hr16 commited on
Commit
f615462
1 Parent(s): a57aa77

Training in progress, epoch 1, checkpoint

Browse files
checkpoint-1469/config.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "vinai/PhoWhisper-medium",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "apply_spec_augment": false,
6
+ "architectures": [
7
+ "ViSpeechClassification"
8
+ ],
9
+ "attention_dropout": 0.0,
10
+ "begin_suppress_tokens": [
11
+ 220,
12
+ 50257
13
+ ],
14
+ "bos_token_id": 50257,
15
+ "classifier_proj_size": 256,
16
+ "d_model": 1024,
17
+ "decoder_attention_heads": 16,
18
+ "decoder_ffn_dim": 4096,
19
+ "decoder_layerdrop": 0.0,
20
+ "decoder_layers": 24,
21
+ "decoder_start_token_id": 50258,
22
+ "dropout": 0.0,
23
+ "encoder_attention_heads": 16,
24
+ "encoder_ffn_dim": 4096,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 24,
27
+ "eos_token_id": 50257,
28
+ "forced_decoder_ids": null,
29
+ "id2label": {
30
+ "0": "male, northern dialect",
31
+ "1": "male, central dialect",
32
+ "2": "male, highland central dialect",
33
+ "3": "male, southern dialect",
34
+ "4": "female, northern dialect",
35
+ "5": "female, central dialect",
36
+ "6": "female, highland central dialect",
37
+ "7": "female, southern dialect"
38
+ },
39
+ "init_std": 0.02,
40
+ "is_encoder_decoder": true,
41
+ "label2id": {
42
+ "female, central dialect": "5",
43
+ "female, highland central dialect": "6",
44
+ "female, northern dialect": "4",
45
+ "female, southern dialect": "7",
46
+ "male, central dialect": "1",
47
+ "male, highland central dialect": "2",
48
+ "male, northern dialect": "0",
49
+ "male, southern dialect": "3"
50
+ },
51
+ "mask_feature_length": 10,
52
+ "mask_feature_min_masks": 0,
53
+ "mask_feature_prob": 0.0,
54
+ "mask_time_length": 10,
55
+ "mask_time_min_masks": 2,
56
+ "mask_time_prob": 0.05,
57
+ "max_length": 448,
58
+ "max_source_positions": 1500,
59
+ "max_target_positions": 448,
60
+ "median_filter_width": 7,
61
+ "model_type": "whisper",
62
+ "num_hidden_layers": 24,
63
+ "num_mel_bins": 80,
64
+ "pad_token_id": 50257,
65
+ "scale_embedding": false,
66
+ "torch_dtype": "float32",
67
+ "transformers_version": "4.41.2",
68
+ "use_cache": true,
69
+ "use_weighted_layer_sum": false,
70
+ "vocab_size": 51865
71
+ }
checkpoint-1469/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57390c26ca78fa7df1428c07a8a26d1db0713746b8485ee909472e6cbe4f1e58
3
+ size 1235737744
checkpoint-1469/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c0d6709c9d015b60c7e62e95914b3999bc8106ee1158c7ef476cfb6e03dca49
3
+ size 2459418356
checkpoint-1469/preprocessor_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length": 30,
3
+ "feature_extractor_type": "WhisperFeatureExtractor",
4
+ "feature_size": 80,
5
+ "hop_length": 160,
6
+ "n_fft": 400,
7
+ "n_samples": 480000,
8
+ "nb_max_frames": 3000,
9
+ "padding_side": "right",
10
+ "padding_value": 0.0,
11
+ "processor_class": "WhisperProcessor",
12
+ "return_attention_mask": false,
13
+ "sampling_rate": 16000
14
+ }
checkpoint-1469/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7dd282ed118b223ce7335f6545a695708afeb2ef0973e52421614bcd393988c
3
+ size 14244
checkpoint-1469/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05cb1aa66fe604138ab25bd0cfcda48f7f91de1ea3e152310634053d217839c1
3
+ size 1064
checkpoint-1469/trainer_state.json ADDED
@@ -0,0 +1,721 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8819584171696848,
3
+ "best_model_checkpoint": "PhoWhisper-medium-vispeech-classifier-v1/checkpoint-1469",
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1469,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.010211027910142955,
13
+ "grad_norm": 79976.4921875,
14
+ "learning_rate": 6.122448979591837e-08,
15
+ "loss": 2.0788,
16
+ "step": 15
17
+ },
18
+ {
19
+ "epoch": 0.02042205582028591,
20
+ "grad_norm": 49252.46484375,
21
+ "learning_rate": 1.2244897959183673e-07,
22
+ "loss": 2.0787,
23
+ "step": 30
24
+ },
25
+ {
26
+ "epoch": 0.03063308373042886,
27
+ "grad_norm": 66237.7890625,
28
+ "learning_rate": 1.836734693877551e-07,
29
+ "loss": 2.0734,
30
+ "step": 45
31
+ },
32
+ {
33
+ "epoch": 0.04084411164057182,
34
+ "grad_norm": 66530.140625,
35
+ "learning_rate": 2.4489795918367347e-07,
36
+ "loss": 2.0745,
37
+ "step": 60
38
+ },
39
+ {
40
+ "epoch": 0.05105513955071477,
41
+ "grad_norm": 75123.3515625,
42
+ "learning_rate": 3.0612244897959183e-07,
43
+ "loss": 2.0715,
44
+ "step": 75
45
+ },
46
+ {
47
+ "epoch": 0.06126616746085772,
48
+ "grad_norm": 91761.9921875,
49
+ "learning_rate": 3.673469387755102e-07,
50
+ "loss": 2.0702,
51
+ "step": 90
52
+ },
53
+ {
54
+ "epoch": 0.07147719537100068,
55
+ "grad_norm": 76544.7421875,
56
+ "learning_rate": 4.2857142857142857e-07,
57
+ "loss": 2.0659,
58
+ "step": 105
59
+ },
60
+ {
61
+ "epoch": 0.08168822328114364,
62
+ "grad_norm": 85211.5078125,
63
+ "learning_rate": 4.897959183673469e-07,
64
+ "loss": 2.0632,
65
+ "step": 120
66
+ },
67
+ {
68
+ "epoch": 0.09189925119128659,
69
+ "grad_norm": 86731.6171875,
70
+ "learning_rate": 5.510204081632653e-07,
71
+ "loss": 2.0579,
72
+ "step": 135
73
+ },
74
+ {
75
+ "epoch": 0.10211027910142954,
76
+ "grad_norm": 73202.421875,
77
+ "learning_rate": 6.122448979591837e-07,
78
+ "loss": 2.0466,
79
+ "step": 150
80
+ },
81
+ {
82
+ "epoch": 0.1123213070115725,
83
+ "grad_norm": 99795.828125,
84
+ "learning_rate": 6.73469387755102e-07,
85
+ "loss": 2.0266,
86
+ "step": 165
87
+ },
88
+ {
89
+ "epoch": 0.12253233492171545,
90
+ "grad_norm": 78665.6015625,
91
+ "learning_rate": 7.346938775510204e-07,
92
+ "loss": 2.0133,
93
+ "step": 180
94
+ },
95
+ {
96
+ "epoch": 0.13274336283185842,
97
+ "grad_norm": 109804.5546875,
98
+ "learning_rate": 7.959183673469389e-07,
99
+ "loss": 2.0033,
100
+ "step": 195
101
+ },
102
+ {
103
+ "epoch": 0.14295439074200136,
104
+ "grad_norm": 88988.0234375,
105
+ "learning_rate": 8.571428571428571e-07,
106
+ "loss": 1.9912,
107
+ "step": 210
108
+ },
109
+ {
110
+ "epoch": 0.1531654186521443,
111
+ "grad_norm": 163108.234375,
112
+ "learning_rate": 9.183673469387755e-07,
113
+ "loss": 1.9714,
114
+ "step": 225
115
+ },
116
+ {
117
+ "epoch": 0.16337644656228728,
118
+ "grad_norm": 147248.25,
119
+ "learning_rate": 9.795918367346939e-07,
120
+ "loss": 1.9368,
121
+ "step": 240
122
+ },
123
+ {
124
+ "epoch": 0.17358747447243023,
125
+ "grad_norm": 161750.640625,
126
+ "learning_rate": 1.0408163265306123e-06,
127
+ "loss": 1.8955,
128
+ "step": 255
129
+ },
130
+ {
131
+ "epoch": 0.18379850238257317,
132
+ "grad_norm": 189184.296875,
133
+ "learning_rate": 1.1020408163265306e-06,
134
+ "loss": 1.8849,
135
+ "step": 270
136
+ },
137
+ {
138
+ "epoch": 0.19400953029271614,
139
+ "grad_norm": 160164.859375,
140
+ "learning_rate": 1.163265306122449e-06,
141
+ "loss": 1.8496,
142
+ "step": 285
143
+ },
144
+ {
145
+ "epoch": 0.2042205582028591,
146
+ "grad_norm": 147315.296875,
147
+ "learning_rate": 1.2244897959183673e-06,
148
+ "loss": 1.8118,
149
+ "step": 300
150
+ },
151
+ {
152
+ "epoch": 0.21443158611300203,
153
+ "grad_norm": 164890.859375,
154
+ "learning_rate": 1.2857142857142856e-06,
155
+ "loss": 1.7829,
156
+ "step": 315
157
+ },
158
+ {
159
+ "epoch": 0.224642614023145,
160
+ "grad_norm": 195027.578125,
161
+ "learning_rate": 1.346938775510204e-06,
162
+ "loss": 1.7938,
163
+ "step": 330
164
+ },
165
+ {
166
+ "epoch": 0.23485364193328795,
167
+ "grad_norm": 206460.03125,
168
+ "learning_rate": 1.4081632653061225e-06,
169
+ "loss": 1.7258,
170
+ "step": 345
171
+ },
172
+ {
173
+ "epoch": 0.2450646698434309,
174
+ "grad_norm": 247498.296875,
175
+ "learning_rate": 1.4693877551020408e-06,
176
+ "loss": 1.6498,
177
+ "step": 360
178
+ },
179
+ {
180
+ "epoch": 0.25527569775357384,
181
+ "grad_norm": 282144.96875,
182
+ "learning_rate": 1.5306122448979593e-06,
183
+ "loss": 1.6515,
184
+ "step": 375
185
+ },
186
+ {
187
+ "epoch": 0.26548672566371684,
188
+ "grad_norm": 244598.515625,
189
+ "learning_rate": 1.5918367346938777e-06,
190
+ "loss": 1.6327,
191
+ "step": 390
192
+ },
193
+ {
194
+ "epoch": 0.2756977535738598,
195
+ "grad_norm": 169616.5,
196
+ "learning_rate": 1.6530612244897958e-06,
197
+ "loss": 1.576,
198
+ "step": 405
199
+ },
200
+ {
201
+ "epoch": 0.2859087814840027,
202
+ "grad_norm": 186143.078125,
203
+ "learning_rate": 1.7142857142857143e-06,
204
+ "loss": 1.6067,
205
+ "step": 420
206
+ },
207
+ {
208
+ "epoch": 0.29611980939414567,
209
+ "grad_norm": 239726.171875,
210
+ "learning_rate": 1.7755102040816327e-06,
211
+ "loss": 1.5676,
212
+ "step": 435
213
+ },
214
+ {
215
+ "epoch": 0.3063308373042886,
216
+ "grad_norm": 224399.15625,
217
+ "learning_rate": 1.836734693877551e-06,
218
+ "loss": 1.4658,
219
+ "step": 450
220
+ },
221
+ {
222
+ "epoch": 0.31654186521443156,
223
+ "grad_norm": 585805.375,
224
+ "learning_rate": 1.8979591836734695e-06,
225
+ "loss": 1.4288,
226
+ "step": 465
227
+ },
228
+ {
229
+ "epoch": 0.32675289312457456,
230
+ "grad_norm": 204082.328125,
231
+ "learning_rate": 1.9591836734693877e-06,
232
+ "loss": 1.4079,
233
+ "step": 480
234
+ },
235
+ {
236
+ "epoch": 0.3369639210347175,
237
+ "grad_norm": 253897.078125,
238
+ "learning_rate": 2.020408163265306e-06,
239
+ "loss": 1.4658,
240
+ "step": 495
241
+ },
242
+ {
243
+ "epoch": 0.34717494894486045,
244
+ "grad_norm": 322156.15625,
245
+ "learning_rate": 2.0816326530612247e-06,
246
+ "loss": 1.3193,
247
+ "step": 510
248
+ },
249
+ {
250
+ "epoch": 0.3573859768550034,
251
+ "grad_norm": 209002.03125,
252
+ "learning_rate": 2.142857142857143e-06,
253
+ "loss": 1.3258,
254
+ "step": 525
255
+ },
256
+ {
257
+ "epoch": 0.36759700476514634,
258
+ "grad_norm": 239622.390625,
259
+ "learning_rate": 2.204081632653061e-06,
260
+ "loss": 1.3396,
261
+ "step": 540
262
+ },
263
+ {
264
+ "epoch": 0.3778080326752893,
265
+ "grad_norm": 302531.125,
266
+ "learning_rate": 2.2653061224489797e-06,
267
+ "loss": 1.2724,
268
+ "step": 555
269
+ },
270
+ {
271
+ "epoch": 0.3880190605854323,
272
+ "grad_norm": 277807.9375,
273
+ "learning_rate": 2.326530612244898e-06,
274
+ "loss": 1.2506,
275
+ "step": 570
276
+ },
277
+ {
278
+ "epoch": 0.39823008849557523,
279
+ "grad_norm": 543051.25,
280
+ "learning_rate": 2.387755102040816e-06,
281
+ "loss": 1.3114,
282
+ "step": 585
283
+ },
284
+ {
285
+ "epoch": 0.4084411164057182,
286
+ "grad_norm": 278458.59375,
287
+ "learning_rate": 2.4489795918367347e-06,
288
+ "loss": 1.2505,
289
+ "step": 600
290
+ },
291
+ {
292
+ "epoch": 0.4186521443158611,
293
+ "grad_norm": 227854.078125,
294
+ "learning_rate": 2.510204081632653e-06,
295
+ "loss": 1.2032,
296
+ "step": 615
297
+ },
298
+ {
299
+ "epoch": 0.42886317222600406,
300
+ "grad_norm": 669112.0625,
301
+ "learning_rate": 2.571428571428571e-06,
302
+ "loss": 1.2144,
303
+ "step": 630
304
+ },
305
+ {
306
+ "epoch": 0.43907420013614706,
307
+ "grad_norm": 291158.375,
308
+ "learning_rate": 2.6326530612244897e-06,
309
+ "loss": 1.1663,
310
+ "step": 645
311
+ },
312
+ {
313
+ "epoch": 0.44928522804629,
314
+ "grad_norm": 322840.65625,
315
+ "learning_rate": 2.693877551020408e-06,
316
+ "loss": 1.1802,
317
+ "step": 660
318
+ },
319
+ {
320
+ "epoch": 0.45949625595643295,
321
+ "grad_norm": 399617.4375,
322
+ "learning_rate": 2.7551020408163266e-06,
323
+ "loss": 1.0181,
324
+ "step": 675
325
+ },
326
+ {
327
+ "epoch": 0.4697072838665759,
328
+ "grad_norm": 399545.5,
329
+ "learning_rate": 2.816326530612245e-06,
330
+ "loss": 1.2468,
331
+ "step": 690
332
+ },
333
+ {
334
+ "epoch": 0.47991831177671884,
335
+ "grad_norm": 1116027.875,
336
+ "learning_rate": 2.877551020408163e-06,
337
+ "loss": 1.2023,
338
+ "step": 705
339
+ },
340
+ {
341
+ "epoch": 0.4901293396868618,
342
+ "grad_norm": 300487.4375,
343
+ "learning_rate": 2.9387755102040816e-06,
344
+ "loss": 1.147,
345
+ "step": 720
346
+ },
347
+ {
348
+ "epoch": 0.5003403675970047,
349
+ "grad_norm": 413331.65625,
350
+ "learning_rate": 3e-06,
351
+ "loss": 0.9871,
352
+ "step": 735
353
+ },
354
+ {
355
+ "epoch": 0.5105513955071477,
356
+ "grad_norm": 715814.125,
357
+ "learning_rate": 2.993192133131619e-06,
358
+ "loss": 1.1643,
359
+ "step": 750
360
+ },
361
+ {
362
+ "epoch": 0.5207624234172906,
363
+ "grad_norm": 1614147.0,
364
+ "learning_rate": 2.9863842662632373e-06,
365
+ "loss": 0.9421,
366
+ "step": 765
367
+ },
368
+ {
369
+ "epoch": 0.5309734513274337,
370
+ "grad_norm": 639895.0625,
371
+ "learning_rate": 2.9795763993948565e-06,
372
+ "loss": 1.0573,
373
+ "step": 780
374
+ },
375
+ {
376
+ "epoch": 0.5411844792375766,
377
+ "grad_norm": 345273.84375,
378
+ "learning_rate": 2.9727685325264754e-06,
379
+ "loss": 1.0337,
380
+ "step": 795
381
+ },
382
+ {
383
+ "epoch": 0.5513955071477196,
384
+ "grad_norm": 611370.375,
385
+ "learning_rate": 2.9659606656580938e-06,
386
+ "loss": 1.1131,
387
+ "step": 810
388
+ },
389
+ {
390
+ "epoch": 0.5616065350578625,
391
+ "grad_norm": 239903.25,
392
+ "learning_rate": 2.9591527987897126e-06,
393
+ "loss": 0.9076,
394
+ "step": 825
395
+ },
396
+ {
397
+ "epoch": 0.5718175629680055,
398
+ "grad_norm": 257958.703125,
399
+ "learning_rate": 2.9523449319213314e-06,
400
+ "loss": 0.9152,
401
+ "step": 840
402
+ },
403
+ {
404
+ "epoch": 0.5820285908781484,
405
+ "grad_norm": 590921.375,
406
+ "learning_rate": 2.9455370650529502e-06,
407
+ "loss": 0.8795,
408
+ "step": 855
409
+ },
410
+ {
411
+ "epoch": 0.5922396187882913,
412
+ "grad_norm": 363986.1875,
413
+ "learning_rate": 2.938729198184569e-06,
414
+ "loss": 1.0641,
415
+ "step": 870
416
+ },
417
+ {
418
+ "epoch": 0.6024506466984343,
419
+ "grad_norm": 1878926.875,
420
+ "learning_rate": 2.9319213313161875e-06,
421
+ "loss": 0.9414,
422
+ "step": 885
423
+ },
424
+ {
425
+ "epoch": 0.6126616746085772,
426
+ "grad_norm": 2941328.0,
427
+ "learning_rate": 2.9251134644478063e-06,
428
+ "loss": 0.9541,
429
+ "step": 900
430
+ },
431
+ {
432
+ "epoch": 0.6228727025187202,
433
+ "grad_norm": 1327942.75,
434
+ "learning_rate": 2.918305597579425e-06,
435
+ "loss": 0.8319,
436
+ "step": 915
437
+ },
438
+ {
439
+ "epoch": 0.6330837304288631,
440
+ "grad_norm": 161807.125,
441
+ "learning_rate": 2.911497730711044e-06,
442
+ "loss": 0.9268,
443
+ "step": 930
444
+ },
445
+ {
446
+ "epoch": 0.6432947583390062,
447
+ "grad_norm": 899555.875,
448
+ "learning_rate": 2.9046898638426628e-06,
449
+ "loss": 0.8278,
450
+ "step": 945
451
+ },
452
+ {
453
+ "epoch": 0.6535057862491491,
454
+ "grad_norm": 268360.0625,
455
+ "learning_rate": 2.8978819969742816e-06,
456
+ "loss": 0.9021,
457
+ "step": 960
458
+ },
459
+ {
460
+ "epoch": 0.6637168141592921,
461
+ "grad_norm": 555703.0625,
462
+ "learning_rate": 2.8910741301059e-06,
463
+ "loss": 1.0146,
464
+ "step": 975
465
+ },
466
+ {
467
+ "epoch": 0.673927842069435,
468
+ "grad_norm": 222425.40625,
469
+ "learning_rate": 2.884266263237519e-06,
470
+ "loss": 1.0005,
471
+ "step": 990
472
+ },
473
+ {
474
+ "epoch": 0.684138869979578,
475
+ "grad_norm": 681254.4375,
476
+ "learning_rate": 2.877458396369138e-06,
477
+ "loss": 0.842,
478
+ "step": 1005
479
+ },
480
+ {
481
+ "epoch": 0.6943498978897209,
482
+ "grad_norm": 1348221.125,
483
+ "learning_rate": 2.8706505295007565e-06,
484
+ "loss": 0.9394,
485
+ "step": 1020
486
+ },
487
+ {
488
+ "epoch": 0.7045609257998638,
489
+ "grad_norm": 225728.046875,
490
+ "learning_rate": 2.8638426626323753e-06,
491
+ "loss": 0.8289,
492
+ "step": 1035
493
+ },
494
+ {
495
+ "epoch": 0.7147719537100068,
496
+ "grad_norm": 413089.65625,
497
+ "learning_rate": 2.857034795763994e-06,
498
+ "loss": 1.0175,
499
+ "step": 1050
500
+ },
501
+ {
502
+ "epoch": 0.7249829816201497,
503
+ "grad_norm": 279751.71875,
504
+ "learning_rate": 2.8502269288956125e-06,
505
+ "loss": 0.9563,
506
+ "step": 1065
507
+ },
508
+ {
509
+ "epoch": 0.7351940095302927,
510
+ "grad_norm": 422167.125,
511
+ "learning_rate": 2.8434190620272318e-06,
512
+ "loss": 0.8566,
513
+ "step": 1080
514
+ },
515
+ {
516
+ "epoch": 0.7454050374404356,
517
+ "grad_norm": 2015877.125,
518
+ "learning_rate": 2.83661119515885e-06,
519
+ "loss": 0.7012,
520
+ "step": 1095
521
+ },
522
+ {
523
+ "epoch": 0.7556160653505786,
524
+ "grad_norm": 502859.5,
525
+ "learning_rate": 2.829803328290469e-06,
526
+ "loss": 0.7563,
527
+ "step": 1110
528
+ },
529
+ {
530
+ "epoch": 0.7658270932607216,
531
+ "grad_norm": 537785.125,
532
+ "learning_rate": 2.822995461422088e-06,
533
+ "loss": 0.9854,
534
+ "step": 1125
535
+ },
536
+ {
537
+ "epoch": 0.7760381211708646,
538
+ "grad_norm": 390231.125,
539
+ "learning_rate": 2.8161875945537066e-06,
540
+ "loss": 1.0036,
541
+ "step": 1140
542
+ },
543
+ {
544
+ "epoch": 0.7862491490810075,
545
+ "grad_norm": 370246.0625,
546
+ "learning_rate": 2.8093797276853255e-06,
547
+ "loss": 0.8186,
548
+ "step": 1155
549
+ },
550
+ {
551
+ "epoch": 0.7964601769911505,
552
+ "grad_norm": 1279901.375,
553
+ "learning_rate": 2.8025718608169443e-06,
554
+ "loss": 0.9289,
555
+ "step": 1170
556
+ },
557
+ {
558
+ "epoch": 0.8066712049012934,
559
+ "grad_norm": 316319.25,
560
+ "learning_rate": 2.7957639939485627e-06,
561
+ "loss": 0.7542,
562
+ "step": 1185
563
+ },
564
+ {
565
+ "epoch": 0.8168822328114363,
566
+ "grad_norm": 577071.5,
567
+ "learning_rate": 2.7889561270801815e-06,
568
+ "loss": 0.8579,
569
+ "step": 1200
570
+ },
571
+ {
572
+ "epoch": 0.8270932607215793,
573
+ "grad_norm": 412667.71875,
574
+ "learning_rate": 2.7821482602118003e-06,
575
+ "loss": 0.8424,
576
+ "step": 1215
577
+ },
578
+ {
579
+ "epoch": 0.8373042886317222,
580
+ "grad_norm": 555197.125,
581
+ "learning_rate": 2.775340393343419e-06,
582
+ "loss": 0.8225,
583
+ "step": 1230
584
+ },
585
+ {
586
+ "epoch": 0.8475153165418652,
587
+ "grad_norm": 429715.0,
588
+ "learning_rate": 2.768532526475038e-06,
589
+ "loss": 1.1042,
590
+ "step": 1245
591
+ },
592
+ {
593
+ "epoch": 0.8577263444520081,
594
+ "grad_norm": 1407773.25,
595
+ "learning_rate": 2.761724659606657e-06,
596
+ "loss": 0.9892,
597
+ "step": 1260
598
+ },
599
+ {
600
+ "epoch": 0.8679373723621511,
601
+ "grad_norm": 328839.6875,
602
+ "learning_rate": 2.754916792738275e-06,
603
+ "loss": 0.7921,
604
+ "step": 1275
605
+ },
606
+ {
607
+ "epoch": 0.8781484002722941,
608
+ "grad_norm": 196251.03125,
609
+ "learning_rate": 2.748108925869894e-06,
610
+ "loss": 0.6631,
611
+ "step": 1290
612
+ },
613
+ {
614
+ "epoch": 0.8883594281824371,
615
+ "grad_norm": 761081.6875,
616
+ "learning_rate": 2.7413010590015133e-06,
617
+ "loss": 0.8396,
618
+ "step": 1305
619
+ },
620
+ {
621
+ "epoch": 0.89857045609258,
622
+ "grad_norm": 905756.375,
623
+ "learning_rate": 2.7344931921331317e-06,
624
+ "loss": 0.8238,
625
+ "step": 1320
626
+ },
627
+ {
628
+ "epoch": 0.908781484002723,
629
+ "grad_norm": 1221640.875,
630
+ "learning_rate": 2.7276853252647505e-06,
631
+ "loss": 0.7228,
632
+ "step": 1335
633
+ },
634
+ {
635
+ "epoch": 0.9189925119128659,
636
+ "grad_norm": 2633817.0,
637
+ "learning_rate": 2.7208774583963693e-06,
638
+ "loss": 0.8766,
639
+ "step": 1350
640
+ },
641
+ {
642
+ "epoch": 0.9292035398230089,
643
+ "grad_norm": 454206.53125,
644
+ "learning_rate": 2.7140695915279877e-06,
645
+ "loss": 0.922,
646
+ "step": 1365
647
+ },
648
+ {
649
+ "epoch": 0.9394145677331518,
650
+ "grad_norm": 715028.8125,
651
+ "learning_rate": 2.707261724659607e-06,
652
+ "loss": 0.8212,
653
+ "step": 1380
654
+ },
655
+ {
656
+ "epoch": 0.9496255956432947,
657
+ "grad_norm": 384052.40625,
658
+ "learning_rate": 2.7004538577912254e-06,
659
+ "loss": 0.6728,
660
+ "step": 1395
661
+ },
662
+ {
663
+ "epoch": 0.9598366235534377,
664
+ "grad_norm": 268327.0,
665
+ "learning_rate": 2.693645990922844e-06,
666
+ "loss": 0.7077,
667
+ "step": 1410
668
+ },
669
+ {
670
+ "epoch": 0.9700476514635806,
671
+ "grad_norm": 430912.0625,
672
+ "learning_rate": 2.686838124054463e-06,
673
+ "loss": 0.8267,
674
+ "step": 1425
675
+ },
676
+ {
677
+ "epoch": 0.9802586793737236,
678
+ "grad_norm": 448801.65625,
679
+ "learning_rate": 2.680030257186082e-06,
680
+ "loss": 0.6444,
681
+ "step": 1440
682
+ },
683
+ {
684
+ "epoch": 0.9904697072838666,
685
+ "grad_norm": 625466.3125,
686
+ "learning_rate": 2.6732223903177007e-06,
687
+ "loss": 0.8255,
688
+ "step": 1455
689
+ },
690
+ {
691
+ "epoch": 1.0,
692
+ "eval_accuracy": 0.8819584171696848,
693
+ "eval_loss": 0.5119155645370483,
694
+ "eval_runtime": 579.0751,
695
+ "eval_samples_per_second": 5.15,
696
+ "eval_steps_per_second": 0.644,
697
+ "step": 1469
698
+ }
699
+ ],
700
+ "logging_steps": 15,
701
+ "max_steps": 7345,
702
+ "num_input_tokens_seen": 0,
703
+ "num_train_epochs": 5,
704
+ "save_steps": 500,
705
+ "stateful_callbacks": {
706
+ "TrainerControl": {
707
+ "args": {
708
+ "should_epoch_stop": false,
709
+ "should_evaluate": false,
710
+ "should_log": false,
711
+ "should_save": true,
712
+ "should_training_stop": false
713
+ },
714
+ "attributes": {}
715
+ }
716
+ },
717
+ "total_flos": 5.20189308048384e+18,
718
+ "train_batch_size": 8,
719
+ "trial_name": null,
720
+ "trial_params": null
721
+ }
checkpoint-1469/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2de8a88815cd7a4ce539dab9d23a753db8fa0d9213c6e02f1f86779a8bf8950f
3
+ size 5176