Hugging Face's logo

marinone94
/
xls-r-300m-sv-robust

Automatic Speech Recognition
Transformers
PyTorch
Swedish
wav2vec2
mozilla-foundation/common_voice_9_0
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
1
marinone94 commited on
Commit
db1f503
•
1 Parent(s): cbb2361

force stop training

Browse files
Files changed (29) hide show
  1. {checkpoint-600 → checkpoint-2900}/config.json +0 -0
  2. {checkpoint-600 → checkpoint-2900}/optimizer.pt +1 -1
  3. {checkpoint-600 → checkpoint-2900}/preprocessor_config.json +0 -0
  4. {checkpoint-700 → checkpoint-2900}/pytorch_model.bin +1 -1
  5. {checkpoint-800 → checkpoint-2900}/rng_state.pth +1 -1
  6. {checkpoint-800 → checkpoint-2900}/scaler.pt +1 -1
  7. {checkpoint-600 → checkpoint-2900}/scheduler.pt +1 -1
  8. checkpoint-2900/trainer_state.json +1147 -0
  9. {checkpoint-600 → checkpoint-2900}/training_args.bin +0 -0
  10. {checkpoint-700 → checkpoint-3000}/config.json +0 -0
  11. {checkpoint-700 → checkpoint-3000}/optimizer.pt +1 -1
  12. {checkpoint-700 → checkpoint-3000}/preprocessor_config.json +0 -0
  13. {checkpoint-600 → checkpoint-3000}/pytorch_model.bin +1 -1
  14. {checkpoint-700 → checkpoint-3000}/rng_state.pth +1 -1
  15. {checkpoint-700 → checkpoint-3000}/scaler.pt +1 -1
  16. {checkpoint-800 → checkpoint-3000}/scheduler.pt +1 -1
  17. checkpoint-3000/trainer_state.json +1186 -0
  18. {checkpoint-700 → checkpoint-3000}/training_args.bin +0 -0
  19. checkpoint-600/rng_state.pth +0 -3
  20. checkpoint-600/scaler.pt +0 -3
  21. checkpoint-600/trainer_state.json +0 -250
  22. checkpoint-700/scheduler.pt +0 -3
  23. checkpoint-700/trainer_state.json +0 -289
  24. checkpoint-800/config.json +0 -115
  25. checkpoint-800/optimizer.pt +0 -3
  26. checkpoint-800/preprocessor_config.json +0 -9
  27. checkpoint-800/pytorch_model.bin +0 -3
  28. checkpoint-800/trainer_state.json +0 -328
  29. checkpoint-800/training_args.bin +0 -3
{checkpoint-600 → checkpoint-2900}/config.json RENAMED
File without changes
{checkpoint-600 → checkpoint-2900}/optimizer.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:49c514c1b2ff011d50c4a8e3b7e46c2148904652f6afdcc1f4a32fe18f672012
3
  size 2490337809
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33b3d96b481ecf01e0bcb8f96671330442d04aa9d9e6aaf0bbffc19dfbad6595
3
  size 2490337809
{checkpoint-600 → checkpoint-2900}/preprocessor_config.json RENAMED
File without changes
{checkpoint-700 → checkpoint-2900}/pytorch_model.bin RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:63e29b0b3051d81c522bbad5fd8008db0773293cc66ac34f89d0bc18968817eb
3
  size 1262063089
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0052f401fc490352c790ad1c044b6d3276da131bfbf9ec8d804b048cb318fc44
3
  size 1262063089
{checkpoint-800 → checkpoint-2900}/rng_state.pth RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8e1a22fa0fe527ae7807404dd534e45dc9526c41b4d88efd52d4ea3de5ea2d81
3
  size 14567
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9ffbabc68c0abf373bb1d47895150fa4a375b32269a55044934ea72148fae737
3
  size 14567
{checkpoint-800 → checkpoint-2900}/scaler.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:60c51dda6298af7ed612d10ff39b3f233c2bf5ffd32d965c0c08263e14858ea9
3
  size 559
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7552f402ce2d53d5c0d07e1545091df3a2eca05f071addadd058a1312f7ef92a
3
  size 559
{checkpoint-600 → checkpoint-2900}/scheduler.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ac6482ff77366a04d83b9e577d7ad673af6a0a7a1846af277e139c185db59118
3
  size 623
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:270eff0d1c43b4b556fda7d2bfee54b647b82a69d76da331b7b8336cf791b58a
3
  size 623
checkpoint-2900/trainer_state.json ADDED
@@ -0,0 +1,1147 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.3350592567023358,
5
+ "global_step": 2900,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 1.0305343511450381e-05,
13
+ "loss": 13.0539,
14
+ "step": 20
15
+ },
16
+ {
17
+ "epoch": 0.02,
18
+ "learning_rate": 2.118320610687023e-05,
19
+ "loss": 20.9649,
20
+ "step": 40
21
+ },
22
+ {
23
+ "epoch": 0.03,
24
+ "learning_rate": 3.206106870229007e-05,
25
+ "loss": 21.1001,
26
+ "step": 60
27
+ },
28
+ {
29
+ "epoch": 0.04,
30
+ "learning_rate": 4.3511450381679383e-05,
31
+ "loss": 5.5852,
32
+ "step": 80
33
+ },
34
+ {
35
+ "epoch": 0.05,
36
+ "learning_rate": 5.438931297709923e-05,
37
+ "loss": 5.2002,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.05,
42
+ "eval_loss": 3.5950469970703125,
43
+ "eval_runtime": 194.8685,
44
+ "eval_samples_per_second": 24.853,
45
+ "eval_steps_per_second": 0.78,
46
+ "eval_wer": 1.0,
47
+ "step": 100
48
+ },
49
+ {
50
+ "epoch": 0.06,
51
+ "learning_rate": 6.583969465648854e-05,
52
+ "loss": 3.281,
53
+ "step": 120
54
+ },
55
+ {
56
+ "epoch": 0.06,
57
+ "learning_rate": 7.495301487862176e-05,
58
+ "loss": 3.3442,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.07,
63
+ "learning_rate": 7.472983555207517e-05,
64
+ "loss": 3.5229,
65
+ "step": 160
66
+ },
67
+ {
68
+ "epoch": 0.08,
69
+ "learning_rate": 7.449490994518402e-05,
70
+ "loss": 3.089,
71
+ "step": 180
72
+ },
73
+ {
74
+ "epoch": 0.09,
75
+ "learning_rate": 7.427173061863743e-05,
76
+ "loss": 3.482,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 0.09,
81
+ "eval_loss": 3.274627923965454,
82
+ "eval_runtime": 194.5649,
83
+ "eval_samples_per_second": 24.891,
84
+ "eval_steps_per_second": 0.781,
85
+ "eval_wer": 1.0,
86
+ "step": 200
87
+ },
88
+ {
89
+ "epoch": 0.1,
90
+ "learning_rate": 7.403680501174628e-05,
91
+ "loss": 3.1517,
92
+ "step": 220
93
+ },
94
+ {
95
+ "epoch": 0.11,
96
+ "learning_rate": 7.380187940485512e-05,
97
+ "loss": 3.1617,
98
+ "step": 240
99
+ },
100
+ {
101
+ "epoch": 0.12,
102
+ "learning_rate": 7.357870007830852e-05,
103
+ "loss": 3.2971,
104
+ "step": 260
105
+ },
106
+ {
107
+ "epoch": 0.13,
108
+ "learning_rate": 7.334377447141738e-05,
109
+ "loss": 3.022,
110
+ "step": 280
111
+ },
112
+ {
113
+ "epoch": 0.14,
114
+ "learning_rate": 7.312059514487078e-05,
115
+ "loss": 3.3322,
116
+ "step": 300
117
+ },
118
+ {
119
+ "epoch": 0.14,
120
+ "eval_loss": 3.0716073513031006,
121
+ "eval_runtime": 196.1674,
122
+ "eval_samples_per_second": 24.688,
123
+ "eval_steps_per_second": 0.775,
124
+ "eval_wer": 1.0,
125
+ "step": 300
126
+ },
127
+ {
128
+ "epoch": 0.15,
129
+ "learning_rate": 7.288566953797964e-05,
130
+ "loss": 2.9964,
131
+ "step": 320
132
+ },
133
+ {
134
+ "epoch": 0.16,
135
+ "learning_rate": 7.265074393108849e-05,
136
+ "loss": 3.0777,
137
+ "step": 340
138
+ },
139
+ {
140
+ "epoch": 0.17,
141
+ "learning_rate": 7.242756460454189e-05,
142
+ "loss": 3.2703,
143
+ "step": 360
144
+ },
145
+ {
146
+ "epoch": 0.17,
147
+ "learning_rate": 7.219263899765073e-05,
148
+ "loss": 3.1161,
149
+ "step": 380
150
+ },
151
+ {
152
+ "epoch": 0.18,
153
+ "learning_rate": 7.195771339075958e-05,
154
+ "loss": 3.2435,
155
+ "step": 400
156
+ },
157
+ {
158
+ "epoch": 0.18,
159
+ "eval_loss": 3.063333034515381,
160
+ "eval_runtime": 190.5938,
161
+ "eval_samples_per_second": 25.41,
162
+ "eval_steps_per_second": 0.798,
163
+ "eval_wer": 1.0,
164
+ "step": 400
165
+ },
166
+ {
167
+ "epoch": 0.19,
168
+ "learning_rate": 7.172278778386843e-05,
169
+ "loss": 2.9993,
170
+ "step": 420
171
+ },
172
+ {
173
+ "epoch": 0.2,
174
+ "learning_rate": 7.148786217697727e-05,
175
+ "loss": 3.0555,
176
+ "step": 440
177
+ },
178
+ {
179
+ "epoch": 0.21,
180
+ "learning_rate": 7.126468285043069e-05,
181
+ "loss": 3.1879,
182
+ "step": 460
183
+ },
184
+ {
185
+ "epoch": 0.22,
186
+ "learning_rate": 7.102975724353953e-05,
187
+ "loss": 2.9861,
188
+ "step": 480
189
+ },
190
+ {
191
+ "epoch": 0.23,
192
+ "learning_rate": 7.080657791699295e-05,
193
+ "loss": 3.0833,
194
+ "step": 500
195
+ },
196
+ {
197
+ "epoch": 0.23,
198
+ "eval_loss": 2.9769718647003174,
199
+ "eval_runtime": 195.2639,
200
+ "eval_samples_per_second": 24.802,
201
+ "eval_steps_per_second": 0.778,
202
+ "eval_wer": 1.0,
203
+ "step": 500
204
+ },
205
+ {
206
+ "epoch": 0.24,
207
+ "learning_rate": 7.05716523101018e-05,
208
+ "loss": 3.1608,
209
+ "step": 520
210
+ },
211
+ {
212
+ "epoch": 0.25,
213
+ "learning_rate": 7.033672670321064e-05,
214
+ "loss": 3.0206,
215
+ "step": 540
216
+ },
217
+ {
218
+ "epoch": 0.26,
219
+ "learning_rate": 7.01018010963195e-05,
220
+ "loss": 3.0523,
221
+ "step": 560
222
+ },
223
+ {
224
+ "epoch": 0.27,
225
+ "learning_rate": 6.986687548942835e-05,
226
+ "loss": 3.0222,
227
+ "step": 580
228
+ },
229
+ {
230
+ "epoch": 0.28,
231
+ "learning_rate": 6.964369616288175e-05,
232
+ "loss": 3.0913,
233
+ "step": 600
234
+ },
235
+ {
236
+ "epoch": 0.28,
237
+ "eval_loss": 3.052091598510742,
238
+ "eval_runtime": 189.8149,
239
+ "eval_samples_per_second": 25.514,
240
+ "eval_steps_per_second": 0.801,
241
+ "eval_wer": 1.0,
242
+ "step": 600
243
+ },
244
+ {
245
+ "epoch": 0.29,
246
+ "learning_rate": 6.94087705559906e-05,
247
+ "loss": 3.1987,
248
+ "step": 620
249
+ },
250
+ {
251
+ "epoch": 0.29,
252
+ "learning_rate": 6.917384494909944e-05,
253
+ "loss": 3.1504,
254
+ "step": 640
255
+ },
256
+ {
257
+ "epoch": 0.3,
258
+ "learning_rate": 6.893891934220829e-05,
259
+ "loss": 3.2406,
260
+ "step": 660
261
+ },
262
+ {
263
+ "epoch": 0.31,
264
+ "learning_rate": 6.870399373531715e-05,
265
+ "loss": 3.0632,
266
+ "step": 680
267
+ },
268
+ {
269
+ "epoch": 0.32,
270
+ "learning_rate": 6.848081440877055e-05,
271
+ "loss": 3.1487,
272
+ "step": 700
273
+ },
274
+ {
275
+ "epoch": 0.32,
276
+ "eval_loss": 3.1951305866241455,
277
+ "eval_runtime": 196.8636,
278
+ "eval_samples_per_second": 24.601,
279
+ "eval_steps_per_second": 0.772,
280
+ "eval_wer": 1.0,
281
+ "step": 700
282
+ },
283
+ {
284
+ "epoch": 0.33,
285
+ "learning_rate": 6.824588880187941e-05,
286
+ "loss": 3.298,
287
+ "step": 720
288
+ },
289
+ {
290
+ "epoch": 0.34,
291
+ "learning_rate": 6.801096319498825e-05,
292
+ "loss": 3.1669,
293
+ "step": 740
294
+ },
295
+ {
296
+ "epoch": 0.35,
297
+ "learning_rate": 6.778778386844165e-05,
298
+ "loss": 3.2651,
299
+ "step": 760
300
+ },
301
+ {
302
+ "epoch": 0.36,
303
+ "learning_rate": 6.75528582615505e-05,
304
+ "loss": 3.1562,
305
+ "step": 780
306
+ },
307
+ {
308
+ "epoch": 0.37,
309
+ "learning_rate": 6.732967893500391e-05,
310
+ "loss": 3.3254,
311
+ "step": 800
312
+ },
313
+ {
314
+ "epoch": 0.37,
315
+ "eval_loss": 3.0070509910583496,
316
+ "eval_runtime": 192.0978,
317
+ "eval_samples_per_second": 25.211,
318
+ "eval_steps_per_second": 0.791,
319
+ "eval_wer": 1.0,
320
+ "step": 800
321
+ },
322
+ {
323
+ "epoch": 0.38,
324
+ "learning_rate": 6.709475332811276e-05,
325
+ "loss": 3.0487,
326
+ "step": 820
327
+ },
328
+ {
329
+ "epoch": 0.39,
330
+ "learning_rate": 6.68598277212216e-05,
331
+ "loss": 3.1936,
332
+ "step": 840
333
+ },
334
+ {
335
+ "epoch": 0.4,
336
+ "learning_rate": 6.663664839467502e-05,
337
+ "loss": 3.323,
338
+ "step": 860
339
+ },
340
+ {
341
+ "epoch": 0.41,
342
+ "learning_rate": 6.640172278778387e-05,
343
+ "loss": 2.998,
344
+ "step": 880
345
+ },
346
+ {
347
+ "epoch": 0.41,
348
+ "learning_rate": 6.617854346123726e-05,
349
+ "loss": 3.4698,
350
+ "step": 900
351
+ },
352
+ {
353
+ "epoch": 0.41,
354
+ "eval_loss": 2.995044469833374,
355
+ "eval_runtime": 195.1464,
356
+ "eval_samples_per_second": 24.817,
357
+ "eval_steps_per_second": 0.779,
358
+ "eval_wer": 1.0,
359
+ "step": 900
360
+ },
361
+ {
362
+ "epoch": 0.42,
363
+ "learning_rate": 6.594361785434611e-05,
364
+ "loss": 3.0048,
365
+ "step": 920
366
+ },
367
+ {
368
+ "epoch": 0.43,
369
+ "learning_rate": 6.570869224745496e-05,
370
+ "loss": 3.1447,
371
+ "step": 940
372
+ },
373
+ {
374
+ "epoch": 0.44,
375
+ "learning_rate": 6.548551292090837e-05,
376
+ "loss": 3.3347,
377
+ "step": 960
378
+ },
379
+ {
380
+ "epoch": 0.45,
381
+ "learning_rate": 6.525058731401722e-05,
382
+ "loss": 2.9795,
383
+ "step": 980
384
+ },
385
+ {
386
+ "epoch": 0.46,
387
+ "learning_rate": 6.503915426781518e-05,
388
+ "loss": 3.5905,
389
+ "step": 1000
390
+ },
391
+ {
392
+ "epoch": 0.46,
393
+ "eval_loss": 2.9961767196655273,
394
+ "eval_runtime": 188.3505,
395
+ "eval_samples_per_second": 25.713,
396
+ "eval_steps_per_second": 0.807,
397
+ "eval_wer": 1.0,
398
+ "step": 1000
399
+ },
400
+ {
401
+ "epoch": 0.47,
402
+ "learning_rate": 6.480422866092403e-05,
403
+ "loss": 2.9884,
404
+ "step": 1020
405
+ },
406
+ {
407
+ "epoch": 0.48,
408
+ "learning_rate": 6.456930305403288e-05,
409
+ "loss": 3.175,
410
+ "step": 1040
411
+ },
412
+ {
413
+ "epoch": 0.49,
414
+ "learning_rate": 6.434612372748629e-05,
415
+ "loss": 3.3163,
416
+ "step": 1060
417
+ },
418
+ {
419
+ "epoch": 0.5,
420
+ "learning_rate": 6.411119812059513e-05,
421
+ "loss": 3.0063,
422
+ "step": 1080
423
+ },
424
+ {
425
+ "epoch": 0.51,
426
+ "learning_rate": 6.387627251370398e-05,
427
+ "loss": 3.4965,
428
+ "step": 1100
429
+ },
430
+ {
431
+ "epoch": 0.51,
432
+ "eval_loss": 3.0483579635620117,
433
+ "eval_runtime": 193.4167,
434
+ "eval_samples_per_second": 25.039,
435
+ "eval_steps_per_second": 0.786,
436
+ "eval_wer": 1.0,
437
+ "step": 1100
438
+ },
439
+ {
440
+ "epoch": 0.52,
441
+ "learning_rate": 6.364134690681284e-05,
442
+ "loss": 2.9967,
443
+ "step": 1120
444
+ },
445
+ {
446
+ "epoch": 0.52,
447
+ "learning_rate": 6.340642129992169e-05,
448
+ "loss": 3.2329,
449
+ "step": 1140
450
+ },
451
+ {
452
+ "epoch": 0.53,
453
+ "learning_rate": 6.31832419733751e-05,
454
+ "loss": 3.3814,
455
+ "step": 1160
456
+ },
457
+ {
458
+ "epoch": 0.54,
459
+ "learning_rate": 6.294831636648395e-05,
460
+ "loss": 3.0108,
461
+ "step": 1180
462
+ },
463
+ {
464
+ "epoch": 0.55,
465
+ "learning_rate": 6.272513703993735e-05,
466
+ "loss": 3.6227,
467
+ "step": 1200
468
+ },
469
+ {
470
+ "epoch": 0.55,
471
+ "eval_loss": 3.0750181674957275,
472
+ "eval_runtime": 193.4538,
473
+ "eval_samples_per_second": 25.034,
474
+ "eval_steps_per_second": 0.786,
475
+ "eval_wer": 1.0,
476
+ "step": 1200
477
+ },
478
+ {
479
+ "epoch": 0.56,
480
+ "learning_rate": 6.24902114330462e-05,
481
+ "loss": 2.9917,
482
+ "step": 1220
483
+ },
484
+ {
485
+ "epoch": 0.57,
486
+ "learning_rate": 6.225528582615504e-05,
487
+ "loss": 3.1386,
488
+ "step": 1240
489
+ },
490
+ {
491
+ "epoch": 0.58,
492
+ "learning_rate": 6.203210649960845e-05,
493
+ "loss": 3.3022,
494
+ "step": 1260
495
+ },
496
+ {
497
+ "epoch": 0.59,
498
+ "learning_rate": 6.17971808927173e-05,
499
+ "loss": 3.0139,
500
+ "step": 1280
501
+ },
502
+ {
503
+ "epoch": 0.6,
504
+ "learning_rate": 6.157400156617071e-05,
505
+ "loss": 3.2921,
506
+ "step": 1300
507
+ },
508
+ {
509
+ "epoch": 0.6,
510
+ "eval_loss": 3.019115924835205,
511
+ "eval_runtime": 194.2188,
512
+ "eval_samples_per_second": 24.936,
513
+ "eval_steps_per_second": 0.783,
514
+ "eval_wer": 1.0,
515
+ "step": 1300
516
+ },
517
+ {
518
+ "epoch": 0.61,
519
+ "learning_rate": 6.133907595927956e-05,
520
+ "loss": 3.0432,
521
+ "step": 1320
522
+ },
523
+ {
524
+ "epoch": 0.62,
525
+ "learning_rate": 6.11041503523884e-05,
526
+ "loss": 3.0951,
527
+ "step": 1340
528
+ },
529
+ {
530
+ "epoch": 0.63,
531
+ "learning_rate": 6.088097102584181e-05,
532
+ "loss": 3.189,
533
+ "step": 1360
534
+ },
535
+ {
536
+ "epoch": 0.64,
537
+ "learning_rate": 6.064604541895066e-05,
538
+ "loss": 3.1296,
539
+ "step": 1380
540
+ },
541
+ {
542
+ "epoch": 0.64,
543
+ "learning_rate": 6.0422866092404065e-05,
544
+ "loss": 3.2907,
545
+ "step": 1400
546
+ },
547
+ {
548
+ "epoch": 0.64,
549
+ "eval_loss": 3.06713604927063,
550
+ "eval_runtime": 192.5311,
551
+ "eval_samples_per_second": 25.154,
552
+ "eval_steps_per_second": 0.789,
553
+ "eval_wer": 1.0,
554
+ "step": 1400
555
+ },
556
+ {
557
+ "epoch": 0.65,
558
+ "learning_rate": 6.018794048551292e-05,
559
+ "loss": 3.1849,
560
+ "step": 1420
561
+ },
562
+ {
563
+ "epoch": 0.66,
564
+ "learning_rate": 5.9953014878621765e-05,
565
+ "loss": 3.1682,
566
+ "step": 1440
567
+ },
568
+ {
569
+ "epoch": 0.67,
570
+ "learning_rate": 5.9729835552075165e-05,
571
+ "loss": 3.2002,
572
+ "step": 1460
573
+ },
574
+ {
575
+ "epoch": 0.68,
576
+ "learning_rate": 5.949490994518402e-05,
577
+ "loss": 3.2207,
578
+ "step": 1480
579
+ },
580
+ {
581
+ "epoch": 0.69,
582
+ "learning_rate": 5.9271730618637424e-05,
583
+ "loss": 3.1716,
584
+ "step": 1500
585
+ },
586
+ {
587
+ "epoch": 0.69,
588
+ "eval_loss": 3.093482732772827,
589
+ "eval_runtime": 193.22,
590
+ "eval_samples_per_second": 25.065,
591
+ "eval_steps_per_second": 0.787,
592
+ "eval_wer": 1.0,
593
+ "step": 1500
594
+ },
595
+ {
596
+ "epoch": 0.7,
597
+ "learning_rate": 5.903680501174628e-05,
598
+ "loss": 3.2105,
599
+ "step": 1520
600
+ },
601
+ {
602
+ "epoch": 0.71,
603
+ "learning_rate": 5.8801879404855124e-05,
604
+ "loss": 3.173,
605
+ "step": 1540
606
+ },
607
+ {
608
+ "epoch": 0.72,
609
+ "learning_rate": 5.857870007830853e-05,
610
+ "loss": 3.2066,
611
+ "step": 1560
612
+ },
613
+ {
614
+ "epoch": 0.73,
615
+ "learning_rate": 5.8343774471417384e-05,
616
+ "loss": 3.2196,
617
+ "step": 1580
618
+ },
619
+ {
620
+ "epoch": 0.74,
621
+ "learning_rate": 5.812059514487078e-05,
622
+ "loss": 3.1788,
623
+ "step": 1600
624
+ },
625
+ {
626
+ "epoch": 0.74,
627
+ "eval_loss": 3.092198133468628,
628
+ "eval_runtime": 194.0448,
629
+ "eval_samples_per_second": 24.958,
630
+ "eval_steps_per_second": 0.783,
631
+ "eval_wer": 1.0,
632
+ "step": 1600
633
+ },
634
+ {
635
+ "epoch": 0.75,
636
+ "learning_rate": 5.788566953797963e-05,
637
+ "loss": 3.2075,
638
+ "step": 1620
639
+ },
640
+ {
641
+ "epoch": 0.75,
642
+ "learning_rate": 5.765074393108848e-05,
643
+ "loss": 3.1772,
644
+ "step": 1640
645
+ },
646
+ {
647
+ "epoch": 0.76,
648
+ "learning_rate": 5.742756460454189e-05,
649
+ "loss": 3.2101,
650
+ "step": 1660
651
+ },
652
+ {
653
+ "epoch": 0.77,
654
+ "learning_rate": 5.7192638997650736e-05,
655
+ "loss": 3.2115,
656
+ "step": 1680
657
+ },
658
+ {
659
+ "epoch": 0.78,
660
+ "learning_rate": 5.69812059514487e-05,
661
+ "loss": 3.2986,
662
+ "step": 1700
663
+ },
664
+ {
665
+ "epoch": 0.78,
666
+ "eval_loss": 3.092197895050049,
667
+ "eval_runtime": 194.3305,
668
+ "eval_samples_per_second": 24.921,
669
+ "eval_steps_per_second": 0.782,
670
+ "eval_wer": 1.0,
671
+ "step": 1700
672
+ },
673
+ {
674
+ "epoch": 0.79,
675
+ "learning_rate": 5.674628034455755e-05,
676
+ "loss": 3.2063,
677
+ "step": 1720
678
+ },
679
+ {
680
+ "epoch": 0.8,
681
+ "learning_rate": 5.65113547376664e-05,
682
+ "loss": 3.1723,
683
+ "step": 1740
684
+ },
685
+ {
686
+ "epoch": 0.81,
687
+ "learning_rate": 5.627642913077525e-05,
688
+ "loss": 3.0945,
689
+ "step": 1760
690
+ },
691
+ {
692
+ "epoch": 0.82,
693
+ "learning_rate": 5.6041503523884095e-05,
694
+ "loss": 3.2153,
695
+ "step": 1780
696
+ },
697
+ {
698
+ "epoch": 0.83,
699
+ "learning_rate": 5.580657791699295e-05,
700
+ "loss": 3.0761,
701
+ "step": 1800
702
+ },
703
+ {
704
+ "epoch": 0.83,
705
+ "eval_loss": 3.092197895050049,
706
+ "eval_runtime": 192.7837,
707
+ "eval_samples_per_second": 25.121,
708
+ "eval_steps_per_second": 0.788,
709
+ "eval_wer": 1.0,
710
+ "step": 1800
711
+ },
712
+ {
713
+ "epoch": 0.84,
714
+ "learning_rate": 5.5571652310101795e-05,
715
+ "loss": 3.2061,
716
+ "step": 1820
717
+ },
718
+ {
719
+ "epoch": 0.85,
720
+ "learning_rate": 5.533672670321064e-05,
721
+ "loss": 3.1768,
722
+ "step": 1840
723
+ },
724
+ {
725
+ "epoch": 0.86,
726
+ "learning_rate": 5.5101801096319495e-05,
727
+ "loss": 3.096,
728
+ "step": 1860
729
+ },
730
+ {
731
+ "epoch": 0.87,
732
+ "learning_rate": 5.486687548942834e-05,
733
+ "loss": 3.213,
734
+ "step": 1880
735
+ },
736
+ {
737
+ "epoch": 0.87,
738
+ "learning_rate": 5.464369616288175e-05,
739
+ "loss": 3.1832,
740
+ "step": 1900
741
+ },
742
+ {
743
+ "epoch": 0.87,
744
+ "eval_loss": 3.092197895050049,
745
+ "eval_runtime": 196.1769,
746
+ "eval_samples_per_second": 24.687,
747
+ "eval_steps_per_second": 0.775,
748
+ "eval_wer": 1.0,
749
+ "step": 1900
750
+ },
751
+ {
752
+ "epoch": 0.88,
753
+ "learning_rate": 5.44087705559906e-05,
754
+ "loss": 3.2046,
755
+ "step": 1920
756
+ },
757
+ {
758
+ "epoch": 0.89,
759
+ "learning_rate": 5.417384494909945e-05,
760
+ "loss": 3.1743,
761
+ "step": 1940
762
+ },
763
+ {
764
+ "epoch": 0.9,
765
+ "learning_rate": 5.395066562255285e-05,
766
+ "loss": 3.2055,
767
+ "step": 1960
768
+ },
769
+ {
770
+ "epoch": 0.91,
771
+ "learning_rate": 5.37157400156617e-05,
772
+ "loss": 3.2152,
773
+ "step": 1980
774
+ },
775
+ {
776
+ "epoch": 0.92,
777
+ "learning_rate": 5.349256068911511e-05,
778
+ "loss": 3.1845,
779
+ "step": 2000
780
+ },
781
+ {
782
+ "epoch": 0.92,
783
+ "eval_loss": 3.092197895050049,
784
+ "eval_runtime": 195.9315,
785
+ "eval_samples_per_second": 24.718,
786
+ "eval_steps_per_second": 0.776,
787
+ "eval_wer": 1.0,
788
+ "step": 2000
789
+ },
790
+ {
791
+ "epoch": 0.93,
792
+ "learning_rate": 5.3257635082223954e-05,
793
+ "loss": 3.2093,
794
+ "step": 2020
795
+ },
796
+ {
797
+ "epoch": 0.94,
798
+ "learning_rate": 5.302270947533281e-05,
799
+ "loss": 3.1748,
800
+ "step": 2040
801
+ },
802
+ {
803
+ "epoch": 0.95,
804
+ "learning_rate": 5.279953014878621e-05,
805
+ "loss": 3.199,
806
+ "step": 2060
807
+ },
808
+ {
809
+ "epoch": 0.96,
810
+ "learning_rate": 5.2564604541895067e-05,
811
+ "loss": 3.2169,
812
+ "step": 2080
813
+ },
814
+ {
815
+ "epoch": 0.97,
816
+ "learning_rate": 5.2341425215348466e-05,
817
+ "loss": 3.1817,
818
+ "step": 2100
819
+ },
820
+ {
821
+ "epoch": 0.97,
822
+ "eval_loss": 3.092197895050049,
823
+ "eval_runtime": 193.8566,
824
+ "eval_samples_per_second": 24.982,
825
+ "eval_steps_per_second": 0.784,
826
+ "eval_wer": 1.0,
827
+ "step": 2100
828
+ },
829
+ {
830
+ "epoch": 0.98,
831
+ "learning_rate": 5.210649960845731e-05,
832
+ "loss": 3.2075,
833
+ "step": 2120
834
+ },
835
+ {
836
+ "epoch": 0.98,
837
+ "learning_rate": 5.1871574001566166e-05,
838
+ "loss": 3.1693,
839
+ "step": 2140
840
+ },
841
+ {
842
+ "epoch": 0.99,
843
+ "learning_rate": 5.164839467501957e-05,
844
+ "loss": 3.2049,
845
+ "step": 2160
846
+ },
847
+ {
848
+ "epoch": 1.0,
849
+ "learning_rate": 5.1425215348472984e-05,
850
+ "loss": 3.3657,
851
+ "step": 2180
852
+ },
853
+ {
854
+ "epoch": 1.01,
855
+ "learning_rate": 5.119028974158183e-05,
856
+ "loss": 3.2209,
857
+ "step": 2200
858
+ },
859
+ {
860
+ "epoch": 1.01,
861
+ "eval_loss": 3.092197895050049,
862
+ "eval_runtime": 194.751,
863
+ "eval_samples_per_second": 24.868,
864
+ "eval_steps_per_second": 0.78,
865
+ "eval_wer": 1.0,
866
+ "step": 2200
867
+ },
868
+ {
869
+ "epoch": 1.02,
870
+ "learning_rate": 5.095536413469068e-05,
871
+ "loss": 3.0961,
872
+ "step": 2220
873
+ },
874
+ {
875
+ "epoch": 1.03,
876
+ "learning_rate": 5.073218480814408e-05,
877
+ "loss": 3.2735,
878
+ "step": 2240
879
+ },
880
+ {
881
+ "epoch": 1.04,
882
+ "learning_rate": 5.049725920125293e-05,
883
+ "loss": 3.1904,
884
+ "step": 2260
885
+ },
886
+ {
887
+ "epoch": 1.05,
888
+ "learning_rate": 5.0274079874706336e-05,
889
+ "loss": 3.1892,
890
+ "step": 2280
891
+ },
892
+ {
893
+ "epoch": 1.06,
894
+ "learning_rate": 5.003915426781519e-05,
895
+ "loss": 3.2138,
896
+ "step": 2300
897
+ },
898
+ {
899
+ "epoch": 1.06,
900
+ "eval_loss": 3.092197895050049,
901
+ "eval_runtime": 195.285,
902
+ "eval_samples_per_second": 24.8,
903
+ "eval_steps_per_second": 0.778,
904
+ "eval_wer": 1.0,
905
+ "step": 2300
906
+ },
907
+ {
908
+ "epoch": 1.07,
909
+ "learning_rate": 4.980422866092404e-05,
910
+ "loss": 3.0819,
911
+ "step": 2320
912
+ },
913
+ {
914
+ "epoch": 1.08,
915
+ "learning_rate": 4.958104933437744e-05,
916
+ "loss": 3.272,
917
+ "step": 2340
918
+ },
919
+ {
920
+ "epoch": 1.09,
921
+ "learning_rate": 4.9346123727486296e-05,
922
+ "loss": 3.2011,
923
+ "step": 2360
924
+ },
925
+ {
926
+ "epoch": 1.1,
927
+ "learning_rate": 4.9122944400939695e-05,
928
+ "loss": 3.1918,
929
+ "step": 2380
930
+ },
931
+ {
932
+ "epoch": 1.1,
933
+ "learning_rate": 4.888801879404854e-05,
934
+ "loss": 3.2154,
935
+ "step": 2400
936
+ },
937
+ {
938
+ "epoch": 1.1,
939
+ "eval_loss": 3.092197895050049,
940
+ "eval_runtime": 196.4218,
941
+ "eval_samples_per_second": 24.656,
942
+ "eval_steps_per_second": 0.774,
943
+ "eval_wer": 1.0,
944
+ "step": 2400
945
+ },
946
+ {
947
+ "epoch": 1.11,
948
+ "learning_rate": 4.8653093187157396e-05,
949
+ "loss": 3.0946,
950
+ "step": 2420
951
+ },
952
+ {
953
+ "epoch": 1.12,
954
+ "learning_rate": 4.84299138606108e-05,
955
+ "loss": 3.2753,
956
+ "step": 2440
957
+ },
958
+ {
959
+ "epoch": 1.13,
960
+ "learning_rate": 4.8194988253719655e-05,
961
+ "loss": 3.1829,
962
+ "step": 2460
963
+ },
964
+ {
965
+ "epoch": 1.14,
966
+ "learning_rate": 4.797180892717306e-05,
967
+ "loss": 3.1839,
968
+ "step": 2480
969
+ },
970
+ {
971
+ "epoch": 1.15,
972
+ "learning_rate": 4.773688332028191e-05,
973
+ "loss": 3.2185,
974
+ "step": 2500
975
+ },
976
+ {
977
+ "epoch": 1.15,
978
+ "eval_loss": 3.092197895050049,
979
+ "eval_runtime": 192.8929,
980
+ "eval_samples_per_second": 25.107,
981
+ "eval_steps_per_second": 0.788,
982
+ "eval_wer": 1.0,
983
+ "step": 2500
984
+ },
985
+ {
986
+ "epoch": 1.16,
987
+ "learning_rate": 4.750195771339076e-05,
988
+ "loss": 3.0971,
989
+ "step": 2520
990
+ },
991
+ {
992
+ "epoch": 1.17,
993
+ "learning_rate": 4.727877838684416e-05,
994
+ "loss": 3.2719,
995
+ "step": 2540
996
+ },
997
+ {
998
+ "epoch": 1.18,
999
+ "learning_rate": 4.704385277995301e-05,
1000
+ "loss": 3.1974,
1001
+ "step": 2560
1002
+ },
1003
+ {
1004
+ "epoch": 1.19,
1005
+ "learning_rate": 4.680892717306186e-05,
1006
+ "loss": 3.0775,
1007
+ "step": 2580
1008
+ },
1009
+ {
1010
+ "epoch": 1.2,
1011
+ "learning_rate": 4.657400156617071e-05,
1012
+ "loss": 3.2201,
1013
+ "step": 2600
1014
+ },
1015
+ {
1016
+ "epoch": 1.2,
1017
+ "eval_loss": 3.092197895050049,
1018
+ "eval_runtime": 192.9488,
1019
+ "eval_samples_per_second": 25.1,
1020
+ "eval_steps_per_second": 0.788,
1021
+ "eval_wer": 1.0,
1022
+ "step": 2600
1023
+ },
1024
+ {
1025
+ "epoch": 1.21,
1026
+ "learning_rate": 4.6339075959279554e-05,
1027
+ "loss": 3.0946,
1028
+ "step": 2620
1029
+ },
1030
+ {
1031
+ "epoch": 1.22,
1032
+ "learning_rate": 4.611589663273297e-05,
1033
+ "loss": 3.2708,
1034
+ "step": 2640
1035
+ },
1036
+ {
1037
+ "epoch": 1.22,
1038
+ "learning_rate": 4.5880971025841814e-05,
1039
+ "loss": 3.1949,
1040
+ "step": 2660
1041
+ },
1042
+ {
1043
+ "epoch": 1.23,
1044
+ "learning_rate": 4.564604541895066e-05,
1045
+ "loss": 3.0763,
1046
+ "step": 2680
1047
+ },
1048
+ {
1049
+ "epoch": 1.24,
1050
+ "learning_rate": 4.5411119812059514e-05,
1051
+ "loss": 3.2184,
1052
+ "step": 2700
1053
+ },
1054
+ {
1055
+ "epoch": 1.24,
1056
+ "eval_loss": 3.092197895050049,
1057
+ "eval_runtime": 189.7646,
1058
+ "eval_samples_per_second": 25.521,
1059
+ "eval_steps_per_second": 0.801,
1060
+ "eval_wer": 1.0,
1061
+ "step": 2700
1062
+ },
1063
+ {
1064
+ "epoch": 1.25,
1065
+ "learning_rate": 4.517619420516836e-05,
1066
+ "loss": 3.0962,
1067
+ "step": 2720
1068
+ },
1069
+ {
1070
+ "epoch": 1.26,
1071
+ "learning_rate": 4.495301487862176e-05,
1072
+ "loss": 3.2795,
1073
+ "step": 2740
1074
+ },
1075
+ {
1076
+ "epoch": 1.27,
1077
+ "learning_rate": 4.4718089271730614e-05,
1078
+ "loss": 3.1901,
1079
+ "step": 2760
1080
+ },
1081
+ {
1082
+ "epoch": 1.28,
1083
+ "learning_rate": 4.449490994518402e-05,
1084
+ "loss": 3.1902,
1085
+ "step": 2780
1086
+ },
1087
+ {
1088
+ "epoch": 1.29,
1089
+ "learning_rate": 4.425998433829287e-05,
1090
+ "loss": 3.2196,
1091
+ "step": 2800
1092
+ },
1093
+ {
1094
+ "epoch": 1.29,
1095
+ "eval_loss": 3.092197895050049,
1096
+ "eval_runtime": 192.1661,
1097
+ "eval_samples_per_second": 25.202,
1098
+ "eval_steps_per_second": 0.791,
1099
+ "eval_wer": 1.0,
1100
+ "step": 2800
1101
+ },
1102
+ {
1103
+ "epoch": 1.3,
1104
+ "learning_rate": 4.402505873140172e-05,
1105
+ "loss": 3.0923,
1106
+ "step": 2820
1107
+ },
1108
+ {
1109
+ "epoch": 1.31,
1110
+ "learning_rate": 4.3801879404855126e-05,
1111
+ "loss": 3.2783,
1112
+ "step": 2840
1113
+ },
1114
+ {
1115
+ "epoch": 1.32,
1116
+ "learning_rate": 4.356695379796398e-05,
1117
+ "loss": 3.195,
1118
+ "step": 2860
1119
+ },
1120
+ {
1121
+ "epoch": 1.33,
1122
+ "learning_rate": 4.335552075176194e-05,
1123
+ "loss": 3.3041,
1124
+ "step": 2880
1125
+ },
1126
+ {
1127
+ "epoch": 1.34,
1128
+ "learning_rate": 4.3120595144870784e-05,
1129
+ "loss": 3.2158,
1130
+ "step": 2900
1131
+ },
1132
+ {
1133
+ "epoch": 1.34,
1134
+ "eval_loss": 3.092197895050049,
1135
+ "eval_runtime": 194.7934,
1136
+ "eval_samples_per_second": 24.862,
1137
+ "eval_steps_per_second": 0.78,
1138
+ "eval_wer": 1.0,
1139
+ "step": 2900
1140
+ }
1141
+ ],
1142
+ "max_steps": 6516,
1143
+ "num_train_epochs": 3,
1144
+ "total_flos": 5.539671173723811e+19,
1145
+ "trial_name": null,
1146
+ "trial_params": null
1147
+ }
{checkpoint-600 → checkpoint-2900}/training_args.bin RENAMED
File without changes
{checkpoint-700 → checkpoint-3000}/config.json RENAMED
File without changes
{checkpoint-700 → checkpoint-3000}/optimizer.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fe48dc95f416a61e29c9b384b89669597e505aff37f490432393562fd09a7ebc
3
  size 2490337809
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:451c860a843d9f24cd86a79754e01529291d9c93ed7ae857fc15b2165971ef2e
3
  size 2490337809
{checkpoint-700 → checkpoint-3000}/preprocessor_config.json RENAMED
File without changes
{checkpoint-600 → checkpoint-3000}/pytorch_model.bin RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:058906a444aeea71d083f349ade14c36ec6bf31aefcfd633c92657ebe41f3c14
3
  size 1262063089
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0052f401fc490352c790ad1c044b6d3276da131bfbf9ec8d804b048cb318fc44
3
  size 1262063089
{checkpoint-700 → checkpoint-3000}/rng_state.pth RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:299f0081a5cbf561a975f939c328f4686396da29b3233cdb8a83cc793fe92137
3
  size 14567
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b6404ed02f448394a66c9e17203ff3a69646d1570b4570e36ea6110a5d5e330
3
  size 14567
{checkpoint-700 → checkpoint-3000}/scaler.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dd9adbb7bf7a7b94e5da4d383115b816c16f892b14cb7df333d0775ae0f262ae
3
  size 559
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7c4960c2340a2bf713dd0ea0bbca386f6a864f908028378e9ceac66da9c35c7
3
  size 559
{checkpoint-800 → checkpoint-3000}/scheduler.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:179d0711ab8fef56ecf9a9aea25248daf43634aa94dfc5cfe7a4eb224c90b32e
3
  size 623
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:477f93ca257d8ff2b5b08a14ea64bd27150efc48fbd9c9ef1895a607c1cff8b5
3
  size 623
checkpoint-3000/trainer_state.json ADDED
@@ -0,0 +1,1186 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.3810838798757334,
5
+ "global_step": 3000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 1.0305343511450381e-05,
13
+ "loss": 13.0539,
14
+ "step": 20
15
+ },
16
+ {
17
+ "epoch": 0.02,
18
+ "learning_rate": 2.118320610687023e-05,
19
+ "loss": 20.9649,
20
+ "step": 40
21
+ },
22
+ {
23
+ "epoch": 0.03,
24
+ "learning_rate": 3.206106870229007e-05,
25
+ "loss": 21.1001,
26
+ "step": 60
27
+ },
28
+ {
29
+ "epoch": 0.04,
30
+ "learning_rate": 4.3511450381679383e-05,
31
+ "loss": 5.5852,
32
+ "step": 80
33
+ },
34
+ {
35
+ "epoch": 0.05,
36
+ "learning_rate": 5.438931297709923e-05,
37
+ "loss": 5.2002,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.05,
42
+ "eval_loss": 3.5950469970703125,
43
+ "eval_runtime": 194.8685,
44
+ "eval_samples_per_second": 24.853,
45
+ "eval_steps_per_second": 0.78,
46
+ "eval_wer": 1.0,
47
+ "step": 100
48
+ },
49
+ {
50
+ "epoch": 0.06,
51
+ "learning_rate": 6.583969465648854e-05,
52
+ "loss": 3.281,
53
+ "step": 120
54
+ },
55
+ {
56
+ "epoch": 0.06,
57
+ "learning_rate": 7.495301487862176e-05,
58
+ "loss": 3.3442,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.07,
63
+ "learning_rate": 7.472983555207517e-05,
64
+ "loss": 3.5229,
65
+ "step": 160
66
+ },
67
+ {
68
+ "epoch": 0.08,
69
+ "learning_rate": 7.449490994518402e-05,
70
+ "loss": 3.089,
71
+ "step": 180
72
+ },
73
+ {
74
+ "epoch": 0.09,
75
+ "learning_rate": 7.427173061863743e-05,
76
+ "loss": 3.482,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 0.09,
81
+ "eval_loss": 3.274627923965454,
82
+ "eval_runtime": 194.5649,
83
+ "eval_samples_per_second": 24.891,
84
+ "eval_steps_per_second": 0.781,
85
+ "eval_wer": 1.0,
86
+ "step": 200
87
+ },
88
+ {
89
+ "epoch": 0.1,
90
+ "learning_rate": 7.403680501174628e-05,
91
+ "loss": 3.1517,
92
+ "step": 220
93
+ },
94
+ {
95
+ "epoch": 0.11,
96
+ "learning_rate": 7.380187940485512e-05,
97
+ "loss": 3.1617,
98
+ "step": 240
99
+ },
100
+ {
101
+ "epoch": 0.12,
102
+ "learning_rate": 7.357870007830852e-05,
103
+ "loss": 3.2971,
104
+ "step": 260
105
+ },
106
+ {
107
+ "epoch": 0.13,
108
+ "learning_rate": 7.334377447141738e-05,
109
+ "loss": 3.022,
110
+ "step": 280
111
+ },
112
+ {
113
+ "epoch": 0.14,
114
+ "learning_rate": 7.312059514487078e-05,
115
+ "loss": 3.3322,
116
+ "step": 300
117
+ },
118
+ {
119
+ "epoch": 0.14,
120
+ "eval_loss": 3.0716073513031006,
121
+ "eval_runtime": 196.1674,
122
+ "eval_samples_per_second": 24.688,
123
+ "eval_steps_per_second": 0.775,
124
+ "eval_wer": 1.0,
125
+ "step": 300
126
+ },
127
+ {
128
+ "epoch": 0.15,
129
+ "learning_rate": 7.288566953797964e-05,
130
+ "loss": 2.9964,
131
+ "step": 320
132
+ },
133
+ {
134
+ "epoch": 0.16,
135
+ "learning_rate": 7.265074393108849e-05,
136
+ "loss": 3.0777,
137
+ "step": 340
138
+ },
139
+ {
140
+ "epoch": 0.17,
141
+ "learning_rate": 7.242756460454189e-05,
142
+ "loss": 3.2703,
143
+ "step": 360
144
+ },
145
+ {
146
+ "epoch": 0.17,
147
+ "learning_rate": 7.219263899765073e-05,
148
+ "loss": 3.1161,
149
+ "step": 380
150
+ },
151
+ {
152
+ "epoch": 0.18,
153
+ "learning_rate": 7.195771339075958e-05,
154
+ "loss": 3.2435,
155
+ "step": 400
156
+ },
157
+ {
158
+ "epoch": 0.18,
159
+ "eval_loss": 3.063333034515381,
160
+ "eval_runtime": 190.5938,
161
+ "eval_samples_per_second": 25.41,
162
+ "eval_steps_per_second": 0.798,
163
+ "eval_wer": 1.0,
164
+ "step": 400
165
+ },
166
+ {
167
+ "epoch": 0.19,
168
+ "learning_rate": 7.172278778386843e-05,
169
+ "loss": 2.9993,
170
+ "step": 420
171
+ },
172
+ {
173
+ "epoch": 0.2,
174
+ "learning_rate": 7.148786217697727e-05,
175
+ "loss": 3.0555,
176
+ "step": 440
177
+ },
178
+ {
179
+ "epoch": 0.21,
180
+ "learning_rate": 7.126468285043069e-05,
181
+ "loss": 3.1879,
182
+ "step": 460
183
+ },
184
+ {
185
+ "epoch": 0.22,
186
+ "learning_rate": 7.102975724353953e-05,
187
+ "loss": 2.9861,
188
+ "step": 480
189
+ },
190
+ {
191
+ "epoch": 0.23,
192
+ "learning_rate": 7.080657791699295e-05,
193
+ "loss": 3.0833,
194
+ "step": 500
195
+ },
196
+ {
197
+ "epoch": 0.23,
198
+ "eval_loss": 2.9769718647003174,
199
+ "eval_runtime": 195.2639,
200
+ "eval_samples_per_second": 24.802,
201
+ "eval_steps_per_second": 0.778,
202
+ "eval_wer": 1.0,
203
+ "step": 500
204
+ },
205
+ {
206
+ "epoch": 0.24,
207
+ "learning_rate": 7.05716523101018e-05,
208
+ "loss": 3.1608,
209
+ "step": 520
210
+ },
211
+ {
212
+ "epoch": 0.25,
213
+ "learning_rate": 7.033672670321064e-05,
214
+ "loss": 3.0206,
215
+ "step": 540
216
+ },
217
+ {
218
+ "epoch": 0.26,
219
+ "learning_rate": 7.01018010963195e-05,
220
+ "loss": 3.0523,
221
+ "step": 560
222
+ },
223
+ {
224
+ "epoch": 0.27,
225
+ "learning_rate": 6.986687548942835e-05,
226
+ "loss": 3.0222,
227
+ "step": 580
228
+ },
229
+ {
230
+ "epoch": 0.28,
231
+ "learning_rate": 6.964369616288175e-05,
232
+ "loss": 3.0913,
233
+ "step": 600
234
+ },
235
+ {
236
+ "epoch": 0.28,
237
+ "eval_loss": 3.052091598510742,
238
+ "eval_runtime": 189.8149,
239
+ "eval_samples_per_second": 25.514,
240
+ "eval_steps_per_second": 0.801,
241
+ "eval_wer": 1.0,
242
+ "step": 600
243
+ },
244
+ {
245
+ "epoch": 0.29,
246
+ "learning_rate": 6.94087705559906e-05,
247
+ "loss": 3.1987,
248
+ "step": 620
249
+ },
250
+ {
251
+ "epoch": 0.29,
252
+ "learning_rate": 6.917384494909944e-05,
253
+ "loss": 3.1504,
254
+ "step": 640
255
+ },
256
+ {
257
+ "epoch": 0.3,
258
+ "learning_rate": 6.893891934220829e-05,
259
+ "loss": 3.2406,
260
+ "step": 660
261
+ },
262
+ {
263
+ "epoch": 0.31,
264
+ "learning_rate": 6.870399373531715e-05,
265
+ "loss": 3.0632,
266
+ "step": 680
267
+ },
268
+ {
269
+ "epoch": 0.32,
270
+ "learning_rate": 6.848081440877055e-05,
271
+ "loss": 3.1487,
272
+ "step": 700
273
+ },
274
+ {
275
+ "epoch": 0.32,
276
+ "eval_loss": 3.1951305866241455,
277
+ "eval_runtime": 196.8636,
278
+ "eval_samples_per_second": 24.601,
279
+ "eval_steps_per_second": 0.772,
280
+ "eval_wer": 1.0,
281
+ "step": 700
282
+ },
283
+ {
284
+ "epoch": 0.33,
285
+ "learning_rate": 6.824588880187941e-05,
286
+ "loss": 3.298,
287
+ "step": 720
288
+ },
289
+ {
290
+ "epoch": 0.34,
291
+ "learning_rate": 6.801096319498825e-05,
292
+ "loss": 3.1669,
293
+ "step": 740
294
+ },
295
+ {
296
+ "epoch": 0.35,
297
+ "learning_rate": 6.778778386844165e-05,
298
+ "loss": 3.2651,
299
+ "step": 760
300
+ },
301
+ {
302
+ "epoch": 0.36,
303
+ "learning_rate": 6.75528582615505e-05,
304
+ "loss": 3.1562,
305
+ "step": 780
306
+ },
307
+ {
308
+ "epoch": 0.37,
309
+ "learning_rate": 6.732967893500391e-05,
310
+ "loss": 3.3254,
311
+ "step": 800
312
+ },
313
+ {
314
+ "epoch": 0.37,
315
+ "eval_loss": 3.0070509910583496,
316
+ "eval_runtime": 192.0978,
317
+ "eval_samples_per_second": 25.211,
318
+ "eval_steps_per_second": 0.791,
319
+ "eval_wer": 1.0,
320
+ "step": 800
321
+ },
322
+ {
323
+ "epoch": 0.38,
324
+ "learning_rate": 6.709475332811276e-05,
325
+ "loss": 3.0487,
326
+ "step": 820
327
+ },
328
+ {
329
+ "epoch": 0.39,
330
+ "learning_rate": 6.68598277212216e-05,
331
+ "loss": 3.1936,
332
+ "step": 840
333
+ },
334
+ {
335
+ "epoch": 0.4,
336
+ "learning_rate": 6.663664839467502e-05,
337
+ "loss": 3.323,
338
+ "step": 860
339
+ },
340
+ {
341
+ "epoch": 0.41,
342
+ "learning_rate": 6.640172278778387e-05,
343
+ "loss": 2.998,
344
+ "step": 880
345
+ },
346
+ {
347
+ "epoch": 0.41,
348
+ "learning_rate": 6.617854346123726e-05,
349
+ "loss": 3.4698,
350
+ "step": 900
351
+ },
352
+ {
353
+ "epoch": 0.41,
354
+ "eval_loss": 2.995044469833374,
355
+ "eval_runtime": 195.1464,
356
+ "eval_samples_per_second": 24.817,
357
+ "eval_steps_per_second": 0.779,
358
+ "eval_wer": 1.0,
359
+ "step": 900
360
+ },
361
+ {
362
+ "epoch": 0.42,
363
+ "learning_rate": 6.594361785434611e-05,
364
+ "loss": 3.0048,
365
+ "step": 920
366
+ },
367
+ {
368
+ "epoch": 0.43,
369
+ "learning_rate": 6.570869224745496e-05,
370
+ "loss": 3.1447,
371
+ "step": 940
372
+ },
373
+ {
374
+ "epoch": 0.44,
375
+ "learning_rate": 6.548551292090837e-05,
376
+ "loss": 3.3347,
377
+ "step": 960
378
+ },
379
+ {
380
+ "epoch": 0.45,
381
+ "learning_rate": 6.525058731401722e-05,
382
+ "loss": 2.9795,
383
+ "step": 980
384
+ },
385
+ {
386
+ "epoch": 0.46,
387
+ "learning_rate": 6.503915426781518e-05,
388
+ "loss": 3.5905,
389
+ "step": 1000
390
+ },
391
+ {
392
+ "epoch": 0.46,
393
+ "eval_loss": 2.9961767196655273,
394
+ "eval_runtime": 188.3505,
395
+ "eval_samples_per_second": 25.713,
396
+ "eval_steps_per_second": 0.807,
397
+ "eval_wer": 1.0,
398
+ "step": 1000
399
+ },
400
+ {
401
+ "epoch": 0.47,
402
+ "learning_rate": 6.480422866092403e-05,
403
+ "loss": 2.9884,
404
+ "step": 1020
405
+ },
406
+ {
407
+ "epoch": 0.48,
408
+ "learning_rate": 6.456930305403288e-05,
409
+ "loss": 3.175,
410
+ "step": 1040
411
+ },
412
+ {
413
+ "epoch": 0.49,
414
+ "learning_rate": 6.434612372748629e-05,
415
+ "loss": 3.3163,
416
+ "step": 1060
417
+ },
418
+ {
419
+ "epoch": 0.5,
420
+ "learning_rate": 6.411119812059513e-05,
421
+ "loss": 3.0063,
422
+ "step": 1080
423
+ },
424
+ {
425
+ "epoch": 0.51,
426
+ "learning_rate": 6.387627251370398e-05,
427
+ "loss": 3.4965,
428
+ "step": 1100
429
+ },
430
+ {
431
+ "epoch": 0.51,
432
+ "eval_loss": 3.0483579635620117,
433
+ "eval_runtime": 193.4167,
434
+ "eval_samples_per_second": 25.039,
435
+ "eval_steps_per_second": 0.786,
436
+ "eval_wer": 1.0,
437
+ "step": 1100
438
+ },
439
+ {
440
+ "epoch": 0.52,
441
+ "learning_rate": 6.364134690681284e-05,
442
+ "loss": 2.9967,
443
+ "step": 1120
444
+ },
445
+ {
446
+ "epoch": 0.52,
447
+ "learning_rate": 6.340642129992169e-05,
448
+ "loss": 3.2329,
449
+ "step": 1140
450
+ },
451
+ {
452
+ "epoch": 0.53,
453
+ "learning_rate": 6.31832419733751e-05,
454
+ "loss": 3.3814,
455
+ "step": 1160
456
+ },
457
+ {
458
+ "epoch": 0.54,
459
+ "learning_rate": 6.294831636648395e-05,
460
+ "loss": 3.0108,
461
+ "step": 1180
462
+ },
463
+ {
464
+ "epoch": 0.55,
465
+ "learning_rate": 6.272513703993735e-05,
466
+ "loss": 3.6227,
467
+ "step": 1200
468
+ },
469
+ {
470
+ "epoch": 0.55,
471
+ "eval_loss": 3.0750181674957275,
472
+ "eval_runtime": 193.4538,
473
+ "eval_samples_per_second": 25.034,
474
+ "eval_steps_per_second": 0.786,
475
+ "eval_wer": 1.0,
476
+ "step": 1200
477
+ },
478
+ {
479
+ "epoch": 0.56,
480
+ "learning_rate": 6.24902114330462e-05,
481
+ "loss": 2.9917,
482
+ "step": 1220
483
+ },
484
+ {
485
+ "epoch": 0.57,
486
+ "learning_rate": 6.225528582615504e-05,
487
+ "loss": 3.1386,
488
+ "step": 1240
489
+ },
490
+ {
491
+ "epoch": 0.58,
492
+ "learning_rate": 6.203210649960845e-05,
493
+ "loss": 3.3022,
494
+ "step": 1260
495
+ },
496
+ {
497
+ "epoch": 0.59,
498
+ "learning_rate": 6.17971808927173e-05,
499
+ "loss": 3.0139,
500
+ "step": 1280
501
+ },
502
+ {
503
+ "epoch": 0.6,
504
+ "learning_rate": 6.157400156617071e-05,
505
+ "loss": 3.2921,
506
+ "step": 1300
507
+ },
508
+ {
509
+ "epoch": 0.6,
510
+ "eval_loss": 3.019115924835205,
511
+ "eval_runtime": 194.2188,
512
+ "eval_samples_per_second": 24.936,
513
+ "eval_steps_per_second": 0.783,
514
+ "eval_wer": 1.0,
515
+ "step": 1300
516
+ },
517
+ {
518
+ "epoch": 0.61,
519
+ "learning_rate": 6.133907595927956e-05,
520
+ "loss": 3.0432,
521
+ "step": 1320
522
+ },
523
+ {
524
+ "epoch": 0.62,
525
+ "learning_rate": 6.11041503523884e-05,
526
+ "loss": 3.0951,
527
+ "step": 1340
528
+ },
529
+ {
530
+ "epoch": 0.63,
531
+ "learning_rate": 6.088097102584181e-05,
532
+ "loss": 3.189,
533
+ "step": 1360
534
+ },
535
+ {
536
+ "epoch": 0.64,
537
+ "learning_rate": 6.064604541895066e-05,
538
+ "loss": 3.1296,
539
+ "step": 1380
540
+ },
541
+ {
542
+ "epoch": 0.64,
543
+ "learning_rate": 6.0422866092404065e-05,
544
+ "loss": 3.2907,
545
+ "step": 1400
546
+ },
547
+ {
548
+ "epoch": 0.64,
549
+ "eval_loss": 3.06713604927063,
550
+ "eval_runtime": 192.5311,
551
+ "eval_samples_per_second": 25.154,
552
+ "eval_steps_per_second": 0.789,
553
+ "eval_wer": 1.0,
554
+ "step": 1400
555
+ },
556
+ {
557
+ "epoch": 0.65,
558
+ "learning_rate": 6.018794048551292e-05,
559
+ "loss": 3.1849,
560
+ "step": 1420
561
+ },
562
+ {
563
+ "epoch": 0.66,
564
+ "learning_rate": 5.9953014878621765e-05,
565
+ "loss": 3.1682,
566
+ "step": 1440
567
+ },
568
+ {
569
+ "epoch": 0.67,
570
+ "learning_rate": 5.9729835552075165e-05,
571
+ "loss": 3.2002,
572
+ "step": 1460
573
+ },
574
+ {
575
+ "epoch": 0.68,
576
+ "learning_rate": 5.949490994518402e-05,
577
+ "loss": 3.2207,
578
+ "step": 1480
579
+ },
580
+ {
581
+ "epoch": 0.69,
582
+ "learning_rate": 5.9271730618637424e-05,
583
+ "loss": 3.1716,
584
+ "step": 1500
585
+ },
586
+ {
587
+ "epoch": 0.69,
588
+ "eval_loss": 3.093482732772827,
589
+ "eval_runtime": 193.22,
590
+ "eval_samples_per_second": 25.065,
591
+ "eval_steps_per_second": 0.787,
592
+ "eval_wer": 1.0,
593
+ "step": 1500
594
+ },
595
+ {
596
+ "epoch": 0.7,
597
+ "learning_rate": 5.903680501174628e-05,
598
+ "loss": 3.2105,
599
+ "step": 1520
600
+ },
601
+ {
602
+ "epoch": 0.71,
603
+ "learning_rate": 5.8801879404855124e-05,
604
+ "loss": 3.173,
605
+ "step": 1540
606
+ },
607
+ {
608
+ "epoch": 0.72,
609
+ "learning_rate": 5.857870007830853e-05,
610
+ "loss": 3.2066,
611
+ "step": 1560
612
+ },
613
+ {
614
+ "epoch": 0.73,
615
+ "learning_rate": 5.8343774471417384e-05,
616
+ "loss": 3.2196,
617
+ "step": 1580
618
+ },
619
+ {
620
+ "epoch": 0.74,
621
+ "learning_rate": 5.812059514487078e-05,
622
+ "loss": 3.1788,
623
+ "step": 1600
624
+ },
625
+ {
626
+ "epoch": 0.74,
627
+ "eval_loss": 3.092198133468628,
628
+ "eval_runtime": 194.0448,
629
+ "eval_samples_per_second": 24.958,
630
+ "eval_steps_per_second": 0.783,
631
+ "eval_wer": 1.0,
632
+ "step": 1600
633
+ },
634
+ {
635
+ "epoch": 0.75,
636
+ "learning_rate": 5.788566953797963e-05,
637
+ "loss": 3.2075,
638
+ "step": 1620
639
+ },
640
+ {
641
+ "epoch": 0.75,
642
+ "learning_rate": 5.765074393108848e-05,
643
+ "loss": 3.1772,
644
+ "step": 1640
645
+ },
646
+ {
647
+ "epoch": 0.76,
648
+ "learning_rate": 5.742756460454189e-05,
649
+ "loss": 3.2101,
650
+ "step": 1660
651
+ },
652
+ {
653
+ "epoch": 0.77,
654
+ "learning_rate": 5.7192638997650736e-05,
655
+ "loss": 3.2115,
656
+ "step": 1680
657
+ },
658
+ {
659
+ "epoch": 0.78,
660
+ "learning_rate": 5.69812059514487e-05,
661
+ "loss": 3.2986,
662
+ "step": 1700
663
+ },
664
+ {
665
+ "epoch": 0.78,
666
+ "eval_loss": 3.092197895050049,
667
+ "eval_runtime": 194.3305,
668
+ "eval_samples_per_second": 24.921,
669
+ "eval_steps_per_second": 0.782,
670
+ "eval_wer": 1.0,
671
+ "step": 1700
672
+ },
673
+ {
674
+ "epoch": 0.79,
675
+ "learning_rate": 5.674628034455755e-05,
676
+ "loss": 3.2063,
677
+ "step": 1720
678
+ },
679
+ {
680
+ "epoch": 0.8,
681
+ "learning_rate": 5.65113547376664e-05,
682
+ "loss": 3.1723,
683
+ "step": 1740
684
+ },
685
+ {
686
+ "epoch": 0.81,
687
+ "learning_rate": 5.627642913077525e-05,
688
+ "loss": 3.0945,
689
+ "step": 1760
690
+ },
691
+ {
692
+ "epoch": 0.82,
693
+ "learning_rate": 5.6041503523884095e-05,
694
+ "loss": 3.2153,
695
+ "step": 1780
696
+ },
697
+ {
698
+ "epoch": 0.83,
699
+ "learning_rate": 5.580657791699295e-05,
700
+ "loss": 3.0761,
701
+ "step": 1800
702
+ },
703
+ {
704
+ "epoch": 0.83,
705
+ "eval_loss": 3.092197895050049,
706
+ "eval_runtime": 192.7837,
707
+ "eval_samples_per_second": 25.121,
708
+ "eval_steps_per_second": 0.788,
709
+ "eval_wer": 1.0,
710
+ "step": 1800
711
+ },
712
+ {
713
+ "epoch": 0.84,
714
+ "learning_rate": 5.5571652310101795e-05,
715
+ "loss": 3.2061,
716
+ "step": 1820
717
+ },
718
+ {
719
+ "epoch": 0.85,
720
+ "learning_rate": 5.533672670321064e-05,
721
+ "loss": 3.1768,
722
+ "step": 1840
723
+ },
724
+ {
725
+ "epoch": 0.86,
726
+ "learning_rate": 5.5101801096319495e-05,
727
+ "loss": 3.096,
728
+ "step": 1860
729
+ },
730
+ {
731
+ "epoch": 0.87,
732
+ "learning_rate": 5.486687548942834e-05,
733
+ "loss": 3.213,
734
+ "step": 1880
735
+ },
736
+ {
737
+ "epoch": 0.87,
738
+ "learning_rate": 5.464369616288175e-05,
739
+ "loss": 3.1832,
740
+ "step": 1900
741
+ },
742
+ {
743
+ "epoch": 0.87,
744
+ "eval_loss": 3.092197895050049,
745
+ "eval_runtime": 196.1769,
746
+ "eval_samples_per_second": 24.687,
747
+ "eval_steps_per_second": 0.775,
748
+ "eval_wer": 1.0,
749
+ "step": 1900
750
+ },
751
+ {
752
+ "epoch": 0.88,
753
+ "learning_rate": 5.44087705559906e-05,
754
+ "loss": 3.2046,
755
+ "step": 1920
756
+ },
757
+ {
758
+ "epoch": 0.89,
759
+ "learning_rate": 5.417384494909945e-05,
760
+ "loss": 3.1743,
761
+ "step": 1940
762
+ },
763
+ {
764
+ "epoch": 0.9,
765
+ "learning_rate": 5.395066562255285e-05,
766
+ "loss": 3.2055,
767
+ "step": 1960
768
+ },
769
+ {
770
+ "epoch": 0.91,
771
+ "learning_rate": 5.37157400156617e-05,
772
+ "loss": 3.2152,
773
+ "step": 1980
774
+ },
775
+ {
776
+ "epoch": 0.92,
777
+ "learning_rate": 5.349256068911511e-05,
778
+ "loss": 3.1845,
779
+ "step": 2000
780
+ },
781
+ {
782
+ "epoch": 0.92,
783
+ "eval_loss": 3.092197895050049,
784
+ "eval_runtime": 195.9315,
785
+ "eval_samples_per_second": 24.718,
786
+ "eval_steps_per_second": 0.776,
787
+ "eval_wer": 1.0,
788
+ "step": 2000
789
+ },
790
+ {
791
+ "epoch": 0.93,
792
+ "learning_rate": 5.3257635082223954e-05,
793
+ "loss": 3.2093,
794
+ "step": 2020
795
+ },
796
+ {
797
+ "epoch": 0.94,
798
+ "learning_rate": 5.302270947533281e-05,
799
+ "loss": 3.1748,
800
+ "step": 2040
801
+ },
802
+ {
803
+ "epoch": 0.95,
804
+ "learning_rate": 5.279953014878621e-05,
805
+ "loss": 3.199,
806
+ "step": 2060
807
+ },
808
+ {
809
+ "epoch": 0.96,
810
+ "learning_rate": 5.2564604541895067e-05,
811
+ "loss": 3.2169,
812
+ "step": 2080
813
+ },
814
+ {
815
+ "epoch": 0.97,
816
+ "learning_rate": 5.2341425215348466e-05,
817
+ "loss": 3.1817,
818
+ "step": 2100
819
+ },
820
+ {
821
+ "epoch": 0.97,
822
+ "eval_loss": 3.092197895050049,
823
+ "eval_runtime": 193.8566,
824
+ "eval_samples_per_second": 24.982,
825
+ "eval_steps_per_second": 0.784,
826
+ "eval_wer": 1.0,
827
+ "step": 2100
828
+ },
829
+ {
830
+ "epoch": 0.98,
831
+ "learning_rate": 5.210649960845731e-05,
832
+ "loss": 3.2075,
833
+ "step": 2120
834
+ },
835
+ {
836
+ "epoch": 0.98,
837
+ "learning_rate": 5.1871574001566166e-05,
838
+ "loss": 3.1693,
839
+ "step": 2140
840
+ },
841
+ {
842
+ "epoch": 0.99,
843
+ "learning_rate": 5.164839467501957e-05,
844
+ "loss": 3.2049,
845
+ "step": 2160
846
+ },
847
+ {
848
+ "epoch": 1.0,
849
+ "learning_rate": 5.1425215348472984e-05,
850
+ "loss": 3.3657,
851
+ "step": 2180
852
+ },
853
+ {
854
+ "epoch": 1.01,
855
+ "learning_rate": 5.119028974158183e-05,
856
+ "loss": 3.2209,
857
+ "step": 2200
858
+ },
859
+ {
860
+ "epoch": 1.01,
861
+ "eval_loss": 3.092197895050049,
862
+ "eval_runtime": 194.751,
863
+ "eval_samples_per_second": 24.868,
864
+ "eval_steps_per_second": 0.78,
865
+ "eval_wer": 1.0,
866
+ "step": 2200
867
+ },
868
+ {
869
+ "epoch": 1.02,
870
+ "learning_rate": 5.095536413469068e-05,
871
+ "loss": 3.0961,
872
+ "step": 2220
873
+ },
874
+ {
875
+ "epoch": 1.03,
876
+ "learning_rate": 5.073218480814408e-05,
877
+ "loss": 3.2735,
878
+ "step": 2240
879
+ },
880
+ {
881
+ "epoch": 1.04,
882
+ "learning_rate": 5.049725920125293e-05,
883
+ "loss": 3.1904,
884
+ "step": 2260
885
+ },
886
+ {
887
+ "epoch": 1.05,
888
+ "learning_rate": 5.0274079874706336e-05,
889
+ "loss": 3.1892,
890
+ "step": 2280
891
+ },
892
+ {
893
+ "epoch": 1.06,
894
+ "learning_rate": 5.003915426781519e-05,
895
+ "loss": 3.2138,
896
+ "step": 2300
897
+ },
898
+ {
899
+ "epoch": 1.06,
900
+ "eval_loss": 3.092197895050049,
901
+ "eval_runtime": 195.285,
902
+ "eval_samples_per_second": 24.8,
903
+ "eval_steps_per_second": 0.778,
904
+ "eval_wer": 1.0,
905
+ "step": 2300
906
+ },
907
+ {
908
+ "epoch": 1.07,
909
+ "learning_rate": 4.980422866092404e-05,
910
+ "loss": 3.0819,
911
+ "step": 2320
912
+ },
913
+ {
914
+ "epoch": 1.08,
915
+ "learning_rate": 4.958104933437744e-05,
916
+ "loss": 3.272,
917
+ "step": 2340
918
+ },
919
+ {
920
+ "epoch": 1.09,
921
+ "learning_rate": 4.9346123727486296e-05,
922
+ "loss": 3.2011,
923
+ "step": 2360
924
+ },
925
+ {
926
+ "epoch": 1.1,
927
+ "learning_rate": 4.9122944400939695e-05,
928
+ "loss": 3.1918,
929
+ "step": 2380
930
+ },
931
+ {
932
+ "epoch": 1.1,
933
+ "learning_rate": 4.888801879404854e-05,
934
+ "loss": 3.2154,
935
+ "step": 2400
936
+ },
937
+ {
938
+ "epoch": 1.1,
939
+ "eval_loss": 3.092197895050049,
940
+ "eval_runtime": 196.4218,
941
+ "eval_samples_per_second": 24.656,
942
+ "eval_steps_per_second": 0.774,
943
+ "eval_wer": 1.0,
944
+ "step": 2400
945
+ },
946
+ {
947
+ "epoch": 1.11,
948
+ "learning_rate": 4.8653093187157396e-05,
949
+ "loss": 3.0946,
950
+ "step": 2420
951
+ },
952
+ {
953
+ "epoch": 1.12,
954
+ "learning_rate": 4.84299138606108e-05,
955
+ "loss": 3.2753,
956
+ "step": 2440
957
+ },
958
+ {
959
+ "epoch": 1.13,
960
+ "learning_rate": 4.8194988253719655e-05,
961
+ "loss": 3.1829,
962
+ "step": 2460
963
+ },
964
+ {
965
+ "epoch": 1.14,
966
+ "learning_rate": 4.797180892717306e-05,
967
+ "loss": 3.1839,
968
+ "step": 2480
969
+ },
970
+ {
971
+ "epoch": 1.15,
972
+ "learning_rate": 4.773688332028191e-05,
973
+ "loss": 3.2185,
974
+ "step": 2500
975
+ },
976
+ {
977
+ "epoch": 1.15,
978
+ "eval_loss": 3.092197895050049,
979
+ "eval_runtime": 192.8929,
980
+ "eval_samples_per_second": 25.107,
981
+ "eval_steps_per_second": 0.788,
982
+ "eval_wer": 1.0,
983
+ "step": 2500
984
+ },
985
+ {
986
+ "epoch": 1.16,
987
+ "learning_rate": 4.750195771339076e-05,
988
+ "loss": 3.0971,
989
+ "step": 2520
990
+ },
991
+ {
992
+ "epoch": 1.17,
993
+ "learning_rate": 4.727877838684416e-05,
994
+ "loss": 3.2719,
995
+ "step": 2540
996
+ },
997
+ {
998
+ "epoch": 1.18,
999
+ "learning_rate": 4.704385277995301e-05,
1000
+ "loss": 3.1974,
1001
+ "step": 2560
1002
+ },
1003
+ {
1004
+ "epoch": 1.19,
1005
+ "learning_rate": 4.680892717306186e-05,
1006
+ "loss": 3.0775,
1007
+ "step": 2580
1008
+ },
1009
+ {
1010
+ "epoch": 1.2,
1011
+ "learning_rate": 4.657400156617071e-05,
1012
+ "loss": 3.2201,
1013
+ "step": 2600
1014
+ },
1015
+ {
1016
+ "epoch": 1.2,
1017
+ "eval_loss": 3.092197895050049,
1018
+ "eval_runtime": 192.9488,
1019
+ "eval_samples_per_second": 25.1,
1020
+ "eval_steps_per_second": 0.788,
1021
+ "eval_wer": 1.0,
1022
+ "step": 2600
1023
+ },
1024
+ {
1025
+ "epoch": 1.21,
1026
+ "learning_rate": 4.6339075959279554e-05,
1027
+ "loss": 3.0946,
1028
+ "step": 2620
1029
+ },
1030
+ {
1031
+ "epoch": 1.22,
1032
+ "learning_rate": 4.611589663273297e-05,
1033
+ "loss": 3.2708,
1034
+ "step": 2640
1035
+ },
1036
+ {
1037
+ "epoch": 1.22,
1038
+ "learning_rate": 4.5880971025841814e-05,
1039
+ "loss": 3.1949,
1040
+ "step": 2660
1041
+ },
1042
+ {
1043
+ "epoch": 1.23,
1044
+ "learning_rate": 4.564604541895066e-05,
1045
+ "loss": 3.0763,
1046
+ "step": 2680
1047
+ },
1048
+ {
1049
+ "epoch": 1.24,
1050
+ "learning_rate": 4.5411119812059514e-05,
1051
+ "loss": 3.2184,
1052
+ "step": 2700
1053
+ },
1054
+ {
1055
+ "epoch": 1.24,
1056
+ "eval_loss": 3.092197895050049,
1057
+ "eval_runtime": 189.7646,
1058
+ "eval_samples_per_second": 25.521,
1059
+ "eval_steps_per_second": 0.801,
1060
+ "eval_wer": 1.0,
1061
+ "step": 2700
1062
+ },
1063
+ {
1064
+ "epoch": 1.25,
1065
+ "learning_rate": 4.517619420516836e-05,
1066
+ "loss": 3.0962,
1067
+ "step": 2720
1068
+ },
1069
+ {
1070
+ "epoch": 1.26,
1071
+ "learning_rate": 4.495301487862176e-05,
1072
+ "loss": 3.2795,
1073
+ "step": 2740
1074
+ },
1075
+ {
1076
+ "epoch": 1.27,
1077
+ "learning_rate": 4.4718089271730614e-05,
1078
+ "loss": 3.1901,
1079
+ "step": 2760
1080
+ },
1081
+ {
1082
+ "epoch": 1.28,
1083
+ "learning_rate": 4.449490994518402e-05,
1084
+ "loss": 3.1902,
1085
+ "step": 2780
1086
+ },
1087
+ {
1088
+ "epoch": 1.29,
1089
+ "learning_rate": 4.425998433829287e-05,
1090
+ "loss": 3.2196,
1091
+ "step": 2800
1092
+ },
1093
+ {
1094
+ "epoch": 1.29,
1095
+ "eval_loss": 3.092197895050049,
1096
+ "eval_runtime": 192.1661,
1097
+ "eval_samples_per_second": 25.202,
1098
+ "eval_steps_per_second": 0.791,
1099
+ "eval_wer": 1.0,
1100
+ "step": 2800
1101
+ },
1102
+ {
1103
+ "epoch": 1.3,
1104
+ "learning_rate": 4.402505873140172e-05,
1105
+ "loss": 3.0923,
1106
+ "step": 2820
1107
+ },
1108
+ {
1109
+ "epoch": 1.31,
1110
+ "learning_rate": 4.3801879404855126e-05,
1111
+ "loss": 3.2783,
1112
+ "step": 2840
1113
+ },
1114
+ {
1115
+ "epoch": 1.32,
1116
+ "learning_rate": 4.356695379796398e-05,
1117
+ "loss": 3.195,
1118
+ "step": 2860
1119
+ },
1120
+ {
1121
+ "epoch": 1.33,
1122
+ "learning_rate": 4.335552075176194e-05,
1123
+ "loss": 3.3041,
1124
+ "step": 2880
1125
+ },
1126
+ {
1127
+ "epoch": 1.34,
1128
+ "learning_rate": 4.3120595144870784e-05,
1129
+ "loss": 3.2158,
1130
+ "step": 2900
1131
+ },
1132
+ {
1133
+ "epoch": 1.34,
1134
+ "eval_loss": 3.092197895050049,
1135
+ "eval_runtime": 194.7934,
1136
+ "eval_samples_per_second": 24.862,
1137
+ "eval_steps_per_second": 0.78,
1138
+ "eval_wer": 1.0,
1139
+ "step": 2900
1140
+ },
1141
+ {
1142
+ "epoch": 1.34,
1143
+ "learning_rate": 4.288566953797964e-05,
1144
+ "loss": 3.0976,
1145
+ "step": 2920
1146
+ },
1147
+ {
1148
+ "epoch": 1.35,
1149
+ "learning_rate": 4.2662490211433043e-05,
1150
+ "loss": 3.2737,
1151
+ "step": 2940
1152
+ },
1153
+ {
1154
+ "epoch": 1.36,
1155
+ "learning_rate": 4.24275646045419e-05,
1156
+ "loss": 3.1927,
1157
+ "step": 2960
1158
+ },
1159
+ {
1160
+ "epoch": 1.37,
1161
+ "learning_rate": 4.2204385277995296e-05,
1162
+ "loss": 3.1896,
1163
+ "step": 2980
1164
+ },
1165
+ {
1166
+ "epoch": 1.38,
1167
+ "learning_rate": 4.196945967110414e-05,
1168
+ "loss": 3.2169,
1169
+ "step": 3000
1170
+ },
1171
+ {
1172
+ "epoch": 1.38,
1173
+ "eval_loss": 3.092197895050049,
1174
+ "eval_runtime": 196.5983,
1175
+ "eval_samples_per_second": 24.634,
1176
+ "eval_steps_per_second": 0.773,
1177
+ "eval_wer": 1.0,
1178
+ "step": 3000
1179
+ }
1180
+ ],
1181
+ "max_steps": 6516,
1182
+ "num_train_epochs": 3,
1183
+ "total_flos": 5.730765756456842e+19,
1184
+ "trial_name": null,
1185
+ "trial_params": null
1186
+ }
{checkpoint-700 → checkpoint-3000}/training_args.bin RENAMED
File without changes
checkpoint-600/rng_state.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:796699f569c9be64a692307d8264b44b7adbdf8ac3e91d99def5b22f4808095a
3
- size 14567
 
 
 
 
checkpoint-600/scaler.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:d59c7740536b9634042b591d2769580fe65377ef904907588872cde90375d0b3
3
- size 559
 
 
 
 
checkpoint-600/trainer_state.json DELETED
@@ -1,250 +0,0 @@
1
- {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 0.2761477390403866,
5
- "global_step": 600,
6
- "is_hyper_param_search": false,
7
- "is_local_process_zero": true,
8
- "is_world_process_zero": true,
9
- "log_history": [
10
- {
11
- "epoch": 0.01,
12
- "learning_rate": 1.0305343511450381e-05,
13
- "loss": 13.0539,
14
- "step": 20
15
- },
16
- {
17
- "epoch": 0.02,
18
- "learning_rate": 2.118320610687023e-05,
19
- "loss": 20.9649,
20
- "step": 40
21
- },
22
- {
23
- "epoch": 0.03,
24
- "learning_rate": 3.206106870229007e-05,
25
- "loss": 21.1001,
26
- "step": 60
27
- },
28
- {
29
- "epoch": 0.04,
30
- "learning_rate": 4.3511450381679383e-05,
31
- "loss": 5.5852,
32
- "step": 80
33
- },
34
- {
35
- "epoch": 0.05,
36
- "learning_rate": 5.438931297709923e-05,
37
- "loss": 5.2002,
38
- "step": 100
39
- },
40
- {
41
- "epoch": 0.05,
42
- "eval_loss": 3.5950469970703125,
43
- "eval_runtime": 194.8685,
44
- "eval_samples_per_second": 24.853,
45
- "eval_steps_per_second": 0.78,
46
- "eval_wer": 1.0,
47
- "step": 100
48
- },
49
- {
50
- "epoch": 0.06,
51
- "learning_rate": 6.583969465648854e-05,
52
- "loss": 3.281,
53
- "step": 120
54
- },
55
- {
56
- "epoch": 0.06,
57
- "learning_rate": 7.495301487862176e-05,
58
- "loss": 3.3442,
59
- "step": 140
60
- },
61
- {
62
- "epoch": 0.07,
63
- "learning_rate": 7.472983555207517e-05,
64
- "loss": 3.5229,
65
- "step": 160
66
- },
67
- {
68
- "epoch": 0.08,
69
- "learning_rate": 7.449490994518402e-05,
70
- "loss": 3.089,
71
- "step": 180
72
- },
73
- {
74
- "epoch": 0.09,
75
- "learning_rate": 7.427173061863743e-05,
76
- "loss": 3.482,
77
- "step": 200
78
- },
79
- {
80
- "epoch": 0.09,
81
- "eval_loss": 3.274627923965454,
82
- "eval_runtime": 194.5649,
83
- "eval_samples_per_second": 24.891,
84
- "eval_steps_per_second": 0.781,
85
- "eval_wer": 1.0,
86
- "step": 200
87
- },
88
- {
89
- "epoch": 0.1,
90
- "learning_rate": 7.403680501174628e-05,
91
- "loss": 3.1517,
92
- "step": 220
93
- },
94
- {
95
- "epoch": 0.11,
96
- "learning_rate": 7.380187940485512e-05,
97
- "loss": 3.1617,
98
- "step": 240
99
- },
100
- {
101
- "epoch": 0.12,
102
- "learning_rate": 7.357870007830852e-05,
103
- "loss": 3.2971,
104
- "step": 260
105
- },
106
- {
107
- "epoch": 0.13,
108
- "learning_rate": 7.334377447141738e-05,
109
- "loss": 3.022,
110
- "step": 280
111
- },
112
- {
113
- "epoch": 0.14,
114
- "learning_rate": 7.312059514487078e-05,
115
- "loss": 3.3322,
116
- "step": 300
117
- },
118
- {
119
- "epoch": 0.14,
120
- "eval_loss": 3.0716073513031006,
121
- "eval_runtime": 196.1674,
122
- "eval_samples_per_second": 24.688,
123
- "eval_steps_per_second": 0.775,
124
- "eval_wer": 1.0,
125
- "step": 300
126
- },
127
- {
128
- "epoch": 0.15,
129
- "learning_rate": 7.288566953797964e-05,
130
- "loss": 2.9964,
131
- "step": 320
132
- },
133
- {
134
- "epoch": 0.16,
135
- "learning_rate": 7.265074393108849e-05,
136
- "loss": 3.0777,
137
- "step": 340
138
- },
139
- {
140
- "epoch": 0.17,
141
- "learning_rate": 7.242756460454189e-05,
142
- "loss": 3.2703,
143
- "step": 360
144
- },
145
- {
146
- "epoch": 0.17,
147
- "learning_rate": 7.219263899765073e-05,
148
- "loss": 3.1161,
149
- "step": 380
150
- },
151
- {
152
- "epoch": 0.18,
153
- "learning_rate": 7.195771339075958e-05,
154
- "loss": 3.2435,
155
- "step": 400
156
- },
157
- {
158
- "epoch": 0.18,
159
- "eval_loss": 3.063333034515381,
160
- "eval_runtime": 190.5938,
161
- "eval_samples_per_second": 25.41,
162
- "eval_steps_per_second": 0.798,
163
- "eval_wer": 1.0,
164
- "step": 400
165
- },
166
- {
167
- "epoch": 0.19,
168
- "learning_rate": 7.172278778386843e-05,
169
- "loss": 2.9993,
170
- "step": 420
171
- },
172
- {
173
- "epoch": 0.2,
174
- "learning_rate": 7.148786217697727e-05,
175
- "loss": 3.0555,
176
- "step": 440
177
- },
178
- {
179
- "epoch": 0.21,
180
- "learning_rate": 7.126468285043069e-05,
181
- "loss": 3.1879,
182
- "step": 460
183
- },
184
- {
185
- "epoch": 0.22,
186
- "learning_rate": 7.102975724353953e-05,
187
- "loss": 2.9861,
188
- "step": 480
189
- },
190
- {
191
- "epoch": 0.23,
192
- "learning_rate": 7.080657791699295e-05,
193
- "loss": 3.0833,
194
- "step": 500
195
- },
196
- {
197
- "epoch": 0.23,
198
- "eval_loss": 2.9769718647003174,
199
- "eval_runtime": 195.2639,
200
- "eval_samples_per_second": 24.802,
201
- "eval_steps_per_second": 0.778,
202
- "eval_wer": 1.0,
203
- "step": 500
204
- },
205
- {
206
- "epoch": 0.24,
207
- "learning_rate": 7.05716523101018e-05,
208
- "loss": 3.1608,
209
- "step": 520
210
- },
211
- {
212
- "epoch": 0.25,
213
- "learning_rate": 7.033672670321064e-05,
214
- "loss": 3.0206,
215
- "step": 540
216
- },
217
- {
218
- "epoch": 0.26,
219
- "learning_rate": 7.01018010963195e-05,
220
- "loss": 3.0523,
221
- "step": 560
222
- },
223
- {
224
- "epoch": 0.27,
225
- "learning_rate": 6.986687548942835e-05,
226
- "loss": 3.0222,
227
- "step": 580
228
- },
229
- {
230
- "epoch": 0.28,
231
- "learning_rate": 6.964369616288175e-05,
232
- "loss": 3.0913,
233
- "step": 600
234
- },
235
- {
236
- "epoch": 0.28,
237
- "eval_loss": 3.052091598510742,
238
- "eval_runtime": 189.8149,
239
- "eval_samples_per_second": 25.514,
240
- "eval_steps_per_second": 0.801,
241
- "eval_wer": 1.0,
242
- "step": 600
243
- }
244
- ],
245
- "max_steps": 6516,
246
- "num_train_epochs": 3,
247
- "total_flos": 1.1416289792484741e+19,
248
- "trial_name": null,
249
- "trial_params": null
250
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
checkpoint-700/scheduler.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:e5e288f5acef38cf681fd34cf5dbd19393d9b7a69196a87d07c084d3bc82290e
3
- size 623
 
 
 
 
checkpoint-700/trainer_state.json DELETED
@@ -1,289 +0,0 @@
1
- {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 0.32217236221378437,
5
- "global_step": 700,
6
- "is_hyper_param_search": false,
7
- "is_local_process_zero": true,
8
- "is_world_process_zero": true,
9
- "log_history": [
10
- {
11
- "epoch": 0.01,
12
- "learning_rate": 1.0305343511450381e-05,
13
- "loss": 13.0539,
14
- "step": 20
15
- },
16
- {
17
- "epoch": 0.02,
18
- "learning_rate": 2.118320610687023e-05,
19
- "loss": 20.9649,
20
- "step": 40
21
- },
22
- {
23
- "epoch": 0.03,
24
- "learning_rate": 3.206106870229007e-05,
25
- "loss": 21.1001,
26
- "step": 60
27
- },
28
- {
29
- "epoch": 0.04,
30
- "learning_rate": 4.3511450381679383e-05,
31
- "loss": 5.5852,
32
- "step": 80
33
- },
34
- {
35
- "epoch": 0.05,
36
- "learning_rate": 5.438931297709923e-05,
37
- "loss": 5.2002,
38
- "step": 100
39
- },
40
- {
41
- "epoch": 0.05,
42
- "eval_loss": 3.5950469970703125,
43
- "eval_runtime": 194.8685,
44
- "eval_samples_per_second": 24.853,
45
- "eval_steps_per_second": 0.78,
46
- "eval_wer": 1.0,
47
- "step": 100
48
- },
49
- {
50
- "epoch": 0.06,
51
- "learning_rate": 6.583969465648854e-05,
52
- "loss": 3.281,
53
- "step": 120
54
- },
55
- {
56
- "epoch": 0.06,
57
- "learning_rate": 7.495301487862176e-05,
58
- "loss": 3.3442,
59
- "step": 140
60
- },
61
- {
62
- "epoch": 0.07,
63
- "learning_rate": 7.472983555207517e-05,
64
- "loss": 3.5229,
65
- "step": 160
66
- },
67
- {
68
- "epoch": 0.08,
69
- "learning_rate": 7.449490994518402e-05,
70
- "loss": 3.089,
71
- "step": 180
72
- },
73
- {
74
- "epoch": 0.09,
75
- "learning_rate": 7.427173061863743e-05,
76
- "loss": 3.482,
77
- "step": 200
78
- },
79
- {
80
- "epoch": 0.09,
81
- "eval_loss": 3.274627923965454,
82
- "eval_runtime": 194.5649,
83
- "eval_samples_per_second": 24.891,
84
- "eval_steps_per_second": 0.781,
85
- "eval_wer": 1.0,
86
- "step": 200
87
- },
88
- {
89
- "epoch": 0.1,
90
- "learning_rate": 7.403680501174628e-05,
91
- "loss": 3.1517,
92
- "step": 220
93
- },
94
- {
95
- "epoch": 0.11,
96
- "learning_rate": 7.380187940485512e-05,
97
- "loss": 3.1617,
98
- "step": 240
99
- },
100
- {
101
- "epoch": 0.12,
102
- "learning_rate": 7.357870007830852e-05,
103
- "loss": 3.2971,
104
- "step": 260
105
- },
106
- {
107
- "epoch": 0.13,
108
- "learning_rate": 7.334377447141738e-05,
109
- "loss": 3.022,
110
- "step": 280
111
- },
112
- {
113
- "epoch": 0.14,
114
- "learning_rate": 7.312059514487078e-05,
115
- "loss": 3.3322,
116
- "step": 300
117
- },
118
- {
119
- "epoch": 0.14,
120
- "eval_loss": 3.0716073513031006,
121
- "eval_runtime": 196.1674,
122
- "eval_samples_per_second": 24.688,
123
- "eval_steps_per_second": 0.775,
124
- "eval_wer": 1.0,
125
- "step": 300
126
- },
127
- {
128
- "epoch": 0.15,
129
- "learning_rate": 7.288566953797964e-05,
130
- "loss": 2.9964,
131
- "step": 320
132
- },
133
- {
134
- "epoch": 0.16,
135
- "learning_rate": 7.265074393108849e-05,
136
- "loss": 3.0777,
137
- "step": 340
138
- },
139
- {
140
- "epoch": 0.17,
141
- "learning_rate": 7.242756460454189e-05,
142
- "loss": 3.2703,
143
- "step": 360
144
- },
145
- {
146
- "epoch": 0.17,
147
- "learning_rate": 7.219263899765073e-05,
148
- "loss": 3.1161,
149
- "step": 380
150
- },
151
- {
152
- "epoch": 0.18,
153
- "learning_rate": 7.195771339075958e-05,
154
- "loss": 3.2435,
155
- "step": 400
156
- },
157
- {
158
- "epoch": 0.18,
159
- "eval_loss": 3.063333034515381,
160
- "eval_runtime": 190.5938,
161
- "eval_samples_per_second": 25.41,
162
- "eval_steps_per_second": 0.798,
163
- "eval_wer": 1.0,
164
- "step": 400
165
- },
166
- {
167
- "epoch": 0.19,
168
- "learning_rate": 7.172278778386843e-05,
169
- "loss": 2.9993,
170
- "step": 420
171
- },
172
- {
173
- "epoch": 0.2,
174
- "learning_rate": 7.148786217697727e-05,
175
- "loss": 3.0555,
176
- "step": 440
177
- },
178
- {
179
- "epoch": 0.21,
180
- "learning_rate": 7.126468285043069e-05,
181
- "loss": 3.1879,
182
- "step": 460
183
- },
184
- {
185
- "epoch": 0.22,
186
- "learning_rate": 7.102975724353953e-05,
187
- "loss": 2.9861,
188
- "step": 480
189
- },
190
- {
191
- "epoch": 0.23,
192
- "learning_rate": 7.080657791699295e-05,
193
- "loss": 3.0833,
194
- "step": 500
195
- },
196
- {
197
- "epoch": 0.23,
198
- "eval_loss": 2.9769718647003174,
199
- "eval_runtime": 195.2639,
200
- "eval_samples_per_second": 24.802,
201
- "eval_steps_per_second": 0.778,
202
- "eval_wer": 1.0,
203
- "step": 500
204
- },
205
- {
206
- "epoch": 0.24,
207
- "learning_rate": 7.05716523101018e-05,
208
- "loss": 3.1608,
209
- "step": 520
210
- },
211
- {
212
- "epoch": 0.25,
213
- "learning_rate": 7.033672670321064e-05,
214
- "loss": 3.0206,
215
- "step": 540
216
- },
217
- {
218
- "epoch": 0.26,
219
- "learning_rate": 7.01018010963195e-05,
220
- "loss": 3.0523,
221
- "step": 560
222
- },
223
- {
224
- "epoch": 0.27,
225
- "learning_rate": 6.986687548942835e-05,
226
- "loss": 3.0222,
227
- "step": 580
228
- },
229
- {
230
- "epoch": 0.28,
231
- "learning_rate": 6.964369616288175e-05,
232
- "loss": 3.0913,
233
- "step": 600
234
- },
235
- {
236
- "epoch": 0.28,
237
- "eval_loss": 3.052091598510742,
238
- "eval_runtime": 189.8149,
239
- "eval_samples_per_second": 25.514,
240
- "eval_steps_per_second": 0.801,
241
- "eval_wer": 1.0,
242
- "step": 600
243
- },
244
- {
245
- "epoch": 0.29,
246
- "learning_rate": 6.94087705559906e-05,
247
- "loss": 3.1987,
248
- "step": 620
249
- },
250
- {
251
- "epoch": 0.29,
252
- "learning_rate": 6.917384494909944e-05,
253
- "loss": 3.1504,
254
- "step": 640
255
- },
256
- {
257
- "epoch": 0.3,
258
- "learning_rate": 6.893891934220829e-05,
259
- "loss": 3.2406,
260
- "step": 660
261
- },
262
- {
263
- "epoch": 0.31,
264
- "learning_rate": 6.870399373531715e-05,
265
- "loss": 3.0632,
266
- "step": 680
267
- },
268
- {
269
- "epoch": 0.32,
270
- "learning_rate": 6.848081440877055e-05,
271
- "loss": 3.1487,
272
- "step": 700
273
- },
274
- {
275
- "epoch": 0.32,
276
- "eval_loss": 3.1951305866241455,
277
- "eval_runtime": 196.8636,
278
- "eval_samples_per_second": 24.601,
279
- "eval_steps_per_second": 0.772,
280
- "eval_wer": 1.0,
281
- "step": 700
282
- }
283
- ],
284
- "max_steps": 6516,
285
- "num_train_epochs": 3,
286
- "total_flos": 1.3316117705308692e+19,
287
- "trial_name": null,
288
- "trial_params": null
289
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
checkpoint-800/config.json DELETED
@@ -1,115 +0,0 @@
1
- {
2
- "_name_or_path": "KBLab/wav2vec2-large-voxrex",
3
- "activation_dropout": 0.1,
4
- "adapter_kernel_size": 3,
5
- "adapter_stride": 2,
6
- "add_adapter": false,
7
- "apply_spec_augment": true,
8
- "architectures": [
9
- "Wav2Vec2ForCTC"
10
- ],
11
- "attention_dropout": 0.0,
12
- "bos_token_id": 1,
13
- "classifier_proj_size": 256,
14
- "codevector_dim": 768,
15
- "contrastive_logits_temperature": 0.1,
16
- "conv_bias": true,
17
- "conv_dim": [
18
- 512,
19
- 512,
20
- 512,
21
- 512,
22
- 512,
23
- 512,
24
- 512
25
- ],
26
- "conv_kernel": [
27
- 10,
28
- 3,
29
- 3,
30
- 3,
31
- 3,
32
- 2,
33
- 2
34
- ],
35
- "conv_stride": [
36
- 5,
37
- 2,
38
- 2,
39
- 2,
40
- 2,
41
- 2,
42
- 2
43
- ],
44
- "ctc_loss_reduction": "mean",
45
- "ctc_zero_infinity": false,
46
- "diversity_loss_weight": 0.1,
47
- "do_stable_layer_norm": true,
48
- "eos_token_id": 2,
49
- "feat_extract_activation": "gelu",
50
- "feat_extract_dropout": 0.0,
51
- "feat_extract_norm": "layer",
52
- "feat_proj_dropout": 0.0,
53
- "feat_quantizer_dropout": 0.0,
54
- "final_dropout": 0.0,
55
- "hidden_act": "gelu",
56
- "hidden_dropout": 0.0,
57
- "hidden_size": 1024,
58
- "initializer_range": 0.02,
59
- "intermediate_size": 4096,
60
- "layer_norm_eps": 1e-05,
61
- "layerdrop": 0.0,
62
- "mask_channel_length": 10,
63
- "mask_channel_min_space": 1,
64
- "mask_channel_other": 0.0,
65
- "mask_channel_prob": 0.0,
66
- "mask_channel_selection": "static",
67
- "mask_feature_length": 64,
68
- "mask_feature_min_masks": 0,
69
- "mask_feature_prob": 0.25,
70
- "mask_time_length": 10,
71
- "mask_time_min_masks": 2,
72
- "mask_time_min_space": 1,
73
- "mask_time_other": 0.0,
74
- "mask_time_prob": 0.75,
75
- "mask_time_selection": "static",
76
- "model_type": "wav2vec2",
77
- "num_adapter_layers": 3,
78
- "num_attention_heads": 16,
79
- "num_codevector_groups": 2,
80
- "num_codevectors_per_group": 320,
81
- "num_conv_pos_embedding_groups": 16,
82
- "num_conv_pos_embeddings": 128,
83
- "num_feat_extract_layers": 7,
84
- "num_hidden_layers": 24,
85
- "num_negatives": 100,
86
- "output_hidden_size": 1024,
87
- "pad_token_id": 31,
88
- "proj_codevector_dim": 768,
89
- "tdnn_dilation": [
90
- 1,
91
- 2,
92
- 3,
93
- 1,
94
- 1
95
- ],
96
- "tdnn_dim": [
97
- 512,
98
- 512,
99
- 512,
100
- 512,
101
- 1500
102
- ],
103
- "tdnn_kernel": [
104
- 5,
105
- 3,
106
- 3,
107
- 1,
108
- 1
109
- ],
110
- "torch_dtype": "float32",
111
- "transformers_version": "4.17.0.dev0",
112
- "use_weighted_layer_sum": false,
113
- "vocab_size": 34,
114
- "xvector_output_dim": 512
115
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
checkpoint-800/optimizer.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:58d40f21a27bc7f1923c7d04f34f438f664d524cd4a03d73ddd945feffa415cb
3
- size 2490337809
 
 
 
 
checkpoint-800/preprocessor_config.json DELETED
@@ -1,9 +0,0 @@
1
- {
2
- "do_normalize": true,
3
- "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
- "feature_size": 1,
5
- "padding_side": "right",
6
- "padding_value": 0,
7
- "return_attention_mask": true,
8
- "sampling_rate": 16000
9
- }
 
 
 
 
 
 
 
 
 
 
checkpoint-800/pytorch_model.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:4cfc19f8f8b247c6b5050f32c8b2bf68b9bdc2656dafb45faf51286ab02eec5e
3
- size 1262063089
 
 
 
 
checkpoint-800/trainer_state.json DELETED
@@ -1,328 +0,0 @@
1
- {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 0.36819698538718215,
5
- "global_step": 800,
6
- "is_hyper_param_search": false,
7
- "is_local_process_zero": true,
8
- "is_world_process_zero": true,
9
- "log_history": [
10
- {
11
- "epoch": 0.01,
12
- "learning_rate": 1.0305343511450381e-05,
13
- "loss": 13.0539,
14
- "step": 20
15
- },
16
- {
17
- "epoch": 0.02,
18
- "learning_rate": 2.118320610687023e-05,
19
- "loss": 20.9649,
20
- "step": 40
21
- },
22
- {
23
- "epoch": 0.03,
24
- "learning_rate": 3.206106870229007e-05,
25
- "loss": 21.1001,
26
- "step": 60
27
- },
28
- {
29
- "epoch": 0.04,
30
- "learning_rate": 4.3511450381679383e-05,
31
- "loss": 5.5852,
32
- "step": 80
33
- },
34
- {
35
- "epoch": 0.05,
36
- "learning_rate": 5.438931297709923e-05,
37
- "loss": 5.2002,
38
- "step": 100
39
- },
40
- {
41
- "epoch": 0.05,
42
- "eval_loss": 3.5950469970703125,
43
- "eval_runtime": 194.8685,
44
- "eval_samples_per_second": 24.853,
45
- "eval_steps_per_second": 0.78,
46
- "eval_wer": 1.0,
47
- "step": 100
48
- },
49
- {
50
- "epoch": 0.06,
51
- "learning_rate": 6.583969465648854e-05,
52
- "loss": 3.281,
53
- "step": 120
54
- },
55
- {
56
- "epoch": 0.06,
57
- "learning_rate": 7.495301487862176e-05,
58
- "loss": 3.3442,
59
- "step": 140
60
- },
61
- {
62
- "epoch": 0.07,
63
- "learning_rate": 7.472983555207517e-05,
64
- "loss": 3.5229,
65
- "step": 160
66
- },
67
- {
68
- "epoch": 0.08,
69
- "learning_rate": 7.449490994518402e-05,
70
- "loss": 3.089,
71
- "step": 180
72
- },
73
- {
74
- "epoch": 0.09,
75
- "learning_rate": 7.427173061863743e-05,
76
- "loss": 3.482,
77
- "step": 200
78
- },
79
- {
80
- "epoch": 0.09,
81
- "eval_loss": 3.274627923965454,
82
- "eval_runtime": 194.5649,
83
- "eval_samples_per_second": 24.891,
84
- "eval_steps_per_second": 0.781,
85
- "eval_wer": 1.0,
86
- "step": 200
87
- },
88
- {
89
- "epoch": 0.1,
90
- "learning_rate": 7.403680501174628e-05,
91
- "loss": 3.1517,
92
- "step": 220
93
- },
94
- {
95
- "epoch": 0.11,
96
- "learning_rate": 7.380187940485512e-05,
97
- "loss": 3.1617,
98
- "step": 240
99
- },
100
- {
101
- "epoch": 0.12,
102
- "learning_rate": 7.357870007830852e-05,
103
- "loss": 3.2971,
104
- "step": 260
105
- },
106
- {
107
- "epoch": 0.13,
108
- "learning_rate": 7.334377447141738e-05,
109
- "loss": 3.022,
110
- "step": 280
111
- },
112
- {
113
- "epoch": 0.14,
114
- "learning_rate": 7.312059514487078e-05,
115
- "loss": 3.3322,
116
- "step": 300
117
- },
118
- {
119
- "epoch": 0.14,
120
- "eval_loss": 3.0716073513031006,
121
- "eval_runtime": 196.1674,
122
- "eval_samples_per_second": 24.688,
123
- "eval_steps_per_second": 0.775,
124
- "eval_wer": 1.0,
125
- "step": 300
126
- },
127
- {
128
- "epoch": 0.15,
129
- "learning_rate": 7.288566953797964e-05,
130
- "loss": 2.9964,
131
- "step": 320
132
- },
133
- {
134
- "epoch": 0.16,
135
- "learning_rate": 7.265074393108849e-05,
136
- "loss": 3.0777,
137
- "step": 340
138
- },
139
- {
140
- "epoch": 0.17,
141
- "learning_rate": 7.242756460454189e-05,
142
- "loss": 3.2703,
143
- "step": 360
144
- },
145
- {
146
- "epoch": 0.17,
147
- "learning_rate": 7.219263899765073e-05,
148
- "loss": 3.1161,
149
- "step": 380
150
- },
151
- {
152
- "epoch": 0.18,
153
- "learning_rate": 7.195771339075958e-05,
154
- "loss": 3.2435,
155
- "step": 400
156
- },
157
- {
158
- "epoch": 0.18,
159
- "eval_loss": 3.063333034515381,
160
- "eval_runtime": 190.5938,
161
- "eval_samples_per_second": 25.41,
162
- "eval_steps_per_second": 0.798,
163
- "eval_wer": 1.0,
164
- "step": 400
165
- },
166
- {
167
- "epoch": 0.19,
168
- "learning_rate": 7.172278778386843e-05,
169
- "loss": 2.9993,
170
- "step": 420
171
- },
172
- {
173
- "epoch": 0.2,
174
- "learning_rate": 7.148786217697727e-05,
175
- "loss": 3.0555,
176
- "step": 440
177
- },
178
- {
179
- "epoch": 0.21,
180
- "learning_rate": 7.126468285043069e-05,
181
- "loss": 3.1879,
182
- "step": 460
183
- },
184
- {
185
- "epoch": 0.22,
186
- "learning_rate": 7.102975724353953e-05,
187
- "loss": 2.9861,
188
- "step": 480
189
- },
190
- {
191
- "epoch": 0.23,
192
- "learning_rate": 7.080657791699295e-05,
193
- "loss": 3.0833,
194
- "step": 500
195
- },
196
- {
197
- "epoch": 0.23,
198
- "eval_loss": 2.9769718647003174,
199
- "eval_runtime": 195.2639,
200
- "eval_samples_per_second": 24.802,
201
- "eval_steps_per_second": 0.778,
202
- "eval_wer": 1.0,
203
- "step": 500
204
- },
205
- {
206
- "epoch": 0.24,
207
- "learning_rate": 7.05716523101018e-05,
208
- "loss": 3.1608,
209
- "step": 520
210
- },
211
- {
212
- "epoch": 0.25,
213
- "learning_rate": 7.033672670321064e-05,
214
- "loss": 3.0206,
215
- "step": 540
216
- },
217
- {
218
- "epoch": 0.26,
219
- "learning_rate": 7.01018010963195e-05,
220
- "loss": 3.0523,
221
- "step": 560
222
- },
223
- {
224
- "epoch": 0.27,
225
- "learning_rate": 6.986687548942835e-05,
226
- "loss": 3.0222,
227
- "step": 580
228
- },
229
- {
230
- "epoch": 0.28,
231
- "learning_rate": 6.964369616288175e-05,
232
- "loss": 3.0913,
233
- "step": 600
234
- },
235
- {
236
- "epoch": 0.28,
237
- "eval_loss": 3.052091598510742,
238
- "eval_runtime": 189.8149,
239
- "eval_samples_per_second": 25.514,
240
- "eval_steps_per_second": 0.801,
241
- "eval_wer": 1.0,
242
- "step": 600
243
- },
244
- {
245
- "epoch": 0.29,
246
- "learning_rate": 6.94087705559906e-05,
247
- "loss": 3.1987,
248
- "step": 620
249
- },
250
- {
251
- "epoch": 0.29,
252
- "learning_rate": 6.917384494909944e-05,
253
- "loss": 3.1504,
254
- "step": 640
255
- },
256
- {
257
- "epoch": 0.3,
258
- "learning_rate": 6.893891934220829e-05,
259
- "loss": 3.2406,
260
- "step": 660
261
- },
262
- {
263
- "epoch": 0.31,
264
- "learning_rate": 6.870399373531715e-05,
265
- "loss": 3.0632,
266
- "step": 680
267
- },
268
- {
269
- "epoch": 0.32,
270
- "learning_rate": 6.848081440877055e-05,
271
- "loss": 3.1487,
272
- "step": 700
273
- },
274
- {
275
- "epoch": 0.32,
276
- "eval_loss": 3.1951305866241455,
277
- "eval_runtime": 196.8636,
278
- "eval_samples_per_second": 24.601,
279
- "eval_steps_per_second": 0.772,
280
- "eval_wer": 1.0,
281
- "step": 700
282
- },
283
- {
284
- "epoch": 0.33,
285
- "learning_rate": 6.824588880187941e-05,
286
- "loss": 3.298,
287
- "step": 720
288
- },
289
- {
290
- "epoch": 0.34,
291
- "learning_rate": 6.801096319498825e-05,
292
- "loss": 3.1669,
293
- "step": 740
294
- },
295
- {
296
- "epoch": 0.35,
297
- "learning_rate": 6.778778386844165e-05,
298
- "loss": 3.2651,
299
- "step": 760
300
- },
301
- {
302
- "epoch": 0.36,
303
- "learning_rate": 6.75528582615505e-05,
304
- "loss": 3.1562,
305
- "step": 780
306
- },
307
- {
308
- "epoch": 0.37,
309
- "learning_rate": 6.732967893500391e-05,
310
- "loss": 3.3254,
311
- "step": 800
312
- },
313
- {
314
- "epoch": 0.37,
315
- "eval_loss": 3.0070509910583496,
316
- "eval_runtime": 192.0978,
317
- "eval_samples_per_second": 25.211,
318
- "eval_steps_per_second": 0.791,
319
- "eval_wer": 1.0,
320
- "step": 800
321
- }
322
- ],
323
- "max_steps": 6516,
324
- "num_train_epochs": 3,
325
- "total_flos": 1.5225400497310126e+19,
326
- "trial_name": null,
327
- "trial_params": null
328
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
checkpoint-800/training_args.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:962dbb61fb0395a025f0dae3475a50ad9719299753b3a01bede5cb391a4aca78
3
- size 3055