rishabhjain16 commited on
Commit
1955041
1 Parent(s): 596eada

End of training

Browse files
README.md CHANGED
@@ -2,20 +2,33 @@
2
  license: apache-2.0
3
  base_model: openai/whisper-large-v3
4
  tags:
 
5
  - generated_from_trainer
 
 
6
  metrics:
7
  - wer
8
  model-index:
9
- - name: openai/whisper-large-v3
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-v3
17
 
18
- This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.3189
21
  - Wer: 23.8109
 
2
  license: apache-2.0
3
  base_model: openai/whisper-large-v3
4
  tags:
5
+ - whisper-event
6
  - generated_from_trainer
7
+ datasets:
8
+ - rishabhjain16/kaggle_albanian
9
  metrics:
10
  - wer
11
  model-index:
12
+ - name: Whisper large V3 to Kaggle Albanian
13
+ results:
14
+ - task:
15
+ name: Automatic Speech Recognition
16
+ type: automatic-speech-recognition
17
+ dataset:
18
+ name: rishabhjain16/kaggle_albanian default
19
+ type: rishabhjain16/kaggle_albanian
20
+ metrics:
21
+ - name: Wer
22
+ type: wer
23
+ value: 23.81094676029683
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 V3 to Kaggle Albanian
30
 
31
+ This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the rishabhjain16/kaggle_albanian default dataset.
32
  It achieves the following results on the evaluation set:
33
  - Loss: 0.3189
34
  - Wer: 23.8109
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.18,
3
+ "eval_loss": 0.31885361671447754,
4
+ "eval_runtime": 2125.9841,
5
+ "eval_samples_per_second": 3.535,
6
+ "eval_steps_per_second": 0.221,
7
+ "eval_wer": 23.81094676029683,
8
+ "train_loss": 0.2540892813682556,
9
+ "train_runtime": 48626.5038,
10
+ "train_samples_per_second": 1.645,
11
+ "train_steps_per_second": 0.103
12
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.18,
3
+ "eval_loss": 0.31885361671447754,
4
+ "eval_runtime": 2125.9841,
5
+ "eval_samples_per_second": 3.535,
6
+ "eval_steps_per_second": 0.221,
7
+ "eval_wer": 23.81094676029683
8
+ }
runs/Feb20_19-32-51_6abce58e1145/events.out.tfevents.1708508512.6abce58e1145.1924.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d3b853ccc8a656a7094821673eadc9523847e6f83efbc1dab7472c1c4488955
3
+ size 358
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.18,
3
+ "train_loss": 0.2540892813682556,
4
+ "train_runtime": 48626.5038,
5
+ "train_samples_per_second": 1.645,
6
+ "train_steps_per_second": 0.103
7
+ }
trainer_state.json ADDED
@@ -0,0 +1,1320 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 23.81094676029683,
3
+ "best_model_checkpoint": "./checkpoint-5000",
4
+ "epoch": 2.1788,
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": 4.6000000000000004e-07,
14
+ "loss": 1.7155,
15
+ "step": 25
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 9.600000000000001e-07,
20
+ "loss": 1.2833,
21
+ "step": 50
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 1.46e-06,
26
+ "loss": 0.6549,
27
+ "step": 75
28
+ },
29
+ {
30
+ "epoch": 0.02,
31
+ "learning_rate": 1.9600000000000003e-06,
32
+ "loss": 0.4167,
33
+ "step": 100
34
+ },
35
+ {
36
+ "epoch": 0.03,
37
+ "learning_rate": 2.46e-06,
38
+ "loss": 0.3397,
39
+ "step": 125
40
+ },
41
+ {
42
+ "epoch": 0.03,
43
+ "learning_rate": 2.96e-06,
44
+ "loss": 0.3208,
45
+ "step": 150
46
+ },
47
+ {
48
+ "epoch": 0.04,
49
+ "learning_rate": 3.46e-06,
50
+ "loss": 0.2839,
51
+ "step": 175
52
+ },
53
+ {
54
+ "epoch": 0.04,
55
+ "learning_rate": 3.96e-06,
56
+ "loss": 0.2953,
57
+ "step": 200
58
+ },
59
+ {
60
+ "epoch": 0.04,
61
+ "learning_rate": 4.4600000000000005e-06,
62
+ "loss": 0.3841,
63
+ "step": 225
64
+ },
65
+ {
66
+ "epoch": 0.05,
67
+ "learning_rate": 4.960000000000001e-06,
68
+ "loss": 0.416,
69
+ "step": 250
70
+ },
71
+ {
72
+ "epoch": 0.06,
73
+ "learning_rate": 5.460000000000001e-06,
74
+ "loss": 0.3926,
75
+ "step": 275
76
+ },
77
+ {
78
+ "epoch": 0.06,
79
+ "learning_rate": 5.9600000000000005e-06,
80
+ "loss": 0.3496,
81
+ "step": 300
82
+ },
83
+ {
84
+ "epoch": 0.07,
85
+ "learning_rate": 6.460000000000001e-06,
86
+ "loss": 0.4149,
87
+ "step": 325
88
+ },
89
+ {
90
+ "epoch": 0.07,
91
+ "learning_rate": 6.96e-06,
92
+ "loss": 0.482,
93
+ "step": 350
94
+ },
95
+ {
96
+ "epoch": 0.07,
97
+ "learning_rate": 7.4600000000000006e-06,
98
+ "loss": 0.4613,
99
+ "step": 375
100
+ },
101
+ {
102
+ "epoch": 0.08,
103
+ "learning_rate": 7.960000000000002e-06,
104
+ "loss": 0.4522,
105
+ "step": 400
106
+ },
107
+ {
108
+ "epoch": 0.09,
109
+ "learning_rate": 8.46e-06,
110
+ "loss": 0.4463,
111
+ "step": 425
112
+ },
113
+ {
114
+ "epoch": 0.09,
115
+ "learning_rate": 8.96e-06,
116
+ "loss": 0.3219,
117
+ "step": 450
118
+ },
119
+ {
120
+ "epoch": 0.1,
121
+ "learning_rate": 9.460000000000001e-06,
122
+ "loss": 0.3347,
123
+ "step": 475
124
+ },
125
+ {
126
+ "epoch": 0.1,
127
+ "learning_rate": 9.960000000000001e-06,
128
+ "loss": 0.2778,
129
+ "step": 500
130
+ },
131
+ {
132
+ "epoch": 0.1,
133
+ "eval_loss": 0.49452078342437744,
134
+ "eval_runtime": 2170.0908,
135
+ "eval_samples_per_second": 3.463,
136
+ "eval_steps_per_second": 0.217,
137
+ "eval_wer": 36.98391354151103,
138
+ "step": 500
139
+ },
140
+ {
141
+ "epoch": 0.1,
142
+ "learning_rate": 9.94888888888889e-06,
143
+ "loss": 0.2984,
144
+ "step": 525
145
+ },
146
+ {
147
+ "epoch": 0.11,
148
+ "learning_rate": 9.893333333333334e-06,
149
+ "loss": 0.3115,
150
+ "step": 550
151
+ },
152
+ {
153
+ "epoch": 0.12,
154
+ "learning_rate": 9.837777777777778e-06,
155
+ "loss": 0.2502,
156
+ "step": 575
157
+ },
158
+ {
159
+ "epoch": 0.12,
160
+ "learning_rate": 9.782222222222222e-06,
161
+ "loss": 0.2514,
162
+ "step": 600
163
+ },
164
+ {
165
+ "epoch": 0.12,
166
+ "learning_rate": 9.726666666666668e-06,
167
+ "loss": 0.3575,
168
+ "step": 625
169
+ },
170
+ {
171
+ "epoch": 0.13,
172
+ "learning_rate": 9.671111111111112e-06,
173
+ "loss": 0.4231,
174
+ "step": 650
175
+ },
176
+ {
177
+ "epoch": 0.14,
178
+ "learning_rate": 9.615555555555558e-06,
179
+ "loss": 0.4043,
180
+ "step": 675
181
+ },
182
+ {
183
+ "epoch": 0.14,
184
+ "learning_rate": 9.56e-06,
185
+ "loss": 0.4514,
186
+ "step": 700
187
+ },
188
+ {
189
+ "epoch": 0.14,
190
+ "learning_rate": 9.504444444444446e-06,
191
+ "loss": 0.4511,
192
+ "step": 725
193
+ },
194
+ {
195
+ "epoch": 0.15,
196
+ "learning_rate": 9.44888888888889e-06,
197
+ "loss": 0.4803,
198
+ "step": 750
199
+ },
200
+ {
201
+ "epoch": 0.15,
202
+ "learning_rate": 9.393333333333334e-06,
203
+ "loss": 0.4437,
204
+ "step": 775
205
+ },
206
+ {
207
+ "epoch": 0.16,
208
+ "learning_rate": 9.33777777777778e-06,
209
+ "loss": 0.4146,
210
+ "step": 800
211
+ },
212
+ {
213
+ "epoch": 0.17,
214
+ "learning_rate": 9.282222222222222e-06,
215
+ "loss": 0.4221,
216
+ "step": 825
217
+ },
218
+ {
219
+ "epoch": 0.17,
220
+ "learning_rate": 9.226666666666668e-06,
221
+ "loss": 0.424,
222
+ "step": 850
223
+ },
224
+ {
225
+ "epoch": 0.17,
226
+ "learning_rate": 9.171111111111112e-06,
227
+ "loss": 0.4471,
228
+ "step": 875
229
+ },
230
+ {
231
+ "epoch": 0.18,
232
+ "learning_rate": 9.115555555555556e-06,
233
+ "loss": 0.4358,
234
+ "step": 900
235
+ },
236
+ {
237
+ "epoch": 0.18,
238
+ "learning_rate": 9.060000000000001e-06,
239
+ "loss": 0.4,
240
+ "step": 925
241
+ },
242
+ {
243
+ "epoch": 0.19,
244
+ "learning_rate": 9.004444444444445e-06,
245
+ "loss": 0.4205,
246
+ "step": 950
247
+ },
248
+ {
249
+ "epoch": 0.2,
250
+ "learning_rate": 8.94888888888889e-06,
251
+ "loss": 0.4146,
252
+ "step": 975
253
+ },
254
+ {
255
+ "epoch": 0.2,
256
+ "learning_rate": 8.893333333333333e-06,
257
+ "loss": 0.4343,
258
+ "step": 1000
259
+ },
260
+ {
261
+ "epoch": 0.2,
262
+ "eval_loss": 0.42902088165283203,
263
+ "eval_runtime": 2158.1828,
264
+ "eval_samples_per_second": 3.483,
265
+ "eval_steps_per_second": 0.218,
266
+ "eval_wer": 35.183525075949994,
267
+ "step": 1000
268
+ },
269
+ {
270
+ "epoch": 0.2,
271
+ "learning_rate": 8.83777777777778e-06,
272
+ "loss": 0.418,
273
+ "step": 1025
274
+ },
275
+ {
276
+ "epoch": 0.21,
277
+ "learning_rate": 8.782222222222223e-06,
278
+ "loss": 0.4076,
279
+ "step": 1050
280
+ },
281
+ {
282
+ "epoch": 0.21,
283
+ "learning_rate": 8.726666666666667e-06,
284
+ "loss": 0.4059,
285
+ "step": 1075
286
+ },
287
+ {
288
+ "epoch": 0.22,
289
+ "learning_rate": 8.671111111111113e-06,
290
+ "loss": 0.3988,
291
+ "step": 1100
292
+ },
293
+ {
294
+ "epoch": 0.23,
295
+ "learning_rate": 8.615555555555555e-06,
296
+ "loss": 0.3707,
297
+ "step": 1125
298
+ },
299
+ {
300
+ "epoch": 0.23,
301
+ "learning_rate": 8.560000000000001e-06,
302
+ "loss": 0.3797,
303
+ "step": 1150
304
+ },
305
+ {
306
+ "epoch": 0.23,
307
+ "learning_rate": 8.504444444444445e-06,
308
+ "loss": 0.3647,
309
+ "step": 1175
310
+ },
311
+ {
312
+ "epoch": 0.24,
313
+ "learning_rate": 8.448888888888889e-06,
314
+ "loss": 0.3669,
315
+ "step": 1200
316
+ },
317
+ {
318
+ "epoch": 0.24,
319
+ "learning_rate": 8.393333333333335e-06,
320
+ "loss": 0.3469,
321
+ "step": 1225
322
+ },
323
+ {
324
+ "epoch": 0.25,
325
+ "learning_rate": 8.337777777777777e-06,
326
+ "loss": 0.336,
327
+ "step": 1250
328
+ },
329
+ {
330
+ "epoch": 0.26,
331
+ "learning_rate": 8.282222222222223e-06,
332
+ "loss": 0.34,
333
+ "step": 1275
334
+ },
335
+ {
336
+ "epoch": 0.26,
337
+ "learning_rate": 8.226666666666667e-06,
338
+ "loss": 0.3328,
339
+ "step": 1300
340
+ },
341
+ {
342
+ "epoch": 0.27,
343
+ "learning_rate": 8.171111111111113e-06,
344
+ "loss": 0.3497,
345
+ "step": 1325
346
+ },
347
+ {
348
+ "epoch": 0.27,
349
+ "learning_rate": 8.115555555555557e-06,
350
+ "loss": 0.3747,
351
+ "step": 1350
352
+ },
353
+ {
354
+ "epoch": 0.28,
355
+ "learning_rate": 8.06e-06,
356
+ "loss": 0.3966,
357
+ "step": 1375
358
+ },
359
+ {
360
+ "epoch": 0.28,
361
+ "learning_rate": 8.004444444444445e-06,
362
+ "loss": 0.3475,
363
+ "step": 1400
364
+ },
365
+ {
366
+ "epoch": 0.28,
367
+ "learning_rate": 7.948888888888889e-06,
368
+ "loss": 0.3441,
369
+ "step": 1425
370
+ },
371
+ {
372
+ "epoch": 0.29,
373
+ "learning_rate": 7.893333333333335e-06,
374
+ "loss": 0.3357,
375
+ "step": 1450
376
+ },
377
+ {
378
+ "epoch": 0.29,
379
+ "learning_rate": 7.837777777777779e-06,
380
+ "loss": 0.378,
381
+ "step": 1475
382
+ },
383
+ {
384
+ "epoch": 0.3,
385
+ "learning_rate": 7.782222222222223e-06,
386
+ "loss": 0.3523,
387
+ "step": 1500
388
+ },
389
+ {
390
+ "epoch": 0.3,
391
+ "eval_loss": 0.3780156672000885,
392
+ "eval_runtime": 2149.1405,
393
+ "eval_samples_per_second": 3.497,
394
+ "eval_steps_per_second": 0.219,
395
+ "eval_wer": 30.651426863887643,
396
+ "step": 1500
397
+ },
398
+ {
399
+ "epoch": 0.3,
400
+ "learning_rate": 7.726666666666667e-06,
401
+ "loss": 0.3923,
402
+ "step": 1525
403
+ },
404
+ {
405
+ "epoch": 0.31,
406
+ "learning_rate": 7.67111111111111e-06,
407
+ "loss": 0.3449,
408
+ "step": 1550
409
+ },
410
+ {
411
+ "epoch": 0.32,
412
+ "learning_rate": 7.6155555555555564e-06,
413
+ "loss": 0.3714,
414
+ "step": 1575
415
+ },
416
+ {
417
+ "epoch": 0.32,
418
+ "learning_rate": 7.5600000000000005e-06,
419
+ "loss": 0.4006,
420
+ "step": 1600
421
+ },
422
+ {
423
+ "epoch": 0.33,
424
+ "learning_rate": 7.504444444444445e-06,
425
+ "loss": 0.3612,
426
+ "step": 1625
427
+ },
428
+ {
429
+ "epoch": 0.33,
430
+ "learning_rate": 7.44888888888889e-06,
431
+ "loss": 0.3639,
432
+ "step": 1650
433
+ },
434
+ {
435
+ "epoch": 0.34,
436
+ "learning_rate": 7.393333333333333e-06,
437
+ "loss": 0.4153,
438
+ "step": 1675
439
+ },
440
+ {
441
+ "epoch": 0.34,
442
+ "learning_rate": 7.337777777777778e-06,
443
+ "loss": 0.3818,
444
+ "step": 1700
445
+ },
446
+ {
447
+ "epoch": 0.34,
448
+ "learning_rate": 7.282222222222222e-06,
449
+ "loss": 0.3564,
450
+ "step": 1725
451
+ },
452
+ {
453
+ "epoch": 0.35,
454
+ "learning_rate": 7.226666666666667e-06,
455
+ "loss": 0.3629,
456
+ "step": 1750
457
+ },
458
+ {
459
+ "epoch": 0.35,
460
+ "learning_rate": 7.171111111111112e-06,
461
+ "loss": 0.2898,
462
+ "step": 1775
463
+ },
464
+ {
465
+ "epoch": 0.36,
466
+ "learning_rate": 7.115555555555557e-06,
467
+ "loss": 0.3372,
468
+ "step": 1800
469
+ },
470
+ {
471
+ "epoch": 0.36,
472
+ "learning_rate": 7.06e-06,
473
+ "loss": 0.3539,
474
+ "step": 1825
475
+ },
476
+ {
477
+ "epoch": 0.37,
478
+ "learning_rate": 7.004444444444445e-06,
479
+ "loss": 0.3791,
480
+ "step": 1850
481
+ },
482
+ {
483
+ "epoch": 0.38,
484
+ "learning_rate": 6.948888888888889e-06,
485
+ "loss": 0.3461,
486
+ "step": 1875
487
+ },
488
+ {
489
+ "epoch": 0.38,
490
+ "learning_rate": 6.893333333333334e-06,
491
+ "loss": 0.3498,
492
+ "step": 1900
493
+ },
494
+ {
495
+ "epoch": 0.39,
496
+ "learning_rate": 6.837777777777779e-06,
497
+ "loss": 0.3502,
498
+ "step": 1925
499
+ },
500
+ {
501
+ "epoch": 0.39,
502
+ "learning_rate": 6.782222222222222e-06,
503
+ "loss": 0.3696,
504
+ "step": 1950
505
+ },
506
+ {
507
+ "epoch": 0.4,
508
+ "learning_rate": 6.726666666666667e-06,
509
+ "loss": 0.3524,
510
+ "step": 1975
511
+ },
512
+ {
513
+ "epoch": 0.4,
514
+ "learning_rate": 6.671111111111112e-06,
515
+ "loss": 0.3833,
516
+ "step": 2000
517
+ },
518
+ {
519
+ "epoch": 0.4,
520
+ "eval_loss": 0.343627005815506,
521
+ "eval_runtime": 2165.5496,
522
+ "eval_samples_per_second": 3.471,
523
+ "eval_steps_per_second": 0.217,
524
+ "eval_wer": 28.031774490761492,
525
+ "step": 2000
526
+ },
527
+ {
528
+ "epoch": 0.41,
529
+ "learning_rate": 6.615555555555556e-06,
530
+ "loss": 0.3653,
531
+ "step": 2025
532
+ },
533
+ {
534
+ "epoch": 0.41,
535
+ "learning_rate": 6.560000000000001e-06,
536
+ "loss": 0.3649,
537
+ "step": 2050
538
+ },
539
+ {
540
+ "epoch": 1.0,
541
+ "learning_rate": 6.504444444444446e-06,
542
+ "loss": 0.3044,
543
+ "step": 2075
544
+ },
545
+ {
546
+ "epoch": 1.01,
547
+ "learning_rate": 6.448888888888889e-06,
548
+ "loss": 0.2725,
549
+ "step": 2100
550
+ },
551
+ {
552
+ "epoch": 1.01,
553
+ "learning_rate": 6.393333333333334e-06,
554
+ "loss": 0.2116,
555
+ "step": 2125
556
+ },
557
+ {
558
+ "epoch": 1.02,
559
+ "learning_rate": 6.3377777777777786e-06,
560
+ "loss": 0.1644,
561
+ "step": 2150
562
+ },
563
+ {
564
+ "epoch": 1.02,
565
+ "learning_rate": 6.282222222222223e-06,
566
+ "loss": 0.1579,
567
+ "step": 2175
568
+ },
569
+ {
570
+ "epoch": 1.03,
571
+ "learning_rate": 6.2266666666666675e-06,
572
+ "loss": 0.1648,
573
+ "step": 2200
574
+ },
575
+ {
576
+ "epoch": 1.03,
577
+ "learning_rate": 6.171111111111112e-06,
578
+ "loss": 0.1634,
579
+ "step": 2225
580
+ },
581
+ {
582
+ "epoch": 1.04,
583
+ "learning_rate": 6.1155555555555555e-06,
584
+ "loss": 0.1412,
585
+ "step": 2250
586
+ },
587
+ {
588
+ "epoch": 1.04,
589
+ "learning_rate": 6.0600000000000004e-06,
590
+ "loss": 0.2006,
591
+ "step": 2275
592
+ },
593
+ {
594
+ "epoch": 1.05,
595
+ "learning_rate": 6.004444444444445e-06,
596
+ "loss": 0.2172,
597
+ "step": 2300
598
+ },
599
+ {
600
+ "epoch": 1.05,
601
+ "learning_rate": 5.948888888888889e-06,
602
+ "loss": 0.1989,
603
+ "step": 2325
604
+ },
605
+ {
606
+ "epoch": 1.06,
607
+ "learning_rate": 5.893333333333334e-06,
608
+ "loss": 0.2036,
609
+ "step": 2350
610
+ },
611
+ {
612
+ "epoch": 1.06,
613
+ "learning_rate": 5.837777777777777e-06,
614
+ "loss": 0.1612,
615
+ "step": 2375
616
+ },
617
+ {
618
+ "epoch": 1.07,
619
+ "learning_rate": 5.782222222222222e-06,
620
+ "loss": 0.192,
621
+ "step": 2400
622
+ },
623
+ {
624
+ "epoch": 1.07,
625
+ "learning_rate": 5.726666666666667e-06,
626
+ "loss": 0.1787,
627
+ "step": 2425
628
+ },
629
+ {
630
+ "epoch": 1.08,
631
+ "learning_rate": 5.671111111111112e-06,
632
+ "loss": 0.1886,
633
+ "step": 2450
634
+ },
635
+ {
636
+ "epoch": 1.08,
637
+ "learning_rate": 5.615555555555556e-06,
638
+ "loss": 0.1996,
639
+ "step": 2475
640
+ },
641
+ {
642
+ "epoch": 1.09,
643
+ "learning_rate": 5.560000000000001e-06,
644
+ "loss": 0.2029,
645
+ "step": 2500
646
+ },
647
+ {
648
+ "epoch": 1.09,
649
+ "eval_loss": 0.33604511618614197,
650
+ "eval_runtime": 2142.8094,
651
+ "eval_samples_per_second": 3.508,
652
+ "eval_steps_per_second": 0.219,
653
+ "eval_wer": 26.51277454056477,
654
+ "step": 2500
655
+ },
656
+ {
657
+ "epoch": 1.09,
658
+ "learning_rate": 5.504444444444444e-06,
659
+ "loss": 0.2064,
660
+ "step": 2525
661
+ },
662
+ {
663
+ "epoch": 1.1,
664
+ "learning_rate": 5.448888888888889e-06,
665
+ "loss": 0.1988,
666
+ "step": 2550
667
+ },
668
+ {
669
+ "epoch": 1.1,
670
+ "learning_rate": 5.393333333333334e-06,
671
+ "loss": 0.1774,
672
+ "step": 2575
673
+ },
674
+ {
675
+ "epoch": 1.11,
676
+ "learning_rate": 5.337777777777779e-06,
677
+ "loss": 0.2277,
678
+ "step": 2600
679
+ },
680
+ {
681
+ "epoch": 1.11,
682
+ "learning_rate": 5.282222222222223e-06,
683
+ "loss": 0.1804,
684
+ "step": 2625
685
+ },
686
+ {
687
+ "epoch": 1.12,
688
+ "learning_rate": 5.226666666666667e-06,
689
+ "loss": 0.2029,
690
+ "step": 2650
691
+ },
692
+ {
693
+ "epoch": 1.12,
694
+ "learning_rate": 5.171111111111111e-06,
695
+ "loss": 0.169,
696
+ "step": 2675
697
+ },
698
+ {
699
+ "epoch": 1.13,
700
+ "learning_rate": 5.115555555555556e-06,
701
+ "loss": 0.1985,
702
+ "step": 2700
703
+ },
704
+ {
705
+ "epoch": 1.13,
706
+ "learning_rate": 5.060000000000001e-06,
707
+ "loss": 0.2063,
708
+ "step": 2725
709
+ },
710
+ {
711
+ "epoch": 1.14,
712
+ "learning_rate": 5.004444444444445e-06,
713
+ "loss": 0.2117,
714
+ "step": 2750
715
+ },
716
+ {
717
+ "epoch": 1.14,
718
+ "learning_rate": 4.94888888888889e-06,
719
+ "loss": 0.2109,
720
+ "step": 2775
721
+ },
722
+ {
723
+ "epoch": 1.15,
724
+ "learning_rate": 4.893333333333334e-06,
725
+ "loss": 0.163,
726
+ "step": 2800
727
+ },
728
+ {
729
+ "epoch": 1.15,
730
+ "learning_rate": 4.837777777777778e-06,
731
+ "loss": 0.1834,
732
+ "step": 2825
733
+ },
734
+ {
735
+ "epoch": 1.16,
736
+ "learning_rate": 4.7822222222222226e-06,
737
+ "loss": 0.196,
738
+ "step": 2850
739
+ },
740
+ {
741
+ "epoch": 1.16,
742
+ "learning_rate": 4.7266666666666674e-06,
743
+ "loss": 0.1746,
744
+ "step": 2875
745
+ },
746
+ {
747
+ "epoch": 1.17,
748
+ "learning_rate": 4.6711111111111115e-06,
749
+ "loss": 0.1844,
750
+ "step": 2900
751
+ },
752
+ {
753
+ "epoch": 1.17,
754
+ "learning_rate": 4.6155555555555555e-06,
755
+ "loss": 0.1548,
756
+ "step": 2925
757
+ },
758
+ {
759
+ "epoch": 1.18,
760
+ "learning_rate": 4.56e-06,
761
+ "loss": 0.1546,
762
+ "step": 2950
763
+ },
764
+ {
765
+ "epoch": 1.18,
766
+ "learning_rate": 4.504444444444444e-06,
767
+ "loss": 0.1539,
768
+ "step": 2975
769
+ },
770
+ {
771
+ "epoch": 1.19,
772
+ "learning_rate": 4.448888888888889e-06,
773
+ "loss": 0.1797,
774
+ "step": 3000
775
+ },
776
+ {
777
+ "epoch": 1.19,
778
+ "eval_loss": 0.3350614607334137,
779
+ "eval_runtime": 2157.9989,
780
+ "eval_samples_per_second": 3.483,
781
+ "eval_steps_per_second": 0.218,
782
+ "eval_wer": 25.892723741222174,
783
+ "step": 3000
784
+ },
785
+ {
786
+ "epoch": 1.19,
787
+ "learning_rate": 4.393333333333334e-06,
788
+ "loss": 0.1737,
789
+ "step": 3025
790
+ },
791
+ {
792
+ "epoch": 1.2,
793
+ "learning_rate": 4.337777777777778e-06,
794
+ "loss": 0.1736,
795
+ "step": 3050
796
+ },
797
+ {
798
+ "epoch": 1.2,
799
+ "learning_rate": 4.282222222222222e-06,
800
+ "loss": 0.2304,
801
+ "step": 3075
802
+ },
803
+ {
804
+ "epoch": 1.21,
805
+ "learning_rate": 4.226666666666667e-06,
806
+ "loss": 0.2147,
807
+ "step": 3100
808
+ },
809
+ {
810
+ "epoch": 1.21,
811
+ "learning_rate": 4.171111111111111e-06,
812
+ "loss": 0.2228,
813
+ "step": 3125
814
+ },
815
+ {
816
+ "epoch": 1.22,
817
+ "learning_rate": 4.115555555555556e-06,
818
+ "loss": 0.238,
819
+ "step": 3150
820
+ },
821
+ {
822
+ "epoch": 1.22,
823
+ "learning_rate": 4.060000000000001e-06,
824
+ "loss": 0.1769,
825
+ "step": 3175
826
+ },
827
+ {
828
+ "epoch": 1.23,
829
+ "learning_rate": 4.004444444444445e-06,
830
+ "loss": 0.1293,
831
+ "step": 3200
832
+ },
833
+ {
834
+ "epoch": 1.23,
835
+ "learning_rate": 3.948888888888889e-06,
836
+ "loss": 0.1206,
837
+ "step": 3225
838
+ },
839
+ {
840
+ "epoch": 1.24,
841
+ "learning_rate": 3.893333333333333e-06,
842
+ "loss": 0.1162,
843
+ "step": 3250
844
+ },
845
+ {
846
+ "epoch": 1.24,
847
+ "learning_rate": 3.837777777777778e-06,
848
+ "loss": 0.1242,
849
+ "step": 3275
850
+ },
851
+ {
852
+ "epoch": 1.25,
853
+ "learning_rate": 3.782222222222223e-06,
854
+ "loss": 0.1193,
855
+ "step": 3300
856
+ },
857
+ {
858
+ "epoch": 1.25,
859
+ "learning_rate": 3.726666666666667e-06,
860
+ "loss": 0.0961,
861
+ "step": 3325
862
+ },
863
+ {
864
+ "epoch": 1.26,
865
+ "learning_rate": 3.6711111111111113e-06,
866
+ "loss": 0.1158,
867
+ "step": 3350
868
+ },
869
+ {
870
+ "epoch": 1.26,
871
+ "learning_rate": 3.615555555555556e-06,
872
+ "loss": 0.1642,
873
+ "step": 3375
874
+ },
875
+ {
876
+ "epoch": 1.27,
877
+ "learning_rate": 3.5600000000000002e-06,
878
+ "loss": 0.1731,
879
+ "step": 3400
880
+ },
881
+ {
882
+ "epoch": 1.27,
883
+ "learning_rate": 3.5044444444444447e-06,
884
+ "loss": 0.1629,
885
+ "step": 3425
886
+ },
887
+ {
888
+ "epoch": 1.28,
889
+ "learning_rate": 3.4488888888888896e-06,
890
+ "loss": 0.1607,
891
+ "step": 3450
892
+ },
893
+ {
894
+ "epoch": 1.28,
895
+ "learning_rate": 3.3933333333333336e-06,
896
+ "loss": 0.1553,
897
+ "step": 3475
898
+ },
899
+ {
900
+ "epoch": 1.29,
901
+ "learning_rate": 3.337777777777778e-06,
902
+ "loss": 0.1687,
903
+ "step": 3500
904
+ },
905
+ {
906
+ "epoch": 1.29,
907
+ "eval_loss": 0.32494643330574036,
908
+ "eval_runtime": 2166.1529,
909
+ "eval_samples_per_second": 3.47,
910
+ "eval_steps_per_second": 0.217,
911
+ "eval_wer": 25.862841774988794,
912
+ "step": 3500
913
+ },
914
+ {
915
+ "epoch": 1.29,
916
+ "learning_rate": 3.282222222222223e-06,
917
+ "loss": 0.1774,
918
+ "step": 3525
919
+ },
920
+ {
921
+ "epoch": 1.3,
922
+ "learning_rate": 3.226666666666667e-06,
923
+ "loss": 0.1959,
924
+ "step": 3550
925
+ },
926
+ {
927
+ "epoch": 1.3,
928
+ "learning_rate": 3.1711111111111114e-06,
929
+ "loss": 0.163,
930
+ "step": 3575
931
+ },
932
+ {
933
+ "epoch": 1.31,
934
+ "learning_rate": 3.1155555555555555e-06,
935
+ "loss": 0.1777,
936
+ "step": 3600
937
+ },
938
+ {
939
+ "epoch": 1.31,
940
+ "learning_rate": 3.0600000000000003e-06,
941
+ "loss": 0.1728,
942
+ "step": 3625
943
+ },
944
+ {
945
+ "epoch": 1.32,
946
+ "learning_rate": 3.004444444444445e-06,
947
+ "loss": 0.1662,
948
+ "step": 3650
949
+ },
950
+ {
951
+ "epoch": 1.32,
952
+ "learning_rate": 2.948888888888889e-06,
953
+ "loss": 0.1561,
954
+ "step": 3675
955
+ },
956
+ {
957
+ "epoch": 1.33,
958
+ "learning_rate": 2.8933333333333337e-06,
959
+ "loss": 0.1787,
960
+ "step": 3700
961
+ },
962
+ {
963
+ "epoch": 1.33,
964
+ "learning_rate": 2.837777777777778e-06,
965
+ "loss": 0.1754,
966
+ "step": 3725
967
+ },
968
+ {
969
+ "epoch": 1.34,
970
+ "learning_rate": 2.7822222222222222e-06,
971
+ "loss": 0.1773,
972
+ "step": 3750
973
+ },
974
+ {
975
+ "epoch": 1.34,
976
+ "learning_rate": 2.726666666666667e-06,
977
+ "loss": 0.1892,
978
+ "step": 3775
979
+ },
980
+ {
981
+ "epoch": 1.35,
982
+ "learning_rate": 2.6711111111111116e-06,
983
+ "loss": 0.1488,
984
+ "step": 3800
985
+ },
986
+ {
987
+ "epoch": 1.35,
988
+ "learning_rate": 2.6155555555555556e-06,
989
+ "loss": 0.1608,
990
+ "step": 3825
991
+ },
992
+ {
993
+ "epoch": 1.36,
994
+ "learning_rate": 2.56e-06,
995
+ "loss": 0.1294,
996
+ "step": 3850
997
+ },
998
+ {
999
+ "epoch": 1.36,
1000
+ "learning_rate": 2.504444444444445e-06,
1001
+ "loss": 0.1756,
1002
+ "step": 3875
1003
+ },
1004
+ {
1005
+ "epoch": 1.37,
1006
+ "learning_rate": 2.448888888888889e-06,
1007
+ "loss": 0.1413,
1008
+ "step": 3900
1009
+ },
1010
+ {
1011
+ "epoch": 1.37,
1012
+ "learning_rate": 2.3933333333333334e-06,
1013
+ "loss": 0.1489,
1014
+ "step": 3925
1015
+ },
1016
+ {
1017
+ "epoch": 1.38,
1018
+ "learning_rate": 2.337777777777778e-06,
1019
+ "loss": 0.1581,
1020
+ "step": 3950
1021
+ },
1022
+ {
1023
+ "epoch": 1.38,
1024
+ "learning_rate": 2.2822222222222223e-06,
1025
+ "loss": 0.1448,
1026
+ "step": 3975
1027
+ },
1028
+ {
1029
+ "epoch": 1.39,
1030
+ "learning_rate": 2.226666666666667e-06,
1031
+ "loss": 0.1603,
1032
+ "step": 4000
1033
+ },
1034
+ {
1035
+ "epoch": 1.39,
1036
+ "eval_loss": 0.3088971972465515,
1037
+ "eval_runtime": 2143.1525,
1038
+ "eval_samples_per_second": 3.507,
1039
+ "eval_steps_per_second": 0.219,
1040
+ "eval_wer": 23.987748393844317,
1041
+ "step": 4000
1042
+ },
1043
+ {
1044
+ "epoch": 1.39,
1045
+ "learning_rate": 2.1711111111111113e-06,
1046
+ "loss": 0.1934,
1047
+ "step": 4025
1048
+ },
1049
+ {
1050
+ "epoch": 1.4,
1051
+ "learning_rate": 2.1155555555555557e-06,
1052
+ "loss": 0.1749,
1053
+ "step": 4050
1054
+ },
1055
+ {
1056
+ "epoch": 1.4,
1057
+ "learning_rate": 2.06e-06,
1058
+ "loss": 0.1772,
1059
+ "step": 4075
1060
+ },
1061
+ {
1062
+ "epoch": 1.41,
1063
+ "learning_rate": 2.0044444444444446e-06,
1064
+ "loss": 0.2196,
1065
+ "step": 4100
1066
+ },
1067
+ {
1068
+ "epoch": 2.0,
1069
+ "learning_rate": 1.948888888888889e-06,
1070
+ "loss": 0.1208,
1071
+ "step": 4125
1072
+ },
1073
+ {
1074
+ "epoch": 2.01,
1075
+ "learning_rate": 1.8933333333333333e-06,
1076
+ "loss": 0.0793,
1077
+ "step": 4150
1078
+ },
1079
+ {
1080
+ "epoch": 2.01,
1081
+ "learning_rate": 1.837777777777778e-06,
1082
+ "loss": 0.0843,
1083
+ "step": 4175
1084
+ },
1085
+ {
1086
+ "epoch": 2.02,
1087
+ "learning_rate": 1.7822222222222225e-06,
1088
+ "loss": 0.0769,
1089
+ "step": 4200
1090
+ },
1091
+ {
1092
+ "epoch": 2.02,
1093
+ "learning_rate": 1.7266666666666667e-06,
1094
+ "loss": 0.0777,
1095
+ "step": 4225
1096
+ },
1097
+ {
1098
+ "epoch": 2.03,
1099
+ "learning_rate": 1.6711111111111112e-06,
1100
+ "loss": 0.086,
1101
+ "step": 4250
1102
+ },
1103
+ {
1104
+ "epoch": 2.03,
1105
+ "learning_rate": 1.6155555555555559e-06,
1106
+ "loss": 0.0766,
1107
+ "step": 4275
1108
+ },
1109
+ {
1110
+ "epoch": 2.04,
1111
+ "learning_rate": 1.56e-06,
1112
+ "loss": 0.0885,
1113
+ "step": 4300
1114
+ },
1115
+ {
1116
+ "epoch": 2.04,
1117
+ "learning_rate": 1.5044444444444446e-06,
1118
+ "loss": 0.0778,
1119
+ "step": 4325
1120
+ },
1121
+ {
1122
+ "epoch": 2.05,
1123
+ "learning_rate": 1.4488888888888892e-06,
1124
+ "loss": 0.088,
1125
+ "step": 4350
1126
+ },
1127
+ {
1128
+ "epoch": 2.05,
1129
+ "learning_rate": 1.3933333333333335e-06,
1130
+ "loss": 0.0715,
1131
+ "step": 4375
1132
+ },
1133
+ {
1134
+ "epoch": 2.06,
1135
+ "learning_rate": 1.337777777777778e-06,
1136
+ "loss": 0.0683,
1137
+ "step": 4400
1138
+ },
1139
+ {
1140
+ "epoch": 2.06,
1141
+ "learning_rate": 1.2822222222222222e-06,
1142
+ "loss": 0.0714,
1143
+ "step": 4425
1144
+ },
1145
+ {
1146
+ "epoch": 2.07,
1147
+ "learning_rate": 1.2266666666666666e-06,
1148
+ "loss": 0.0788,
1149
+ "step": 4450
1150
+ },
1151
+ {
1152
+ "epoch": 2.07,
1153
+ "learning_rate": 1.171111111111111e-06,
1154
+ "loss": 0.0792,
1155
+ "step": 4475
1156
+ },
1157
+ {
1158
+ "epoch": 2.08,
1159
+ "learning_rate": 1.1155555555555558e-06,
1160
+ "loss": 0.0762,
1161
+ "step": 4500
1162
+ },
1163
+ {
1164
+ "epoch": 2.08,
1165
+ "eval_loss": 0.3345198631286621,
1166
+ "eval_runtime": 2127.0754,
1167
+ "eval_samples_per_second": 3.533,
1168
+ "eval_steps_per_second": 0.221,
1169
+ "eval_wer": 24.580407390806318,
1170
+ "step": 4500
1171
+ },
1172
+ {
1173
+ "epoch": 2.08,
1174
+ "learning_rate": 1.06e-06,
1175
+ "loss": 0.0786,
1176
+ "step": 4525
1177
+ },
1178
+ {
1179
+ "epoch": 2.09,
1180
+ "learning_rate": 1.0044444444444445e-06,
1181
+ "loss": 0.0932,
1182
+ "step": 4550
1183
+ },
1184
+ {
1185
+ "epoch": 2.09,
1186
+ "learning_rate": 9.488888888888889e-07,
1187
+ "loss": 0.102,
1188
+ "step": 4575
1189
+ },
1190
+ {
1191
+ "epoch": 2.1,
1192
+ "learning_rate": 8.933333333333334e-07,
1193
+ "loss": 0.0689,
1194
+ "step": 4600
1195
+ },
1196
+ {
1197
+ "epoch": 2.1,
1198
+ "learning_rate": 8.37777777777778e-07,
1199
+ "loss": 0.0841,
1200
+ "step": 4625
1201
+ },
1202
+ {
1203
+ "epoch": 2.11,
1204
+ "learning_rate": 7.822222222222223e-07,
1205
+ "loss": 0.077,
1206
+ "step": 4650
1207
+ },
1208
+ {
1209
+ "epoch": 2.11,
1210
+ "learning_rate": 7.266666666666668e-07,
1211
+ "loss": 0.0685,
1212
+ "step": 4675
1213
+ },
1214
+ {
1215
+ "epoch": 2.12,
1216
+ "learning_rate": 6.711111111111111e-07,
1217
+ "loss": 0.0774,
1218
+ "step": 4700
1219
+ },
1220
+ {
1221
+ "epoch": 2.12,
1222
+ "learning_rate": 6.155555555555556e-07,
1223
+ "loss": 0.0784,
1224
+ "step": 4725
1225
+ },
1226
+ {
1227
+ "epoch": 2.13,
1228
+ "learning_rate": 5.6e-07,
1229
+ "loss": 0.0741,
1230
+ "step": 4750
1231
+ },
1232
+ {
1233
+ "epoch": 2.13,
1234
+ "learning_rate": 5.044444444444445e-07,
1235
+ "loss": 0.0779,
1236
+ "step": 4775
1237
+ },
1238
+ {
1239
+ "epoch": 2.14,
1240
+ "learning_rate": 4.488888888888889e-07,
1241
+ "loss": 0.1124,
1242
+ "step": 4800
1243
+ },
1244
+ {
1245
+ "epoch": 2.14,
1246
+ "learning_rate": 3.9333333333333336e-07,
1247
+ "loss": 0.1224,
1248
+ "step": 4825
1249
+ },
1250
+ {
1251
+ "epoch": 2.15,
1252
+ "learning_rate": 3.3777777777777777e-07,
1253
+ "loss": 0.1299,
1254
+ "step": 4850
1255
+ },
1256
+ {
1257
+ "epoch": 2.15,
1258
+ "learning_rate": 2.822222222222222e-07,
1259
+ "loss": 0.0879,
1260
+ "step": 4875
1261
+ },
1262
+ {
1263
+ "epoch": 2.16,
1264
+ "learning_rate": 2.266666666666667e-07,
1265
+ "loss": 0.0581,
1266
+ "step": 4900
1267
+ },
1268
+ {
1269
+ "epoch": 2.16,
1270
+ "learning_rate": 1.7111111111111114e-07,
1271
+ "loss": 0.0656,
1272
+ "step": 4925
1273
+ },
1274
+ {
1275
+ "epoch": 2.17,
1276
+ "learning_rate": 1.1555555555555556e-07,
1277
+ "loss": 0.0622,
1278
+ "step": 4950
1279
+ },
1280
+ {
1281
+ "epoch": 2.17,
1282
+ "learning_rate": 6.000000000000001e-08,
1283
+ "loss": 0.0549,
1284
+ "step": 4975
1285
+ },
1286
+ {
1287
+ "epoch": 2.18,
1288
+ "learning_rate": 4.444444444444445e-09,
1289
+ "loss": 0.0676,
1290
+ "step": 5000
1291
+ },
1292
+ {
1293
+ "epoch": 2.18,
1294
+ "eval_loss": 0.31885361671447754,
1295
+ "eval_runtime": 2126.4706,
1296
+ "eval_samples_per_second": 3.534,
1297
+ "eval_steps_per_second": 0.221,
1298
+ "eval_wer": 23.81094676029683,
1299
+ "step": 5000
1300
+ },
1301
+ {
1302
+ "epoch": 2.18,
1303
+ "step": 5000,
1304
+ "total_flos": 2.717998645248e+20,
1305
+ "train_loss": 0.2540892813682556,
1306
+ "train_runtime": 48626.5038,
1307
+ "train_samples_per_second": 1.645,
1308
+ "train_steps_per_second": 0.103
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": 2.717998645248e+20,
1317
+ "train_batch_size": 16,
1318
+ "trial_name": null,
1319
+ "trial_params": null
1320
+ }