0.0001,
720
+ "step": 2775
721
+ },
722
+ {
723
+ "epoch": 51.01,
724
+ "learning_rate": 4.904444444444445e-06,
725
+ "loss": 0.0001,
726
+ "step": 2800
727
+ },
728
+ {
729
+ "epoch": 52.0,
730
+ "learning_rate": 4.848888888888889e-06,
731
+ "loss": 0.0001,
732
+ "step": 2825
733
+ },
734
+ {
735
+ "epoch": 52.01,
736
+ "learning_rate": 4.793333333333334e-06,
737
+ "loss": 0.0001,
738
+ "step": 2850
739
+ },
740
+ {
741
+ "epoch": 53.0,
742
+ "learning_rate": 4.737777777777779e-06,
743
+ "loss": 0.0001,
744
+ "step": 2875
745
+ },
746
+ {
747
+ "epoch": 53.01,
748
+ "learning_rate": 4.682222222222223e-06,
749
+ "loss": 0.0001,
750
+ "step": 2900
751
+ },
752
+ {
753
+ "epoch": 54.0,
754
+ "learning_rate": 4.626666666666667e-06,
755
+ "loss": 0.0001,
756
+ "step": 2925
757
+ },
758
+ {
759
+ "epoch": 54.01,
760
+ "learning_rate": 4.571111111111112e-06,
761
+ "loss": 0.0001,
762
+ "step": 2950
763
+ },
764
+ {
765
+ "epoch": 55.0,
766
+ "learning_rate": 4.515555555555556e-06,
767
+ "loss": 0.0001,
768
+ "step": 2975
769
+ },
770
+ {
771
+ "epoch": 55.01,
772
+ "learning_rate": 4.4600000000000005e-06,
773
+ "loss": 0.0001,
774
+ "step": 3000
775
+ },
776
+ {
777
+ "epoch": 55.01,
778
+ "eval_loss": 0.7602861523628235,
779
+ "eval_runtime": 104.4035,
780
+ "eval_samples_per_second": 3.678,
781
+ "eval_steps_per_second": 0.23,
782
+ "eval_wer": 35.03054989816701,
783
+ "step": 3000
784
+ },
785
+ {
786
+ "epoch": 56.0,
787
+ "learning_rate": 4.404444444444445e-06,
788
+ "loss": 0.0001,
789
+ "step": 3025
790
+ },
791
+ {
792
+ "epoch": 56.01,
793
+ "learning_rate": 4.348888888888889e-06,
794
+ "loss": 0.0001,
795
+ "step": 3050
796
+ },
797
+ {
798
+ "epoch": 56.01,
799
+ "learning_rate": 4.2933333333333334e-06,
800
+ "loss": 0.0001,
801
+ "step": 3075
802
+ },
803
+ {
804
+ "epoch": 57.0,
805
+ "learning_rate": 4.2377777777777775e-06,
806
+ "loss": 0.0001,
807
+ "step": 3100
808
+ },
809
+ {
810
+ "epoch": 57.01,
811
+ "learning_rate": 4.182222222222222e-06,
812
+ "loss": 0.0001,
813
+ "step": 3125
814
+ },
815
+ {
816
+ "epoch": 58.0,
817
+ "learning_rate": 4.126666666666667e-06,
818
+ "loss": 0.0001,
819
+ "step": 3150
820
+ },
821
+ {
822
+ "epoch": 58.01,
823
+ "learning_rate": 4.071111111111111e-06,
824
+ "loss": 0.0001,
825
+ "step": 3175
826
+ },
827
+ {
828
+ "epoch": 59.0,
829
+ "learning_rate": 4.015555555555556e-06,
830
+ "loss": 0.0001,
831
+ "step": 3200
832
+ },
833
+ {
834
+ "epoch": 59.01,
835
+ "learning_rate": 3.96e-06,
836
+ "loss": 0.0001,
837
+ "step": 3225
838
+ },
839
+ {
840
+ "epoch": 60.0,
841
+ "learning_rate": 3.904444444444444e-06,
842
+ "loss": 0.0001,
843
+ "step": 3250
844
+ },
845
+ {
846
+ "epoch": 60.01,
847
+ "learning_rate": 3.848888888888889e-06,
848
+ "loss": 0.0001,
849
+ "step": 3275
850
+ },
851
+ {
852
+ "epoch": 61.0,
853
+ "learning_rate": 3.793333333333334e-06,
854
+ "loss": 0.0001,
855
+ "step": 3300
856
+ },
857
+ {
858
+ "epoch": 61.01,
859
+ "learning_rate": 3.737777777777778e-06,
860
+ "loss": 0.0001,
861
+ "step": 3325
862
+ },
863
+ {
864
+ "epoch": 62.0,
865
+ "learning_rate": 3.6822222222222225e-06,
866
+ "loss": 0.0001,
867
+ "step": 3350
868
+ },
869
+ {
870
+ "epoch": 62.01,
871
+ "learning_rate": 3.6266666666666674e-06,
872
+ "loss": 0.0001,
873
+ "step": 3375
874
+ },
875
+ {
876
+ "epoch": 62.01,
877
+ "learning_rate": 3.5711111111111114e-06,
878
+ "loss": 0.0001,
879
+ "step": 3400
880
+ },
881
+ {
882
+ "epoch": 63.0,
883
+ "learning_rate": 3.515555555555556e-06,
884
+ "loss": 0.0001,
885
+ "step": 3425
886
+ },
887
+ {
888
+ "epoch": 63.01,
889
+ "learning_rate": 3.46e-06,
890
+ "loss": 0.0001,
891
+ "step": 3450
892
+ },
893
+ {
894
+ "epoch": 64.0,
895
+ "learning_rate": 3.404444444444445e-06,
896
+ "loss": 0.0001,
897
+ "step": 3475
898
+ },
899
+ {
900
+ "epoch": 64.01,
901
+ "learning_rate": 3.3488888888888892e-06,
902
+ "loss": 0.0001,
903
+ "step": 3500
904
+ },
905
+ {
906
+ "epoch": 64.01,
907
+ "eval_loss": 0.7738750576972961,
908
+ "eval_runtime": 104.8601,
909
+ "eval_samples_per_second": 3.662,
910
+ "eval_steps_per_second": 0.229,
911
+ "eval_wer": 34.86761710794297,
912
+ "step": 3500
913
+ },
914
+ {
915
+ "epoch": 65.0,
916
+ "learning_rate": 3.2933333333333333e-06,
917
+ "loss": 0.0001,
918
+ "step": 3525
919
+ },
920
+ {
921
+ "epoch": 65.01,
922
+ "learning_rate": 3.237777777777778e-06,
923
+ "loss": 0.0001,
924
+ "step": 3550
925
+ },
926
+ {
927
+ "epoch": 66.0,
928
+ "learning_rate": 3.1822222222222226e-06,
929
+ "loss": 0.0001,
930
+ "step": 3575
931
+ },
932
+ {
933
+ "epoch": 66.01,
934
+ "learning_rate": 3.1266666666666667e-06,
935
+ "loss": 0.0001,
936
+ "step": 3600
937
+ },
938
+ {
939
+ "epoch": 67.0,
940
+ "learning_rate": 3.0711111111111115e-06,
941
+ "loss": 0.0001,
942
+ "step": 3625
943
+ },
944
+ {
945
+ "epoch": 67.01,
946
+ "learning_rate": 3.015555555555556e-06,
947
+ "loss": 0.0001,
948
+ "step": 3650
949
+ },
950
+ {
951
+ "epoch": 68.0,
952
+ "learning_rate": 2.96e-06,
953
+ "loss": 0.0001,
954
+ "step": 3675
955
+ },
956
+ {
957
+ "epoch": 68.01,
958
+ "learning_rate": 2.904444444444445e-06,
959
+ "loss": 0.0001,
960
+ "step": 3700
961
+ },
962
+ {
963
+ "epoch": 68.01,
964
+ "learning_rate": 2.8488888888888894e-06,
965
+ "loss": 0.0001,
966
+ "step": 3725
967
+ },
968
+ {
969
+ "epoch": 69.0,
970
+ "learning_rate": 2.7933333333333334e-06,
971
+ "loss": 0.0001,
972
+ "step": 3750
973
+ },
974
+ {
975
+ "epoch": 69.01,
976
+ "learning_rate": 2.7377777777777783e-06,
977
+ "loss": 0.0001,
978
+ "step": 3775
979
+ },
980
+ {
981
+ "epoch": 70.0,
982
+ "learning_rate": 2.6822222222222223e-06,
983
+ "loss": 0.0001,
984
+ "step": 3800
985
+ },
986
+ {
987
+ "epoch": 70.01,
988
+ "learning_rate": 2.6266666666666668e-06,
989
+ "loss": 0.0001,
990
+ "step": 3825
991
+ },
992
+ {
993
+ "epoch": 71.0,
994
+ "learning_rate": 2.5711111111111112e-06,
995
+ "loss": 0.0001,
996
+ "step": 3850
997
+ },
998
+ {
999
+ "epoch": 71.01,
1000
+ "learning_rate": 2.5155555555555557e-06,
1001
+ "loss": 0.0001,
1002
+ "step": 3875
1003
+ },
1004
+ {
1005
+ "epoch": 72.0,
1006
+ "learning_rate": 2.46e-06,
1007
+ "loss": 0.0001,
1008
+ "step": 3900
1009
+ },
1010
+ {
1011
+ "epoch": 72.01,
1012
+ "learning_rate": 2.4044444444444446e-06,
1013
+ "loss": 0.0001,
1014
+ "step": 3925
1015
+ },
1016
+ {
1017
+ "epoch": 73.0,
1018
+ "learning_rate": 2.348888888888889e-06,
1019
+ "loss": 0.0001,
1020
+ "step": 3950
1021
+ },
1022
+ {
1023
+ "epoch": 73.01,
1024
+ "learning_rate": 2.2933333333333335e-06,
1025
+ "loss": 0.0001,
1026
+ "step": 3975
1027
+ },
1028
+ {
1029
+ "epoch": 74.0,
1030
+ "learning_rate": 2.237777777777778e-06,
1031
+ "loss": 0.0001,
1032
+ "step": 4000
1033
+ },
1034
+ {
1035
+ "epoch": 74.0,
1036
+ "eval_loss": 0.7832472920417786,
1037
+ "eval_runtime": 104.6874,
1038
+ "eval_samples_per_second": 3.668,
1039
+ "eval_steps_per_second": 0.229,
1040
+ "eval_wer": 34.82688391038696,
1041
+ "step": 4000
1042
+ },
1043
+ {
1044
+ "epoch": 74.01,
1045
+ "learning_rate": 2.1822222222222225e-06,
1046
+ "loss": 0.0001,
1047
+ "step": 4025
1048
+ },
1049
+ {
1050
+ "epoch": 74.01,
1051
+ "learning_rate": 2.126666666666667e-06,
1052
+ "loss": 0.0001,
1053
+ "step": 4050
1054
+ },
1055
+ {
1056
+ "epoch": 75.0,
1057
+ "learning_rate": 2.0711111111111114e-06,
1058
+ "loss": 0.0001,
1059
+ "step": 4075
1060
+ },
1061
+ {
1062
+ "epoch": 75.01,
1063
+ "learning_rate": 2.0155555555555554e-06,
1064
+ "loss": 0.0001,
1065
+ "step": 4100
1066
+ },
1067
+ {
1068
+ "epoch": 76.0,
1069
+ "learning_rate": 1.9600000000000003e-06,
1070
+ "loss": 0.0001,
1071
+ "step": 4125
1072
+ },
1073
+ {
1074
+ "epoch": 76.01,
1075
+ "learning_rate": 1.9044444444444445e-06,
1076
+ "loss": 0.0001,
1077
+ "step": 4150
1078
+ },
1079
+ {
1080
+ "epoch": 77.0,
1081
+ "learning_rate": 1.848888888888889e-06,
1082
+ "loss": 0.0001,
1083
+ "step": 4175
1084
+ },
1085
+ {
1086
+ "epoch": 77.01,
1087
+ "learning_rate": 1.7933333333333337e-06,
1088
+ "loss": 0.0001,
1089
+ "step": 4200
1090
+ },
1091
+ {
1092
+ "epoch": 78.0,
1093
+ "learning_rate": 1.737777777777778e-06,
1094
+ "loss": 0.0001,
1095
+ "step": 4225
1096
+ },
1097
+ {
1098
+ "epoch": 78.01,
1099
+ "learning_rate": 1.6822222222222224e-06,
1100
+ "loss": 0.0001,
1101
+ "step": 4250
1102
+ },
1103
+ {
1104
+ "epoch": 79.0,
1105
+ "learning_rate": 1.6266666666666666e-06,
1106
+ "loss": 0.0001,
1107
+ "step": 4275
1108
+ },
1109
+ {
1110
+ "epoch": 79.01,
1111
+ "learning_rate": 1.5711111111111113e-06,
1112
+ "loss": 0.0001,
1113
+ "step": 4300
1114
+ },
1115
+ {
1116
+ "epoch": 80.0,
1117
+ "learning_rate": 1.5155555555555558e-06,
1118
+ "loss": 0.0001,
1119
+ "step": 4325
1120
+ },
1121
+ {
1122
+ "epoch": 80.01,
1123
+ "learning_rate": 1.46e-06,
1124
+ "loss": 0.0001,
1125
+ "step": 4350
1126
+ },
1127
+ {
1128
+ "epoch": 81.0,
1129
+ "learning_rate": 1.4044444444444447e-06,
1130
+ "loss": 0.0001,
1131
+ "step": 4375
1132
+ },
1133
+ {
1134
+ "epoch": 81.01,
1135
+ "learning_rate": 1.3488888888888891e-06,
1136
+ "loss": 0.0001,
1137
+ "step": 4400
1138
+ },
1139
+ {
1140
+ "epoch": 81.01,
1141
+ "learning_rate": 1.2933333333333334e-06,
1142
+ "loss": 0.0001,
1143
+ "step": 4425
1144
+ },
1145
+ {
1146
+ "epoch": 82.0,
1147
+ "learning_rate": 1.2377777777777778e-06,
1148
+ "loss": 0.0001,
1149
+ "step": 4450
1150
+ },
1151
+ {
1152
+ "epoch": 82.01,
1153
+ "learning_rate": 1.1822222222222223e-06,
1154
+ "loss": 0.0001,
1155
+ "step": 4475
1156
+ },
1157
+ {
1158
+ "epoch": 83.0,
1159
+ "learning_rate": 1.1266666666666667e-06,
1160
+ "loss": 0.0001,
1161
+ "step": 4500
1162
+ },
1163
+ {
1164
+ "epoch": 83.0,
1165
+ "eval_loss": 0.7895247936248779,
1166
+ "eval_runtime": 104.5178,
1167
+ "eval_samples_per_second": 3.674,
1168
+ "eval_steps_per_second": 0.23,
1169
+ "eval_wer": 34.90835030549898,
1170
+ "step": 4500
1171
+ },
1172
+ {
1173
+ "epoch": 83.01,
1174
+ "learning_rate": 1.0711111111111112e-06,
1175
+ "loss": 0.0001,
1176
+ "step": 4525
1177
+ },
1178
+ {
1179
+ "epoch": 84.0,
1180
+ "learning_rate": 1.0155555555555557e-06,
1181
+ "loss": 0.0001,
1182
+ "step": 4550
1183
+ },
1184
+ {
1185
+ "epoch": 84.01,
1186
+ "learning_rate": 9.600000000000001e-07,
1187
+ "loss": 0.0001,
1188
+ "step": 4575
1189
+ },
1190
+ {
1191
+ "epoch": 85.0,
1192
+ "learning_rate": 9.044444444444445e-07,
1193
+ "loss": 0.0001,
1194
+ "step": 4600
1195
+ },
1196
+ {
1197
+ "epoch": 85.01,
1198
+ "learning_rate": 8.488888888888889e-07,
1199
+ "loss": 0.0001,
1200
+ "step": 4625
1201
+ },
1202
+ {
1203
+ "epoch": 86.0,
1204
+ "learning_rate": 7.933333333333335e-07,
1205
+ "loss": 0.0001,
1206
+ "step": 4650
1207
+ },
1208
+ {
1209
+ "epoch": 86.01,
1210
+ "learning_rate": 7.377777777777779e-07,
1211
+ "loss": 0.0001,
1212
+ "step": 4675
1213
+ },
1214
+ {
1215
+ "epoch": 87.0,
1216
+ "learning_rate": 6.822222222222223e-07,
1217
+ "loss": 0.0001,
1218
+ "step": 4700
1219
+ },
1220
+ {
1221
+ "epoch": 87.01,
1222
+ "learning_rate": 6.266666666666667e-07,
1223
+ "loss": 0.0001,
1224
+ "step": 4725
1225
+ },
1226
+ {
1227
+ "epoch": 87.01,
1228
+ "learning_rate": 5.711111111111111e-07,
1229
+ "loss": 0.0001,
1230
+ "step": 4750
1231
+ },
1232
+ {
1233
+ "epoch": 88.0,
1234
+ "learning_rate": 5.155555555555556e-07,
1235
+ "loss": 0.0001,
1236
+ "step": 4775
1237
+ },
1238
+ {
1239
+ "epoch": 88.01,
1240
+ "learning_rate": 4.6000000000000004e-07,
1241
+ "loss": 0.0001,
1242
+ "step": 4800
1243
+ },
1244
+ {
1245
+ "epoch": 89.0,
1246
+ "learning_rate": 4.0444444444444445e-07,
1247
+ "loss": 0.0001,
1248
+ "step": 4825
1249
+ },
1250
+ {
1251
+ "epoch": 89.01,
1252
+ "learning_rate": 3.488888888888889e-07,
1253
+ "loss": 0.0001,
1254
+ "step": 4850
1255
+ },
1256
+ {
1257
+ "epoch": 90.0,
1258
+ "learning_rate": 2.9333333333333337e-07,
1259
+ "loss": 0.0001,
1260
+ "step": 4875
1261
+ },
1262
+ {
1263
+ "epoch": 90.01,
1264
+ "learning_rate": 2.3777777777777777e-07,
1265
+ "loss": 0.0001,
1266
+ "step": 4900
1267
+ },
1268
+ {
1269
+ "epoch": 91.0,
1270
+ "learning_rate": 1.8222222222222226e-07,
1271
+ "loss": 0.0001,
1272
+ "step": 4925
1273
+ },
1274
+ {
1275
+ "epoch": 91.01,
1276
+ "learning_rate": 1.2666666666666666e-07,
1277
+ "loss": 0.0001,
1278
+ "step": 4950
1279
+ },
1280
+ {
1281
+ "epoch": 92.0,
1282
+ "learning_rate": 7.111111111111112e-08,
1283
+ "loss": 0.0001,
1284
+ "step": 4975
1285
+ },
1286
+ {
1287
+ "epoch": 92.01,
1288
+ "learning_rate": 1.5555555555555557e-08,
1289
+ "loss": 0.0001,
1290
+ "step": 5000
1291
+ },
1292
+ {
1293
+ "epoch": 92.01,
1294
+ "eval_loss": 0.7918470501899719,
1295
+ "eval_runtime": 104.4946,
1296
+ "eval_samples_per_second": 3.675,
1297
+ "eval_steps_per_second": 0.23,
1298
+ "eval_wer": 34.623217922606926,
1299
+ "step": 5000
1300
+ },
1301
+ {
1302
+ "epoch": 92.01,
1303
+ "step": 5000,
1304
+ "total_flos": 1.69853681664e+20,
1305
+ "train_loss": 0.04092751889359206,
1306
+ "train_runtime": 27870.4476,
1307
+ "train_samples_per_second": 2.87,
1308
+ "train_steps_per_second": 0.179
1309
+ }
1310
+ ],
1311
+ "logging_steps": 25,
1312
+ "max_steps": 5000,
1313
+ "num_input_tokens_seen": 0,
1314
+ "num_train_epochs": 9223372036854775807,
1315
+ "save_steps": 1000,
1316
+ "total_flos": 1.69853681664e+20,
1317
+ "train_batch_size": 16,
1318
+ "trial_name": null,
1319
+ "trial_params": null
1320
+ }