Santp98 commited on
Commit
09adc1a
1 Parent(s): cfc8c00

Training in progress, step 500

Browse files
README.md ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: dccuchile/distilbert-base-spanish-uncased
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: distilbert-base-spanish-uncased-2023-11-13-18-01
7
+ results: []
8
+ ---
9
+
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
+ # distilbert-base-spanish-uncased-2023-11-13-18-01
14
+
15
+ This model is a fine-tuned version of [dccuchile/distilbert-base-spanish-uncased](https://huggingface.co/dccuchile/distilbert-base-spanish-uncased) on the None dataset.
16
+ It achieves the following results on the evaluation set:
17
+ - Loss: 1.3884
18
+
19
+ ## Model description
20
+
21
+ More information needed
22
+
23
+ ## Intended uses & limitations
24
+
25
+ More information needed
26
+
27
+ ## Training and evaluation data
28
+
29
+ More information needed
30
+
31
+ ## Training procedure
32
+
33
+ ### Training hyperparameters
34
+
35
+ The following hyperparameters were used during training:
36
+ - learning_rate: 5e-05
37
+ - train_batch_size: 16
38
+ - eval_batch_size: 16
39
+ - seed: 42
40
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
+ - lr_scheduler_type: linear
42
+ - num_epochs: 10
43
+
44
+ ### Training results
45
+
46
+ | Training Loss | Epoch | Step | Validation Loss |
47
+ |:-------------:|:-----:|:----:|:---------------:|
48
+ | 2.5316 | 0.59 | 500 | 2.2591 |
49
+ | 2.2197 | 1.19 | 1000 | 1.9800 |
50
+ | 2.0108 | 1.78 | 1500 | 1.8304 |
51
+ | 1.855 | 2.38 | 2000 | 1.7229 |
52
+ | 1.7933 | 2.97 | 2500 | 1.6561 |
53
+ | 1.7466 | 3.56 | 3000 | 1.6101 |
54
+ | 1.6906 | 4.16 | 3500 | 1.5588 |
55
+ | 1.6301 | 4.75 | 4000 | 1.5264 |
56
+ | 1.6131 | 5.34 | 4500 | 1.4893 |
57
+ | 1.5731 | 5.94 | 5000 | 1.4626 |
58
+ | 1.5248 | 6.53 | 5500 | 1.4445 |
59
+ | 1.509 | 7.13 | 6000 | 1.4274 |
60
+ | 1.5083 | 7.72 | 6500 | 1.4109 |
61
+ | 1.5277 | 8.31 | 7000 | 1.4051 |
62
+ | 1.4884 | 8.91 | 7500 | 1.3872 |
63
+ | 1.496 | 9.5 | 8000 | 1.3912 |
64
+
65
+
66
+ ### Framework versions
67
+
68
+ - Transformers 4.35.0
69
+ - Pytorch 2.1.0+cu121
70
+ - Datasets 2.14.6
71
+ - Tokenizers 0.14.1
all_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_loss": 1.3867350816726685,
4
+ "eval_runtime": 34.1096,
5
+ "eval_samples_per_second": 131.547,
6
+ "eval_steps_per_second": 8.238
7
+ }
config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "dccuchile/albert-base-spanish",
3
+ "architectures": [
4
+ "AlbertForMaskedLM"
5
+ ],
6
+ "attention_probs_dropout_prob": 0,
7
+ "bos_token_id": 2,
8
+ "classifier_dropout_prob": 0.1,
9
+ "down_scale_factor": 1,
10
+ "embedding_size": 128,
11
+ "eos_token_id": 3,
12
+ "gap_size": 0,
13
+ "hidden_act": "gelu",
14
+ "hidden_dropout_prob": 0,
15
+ "hidden_size": 768,
16
+ "initializer_range": 0.02,
17
+ "inner_group_num": 1,
18
+ "intermediate_size": 3072,
19
+ "layer_norm_eps": 1e-12,
20
+ "max_position_embeddings": 512,
21
+ "model_type": "albert",
22
+ "net_structure_type": 0,
23
+ "num_attention_heads": 12,
24
+ "num_hidden_groups": 1,
25
+ "num_hidden_layers": 12,
26
+ "num_memory_blocks": 0,
27
+ "pad_token_id": 0,
28
+ "position_embedding_type": "absolute",
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.35.0",
31
+ "type_vocab_size": 2,
32
+ "vocab_size": 31000
33
+ }
mlm_eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_loss": 1.3867350816726685,
4
+ "eval_runtime": 34.1096,
5
+ "eval_samples_per_second": 131.547,
6
+ "eval_steps_per_second": 8.238
7
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da273ea1f686a4ce2a870c54e6c3801520c0c4af3d5fca2a8ec2d3bd2d8c8dd9
3
+ size 45406256
trainer_state.json ADDED
@@ -0,0 +1,676 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.387204885482788,
3
+ "best_model_checkpoint": "output_dir/checkpoint-7500",
4
+ "epoch": 10.0,
5
+ "eval_steps": 500,
6
+ "global_step": 8420,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.12,
13
+ "learning_rate": 4.9406175771971496e-05,
14
+ "loss": 3.5781,
15
+ "step": 100
16
+ },
17
+ {
18
+ "epoch": 0.24,
19
+ "learning_rate": 4.8812351543942996e-05,
20
+ "loss": 3.0221,
21
+ "step": 200
22
+ },
23
+ {
24
+ "epoch": 0.36,
25
+ "learning_rate": 4.821852731591449e-05,
26
+ "loss": 2.7656,
27
+ "step": 300
28
+ },
29
+ {
30
+ "epoch": 0.48,
31
+ "learning_rate": 4.7624703087885984e-05,
32
+ "loss": 2.6432,
33
+ "step": 400
34
+ },
35
+ {
36
+ "epoch": 0.59,
37
+ "learning_rate": 4.7030878859857484e-05,
38
+ "loss": 2.5316,
39
+ "step": 500
40
+ },
41
+ {
42
+ "epoch": 0.59,
43
+ "eval_loss": 2.2591474056243896,
44
+ "eval_runtime": 34.7222,
45
+ "eval_samples_per_second": 129.226,
46
+ "eval_steps_per_second": 8.093,
47
+ "step": 500
48
+ },
49
+ {
50
+ "epoch": 0.71,
51
+ "learning_rate": 4.643705463182898e-05,
52
+ "loss": 2.4743,
53
+ "step": 600
54
+ },
55
+ {
56
+ "epoch": 0.83,
57
+ "learning_rate": 4.584323040380048e-05,
58
+ "loss": 2.3371,
59
+ "step": 700
60
+ },
61
+ {
62
+ "epoch": 0.95,
63
+ "learning_rate": 4.524940617577197e-05,
64
+ "loss": 2.3233,
65
+ "step": 800
66
+ },
67
+ {
68
+ "epoch": 1.07,
69
+ "learning_rate": 4.465558194774347e-05,
70
+ "loss": 2.2412,
71
+ "step": 900
72
+ },
73
+ {
74
+ "epoch": 1.19,
75
+ "learning_rate": 4.406175771971497e-05,
76
+ "loss": 2.2197,
77
+ "step": 1000
78
+ },
79
+ {
80
+ "epoch": 1.19,
81
+ "eval_loss": 1.979953408241272,
82
+ "eval_runtime": 34.803,
83
+ "eval_samples_per_second": 128.926,
84
+ "eval_steps_per_second": 8.074,
85
+ "step": 1000
86
+ },
87
+ {
88
+ "epoch": 1.31,
89
+ "learning_rate": 4.3467933491686466e-05,
90
+ "loss": 2.1312,
91
+ "step": 1100
92
+ },
93
+ {
94
+ "epoch": 1.43,
95
+ "learning_rate": 4.287410926365796e-05,
96
+ "loss": 2.1109,
97
+ "step": 1200
98
+ },
99
+ {
100
+ "epoch": 1.54,
101
+ "learning_rate": 4.228028503562946e-05,
102
+ "loss": 2.0618,
103
+ "step": 1300
104
+ },
105
+ {
106
+ "epoch": 1.66,
107
+ "learning_rate": 4.168646080760095e-05,
108
+ "loss": 2.0827,
109
+ "step": 1400
110
+ },
111
+ {
112
+ "epoch": 1.78,
113
+ "learning_rate": 4.109263657957245e-05,
114
+ "loss": 2.0108,
115
+ "step": 1500
116
+ },
117
+ {
118
+ "epoch": 1.78,
119
+ "eval_loss": 1.830361008644104,
120
+ "eval_runtime": 34.8169,
121
+ "eval_samples_per_second": 128.874,
122
+ "eval_steps_per_second": 8.071,
123
+ "step": 1500
124
+ },
125
+ {
126
+ "epoch": 1.9,
127
+ "learning_rate": 4.049881235154395e-05,
128
+ "loss": 1.9889,
129
+ "step": 1600
130
+ },
131
+ {
132
+ "epoch": 2.02,
133
+ "learning_rate": 3.990498812351544e-05,
134
+ "loss": 1.9597,
135
+ "step": 1700
136
+ },
137
+ {
138
+ "epoch": 2.14,
139
+ "learning_rate": 3.9311163895486934e-05,
140
+ "loss": 1.9155,
141
+ "step": 1800
142
+ },
143
+ {
144
+ "epoch": 2.26,
145
+ "learning_rate": 3.8717339667458435e-05,
146
+ "loss": 1.9138,
147
+ "step": 1900
148
+ },
149
+ {
150
+ "epoch": 2.38,
151
+ "learning_rate": 3.812351543942993e-05,
152
+ "loss": 1.855,
153
+ "step": 2000
154
+ },
155
+ {
156
+ "epoch": 2.38,
157
+ "eval_loss": 1.722902536392212,
158
+ "eval_runtime": 34.0608,
159
+ "eval_samples_per_second": 131.735,
160
+ "eval_steps_per_second": 8.25,
161
+ "step": 2000
162
+ },
163
+ {
164
+ "epoch": 2.49,
165
+ "learning_rate": 3.752969121140142e-05,
166
+ "loss": 1.8592,
167
+ "step": 2100
168
+ },
169
+ {
170
+ "epoch": 2.61,
171
+ "learning_rate": 3.693586698337292e-05,
172
+ "loss": 1.8951,
173
+ "step": 2200
174
+ },
175
+ {
176
+ "epoch": 2.73,
177
+ "learning_rate": 3.6342042755344416e-05,
178
+ "loss": 1.8651,
179
+ "step": 2300
180
+ },
181
+ {
182
+ "epoch": 2.85,
183
+ "learning_rate": 3.5748218527315916e-05,
184
+ "loss": 1.808,
185
+ "step": 2400
186
+ },
187
+ {
188
+ "epoch": 2.97,
189
+ "learning_rate": 3.515439429928741e-05,
190
+ "loss": 1.7933,
191
+ "step": 2500
192
+ },
193
+ {
194
+ "epoch": 2.97,
195
+ "eval_loss": 1.6560965776443481,
196
+ "eval_runtime": 34.0703,
197
+ "eval_samples_per_second": 131.698,
198
+ "eval_steps_per_second": 8.248,
199
+ "step": 2500
200
+ },
201
+ {
202
+ "epoch": 3.09,
203
+ "learning_rate": 3.456057007125891e-05,
204
+ "loss": 1.8124,
205
+ "step": 2600
206
+ },
207
+ {
208
+ "epoch": 3.21,
209
+ "learning_rate": 3.396674584323041e-05,
210
+ "loss": 1.7598,
211
+ "step": 2700
212
+ },
213
+ {
214
+ "epoch": 3.33,
215
+ "learning_rate": 3.3372921615201904e-05,
216
+ "loss": 1.7716,
217
+ "step": 2800
218
+ },
219
+ {
220
+ "epoch": 3.44,
221
+ "learning_rate": 3.27790973871734e-05,
222
+ "loss": 1.7557,
223
+ "step": 2900
224
+ },
225
+ {
226
+ "epoch": 3.56,
227
+ "learning_rate": 3.21852731591449e-05,
228
+ "loss": 1.7466,
229
+ "step": 3000
230
+ },
231
+ {
232
+ "epoch": 3.56,
233
+ "eval_loss": 1.6100554466247559,
234
+ "eval_runtime": 34.1759,
235
+ "eval_samples_per_second": 131.291,
236
+ "eval_steps_per_second": 8.222,
237
+ "step": 3000
238
+ },
239
+ {
240
+ "epoch": 3.68,
241
+ "learning_rate": 3.159144893111639e-05,
242
+ "loss": 1.745,
243
+ "step": 3100
244
+ },
245
+ {
246
+ "epoch": 3.8,
247
+ "learning_rate": 3.0997624703087885e-05,
248
+ "loss": 1.729,
249
+ "step": 3200
250
+ },
251
+ {
252
+ "epoch": 3.92,
253
+ "learning_rate": 3.0403800475059386e-05,
254
+ "loss": 1.7112,
255
+ "step": 3300
256
+ },
257
+ {
258
+ "epoch": 4.04,
259
+ "learning_rate": 2.980997624703088e-05,
260
+ "loss": 1.6804,
261
+ "step": 3400
262
+ },
263
+ {
264
+ "epoch": 4.16,
265
+ "learning_rate": 2.9216152019002373e-05,
266
+ "loss": 1.6906,
267
+ "step": 3500
268
+ },
269
+ {
270
+ "epoch": 4.16,
271
+ "eval_loss": 1.5587605237960815,
272
+ "eval_runtime": 34.0323,
273
+ "eval_samples_per_second": 131.845,
274
+ "eval_steps_per_second": 8.257,
275
+ "step": 3500
276
+ },
277
+ {
278
+ "epoch": 4.28,
279
+ "learning_rate": 2.8622327790973873e-05,
280
+ "loss": 1.6638,
281
+ "step": 3600
282
+ },
283
+ {
284
+ "epoch": 4.39,
285
+ "learning_rate": 2.8028503562945367e-05,
286
+ "loss": 1.6895,
287
+ "step": 3700
288
+ },
289
+ {
290
+ "epoch": 4.51,
291
+ "learning_rate": 2.7434679334916867e-05,
292
+ "loss": 1.6983,
293
+ "step": 3800
294
+ },
295
+ {
296
+ "epoch": 4.63,
297
+ "learning_rate": 2.6840855106888364e-05,
298
+ "loss": 1.6648,
299
+ "step": 3900
300
+ },
301
+ {
302
+ "epoch": 4.75,
303
+ "learning_rate": 2.6247030878859858e-05,
304
+ "loss": 1.6301,
305
+ "step": 4000
306
+ },
307
+ {
308
+ "epoch": 4.75,
309
+ "eval_loss": 1.5263967514038086,
310
+ "eval_runtime": 34.0548,
311
+ "eval_samples_per_second": 131.758,
312
+ "eval_steps_per_second": 8.251,
313
+ "step": 4000
314
+ },
315
+ {
316
+ "epoch": 4.87,
317
+ "learning_rate": 2.5653206650831358e-05,
318
+ "loss": 1.6388,
319
+ "step": 4100
320
+ },
321
+ {
322
+ "epoch": 4.99,
323
+ "learning_rate": 2.5059382422802852e-05,
324
+ "loss": 1.6336,
325
+ "step": 4200
326
+ },
327
+ {
328
+ "epoch": 5.11,
329
+ "learning_rate": 2.446555819477435e-05,
330
+ "loss": 1.6427,
331
+ "step": 4300
332
+ },
333
+ {
334
+ "epoch": 5.23,
335
+ "learning_rate": 2.3871733966745842e-05,
336
+ "loss": 1.6219,
337
+ "step": 4400
338
+ },
339
+ {
340
+ "epoch": 5.34,
341
+ "learning_rate": 2.327790973871734e-05,
342
+ "loss": 1.6131,
343
+ "step": 4500
344
+ },
345
+ {
346
+ "epoch": 5.34,
347
+ "eval_loss": 1.489254355430603,
348
+ "eval_runtime": 34.1712,
349
+ "eval_samples_per_second": 131.309,
350
+ "eval_steps_per_second": 8.223,
351
+ "step": 4500
352
+ },
353
+ {
354
+ "epoch": 5.46,
355
+ "learning_rate": 2.2684085510688836e-05,
356
+ "loss": 1.5886,
357
+ "step": 4600
358
+ },
359
+ {
360
+ "epoch": 5.58,
361
+ "learning_rate": 2.2090261282660333e-05,
362
+ "loss": 1.5906,
363
+ "step": 4700
364
+ },
365
+ {
366
+ "epoch": 5.7,
367
+ "learning_rate": 2.149643705463183e-05,
368
+ "loss": 1.5717,
369
+ "step": 4800
370
+ },
371
+ {
372
+ "epoch": 5.82,
373
+ "learning_rate": 2.0902612826603327e-05,
374
+ "loss": 1.5887,
375
+ "step": 4900
376
+ },
377
+ {
378
+ "epoch": 5.94,
379
+ "learning_rate": 2.0308788598574824e-05,
380
+ "loss": 1.5731,
381
+ "step": 5000
382
+ },
383
+ {
384
+ "epoch": 5.94,
385
+ "eval_loss": 1.4625672101974487,
386
+ "eval_runtime": 34.0832,
387
+ "eval_samples_per_second": 131.649,
388
+ "eval_steps_per_second": 8.245,
389
+ "step": 5000
390
+ },
391
+ {
392
+ "epoch": 6.06,
393
+ "learning_rate": 1.9714964370546318e-05,
394
+ "loss": 1.5607,
395
+ "step": 5100
396
+ },
397
+ {
398
+ "epoch": 6.18,
399
+ "learning_rate": 1.9121140142517815e-05,
400
+ "loss": 1.5905,
401
+ "step": 5200
402
+ },
403
+ {
404
+ "epoch": 6.29,
405
+ "learning_rate": 1.8527315914489312e-05,
406
+ "loss": 1.6074,
407
+ "step": 5300
408
+ },
409
+ {
410
+ "epoch": 6.41,
411
+ "learning_rate": 1.793349168646081e-05,
412
+ "loss": 1.5401,
413
+ "step": 5400
414
+ },
415
+ {
416
+ "epoch": 6.53,
417
+ "learning_rate": 1.7339667458432306e-05,
418
+ "loss": 1.5248,
419
+ "step": 5500
420
+ },
421
+ {
422
+ "epoch": 6.53,
423
+ "eval_loss": 1.4445180892944336,
424
+ "eval_runtime": 33.993,
425
+ "eval_samples_per_second": 131.998,
426
+ "eval_steps_per_second": 8.266,
427
+ "step": 5500
428
+ },
429
+ {
430
+ "epoch": 6.65,
431
+ "learning_rate": 1.6745843230403803e-05,
432
+ "loss": 1.5611,
433
+ "step": 5600
434
+ },
435
+ {
436
+ "epoch": 6.77,
437
+ "learning_rate": 1.61520190023753e-05,
438
+ "loss": 1.5312,
439
+ "step": 5700
440
+ },
441
+ {
442
+ "epoch": 6.89,
443
+ "learning_rate": 1.5558194774346793e-05,
444
+ "loss": 1.5318,
445
+ "step": 5800
446
+ },
447
+ {
448
+ "epoch": 7.01,
449
+ "learning_rate": 1.496437054631829e-05,
450
+ "loss": 1.5366,
451
+ "step": 5900
452
+ },
453
+ {
454
+ "epoch": 7.13,
455
+ "learning_rate": 1.4370546318289787e-05,
456
+ "loss": 1.509,
457
+ "step": 6000
458
+ },
459
+ {
460
+ "epoch": 7.13,
461
+ "eval_loss": 1.4273977279663086,
462
+ "eval_runtime": 33.9883,
463
+ "eval_samples_per_second": 132.016,
464
+ "eval_steps_per_second": 8.268,
465
+ "step": 6000
466
+ },
467
+ {
468
+ "epoch": 7.24,
469
+ "learning_rate": 1.3776722090261283e-05,
470
+ "loss": 1.5425,
471
+ "step": 6100
472
+ },
473
+ {
474
+ "epoch": 7.36,
475
+ "learning_rate": 1.318289786223278e-05,
476
+ "loss": 1.5253,
477
+ "step": 6200
478
+ },
479
+ {
480
+ "epoch": 7.48,
481
+ "learning_rate": 1.2589073634204277e-05,
482
+ "loss": 1.5502,
483
+ "step": 6300
484
+ },
485
+ {
486
+ "epoch": 7.6,
487
+ "learning_rate": 1.1995249406175772e-05,
488
+ "loss": 1.5011,
489
+ "step": 6400
490
+ },
491
+ {
492
+ "epoch": 7.72,
493
+ "learning_rate": 1.1401425178147269e-05,
494
+ "loss": 1.5083,
495
+ "step": 6500
496
+ },
497
+ {
498
+ "epoch": 7.72,
499
+ "eval_loss": 1.410913348197937,
500
+ "eval_runtime": 33.9769,
501
+ "eval_samples_per_second": 132.06,
502
+ "eval_steps_per_second": 8.27,
503
+ "step": 6500
504
+ },
505
+ {
506
+ "epoch": 7.84,
507
+ "learning_rate": 1.0807600950118766e-05,
508
+ "loss": 1.5303,
509
+ "step": 6600
510
+ },
511
+ {
512
+ "epoch": 7.96,
513
+ "learning_rate": 1.0213776722090261e-05,
514
+ "loss": 1.4877,
515
+ "step": 6700
516
+ },
517
+ {
518
+ "epoch": 8.08,
519
+ "learning_rate": 9.619952494061758e-06,
520
+ "loss": 1.5082,
521
+ "step": 6800
522
+ },
523
+ {
524
+ "epoch": 8.19,
525
+ "learning_rate": 9.026128266033253e-06,
526
+ "loss": 1.4751,
527
+ "step": 6900
528
+ },
529
+ {
530
+ "epoch": 8.31,
531
+ "learning_rate": 8.432304038004752e-06,
532
+ "loss": 1.5277,
533
+ "step": 7000
534
+ },
535
+ {
536
+ "epoch": 8.31,
537
+ "eval_loss": 1.4051495790481567,
538
+ "eval_runtime": 33.9346,
539
+ "eval_samples_per_second": 132.225,
540
+ "eval_steps_per_second": 8.281,
541
+ "step": 7000
542
+ },
543
+ {
544
+ "epoch": 8.43,
545
+ "learning_rate": 7.838479809976247e-06,
546
+ "loss": 1.5207,
547
+ "step": 7100
548
+ },
549
+ {
550
+ "epoch": 8.55,
551
+ "learning_rate": 7.244655581947744e-06,
552
+ "loss": 1.4948,
553
+ "step": 7200
554
+ },
555
+ {
556
+ "epoch": 8.67,
557
+ "learning_rate": 6.6508313539192404e-06,
558
+ "loss": 1.4897,
559
+ "step": 7300
560
+ },
561
+ {
562
+ "epoch": 8.79,
563
+ "learning_rate": 6.0570071258907366e-06,
564
+ "loss": 1.4801,
565
+ "step": 7400
566
+ },
567
+ {
568
+ "epoch": 8.91,
569
+ "learning_rate": 5.4631828978622335e-06,
570
+ "loss": 1.4884,
571
+ "step": 7500
572
+ },
573
+ {
574
+ "epoch": 8.91,
575
+ "eval_loss": 1.387204885482788,
576
+ "eval_runtime": 33.9831,
577
+ "eval_samples_per_second": 132.036,
578
+ "eval_steps_per_second": 8.269,
579
+ "step": 7500
580
+ },
581
+ {
582
+ "epoch": 9.03,
583
+ "learning_rate": 4.86935866983373e-06,
584
+ "loss": 1.5037,
585
+ "step": 7600
586
+ },
587
+ {
588
+ "epoch": 9.14,
589
+ "learning_rate": 4.275534441805226e-06,
590
+ "loss": 1.4884,
591
+ "step": 7700
592
+ },
593
+ {
594
+ "epoch": 9.26,
595
+ "learning_rate": 3.681710213776722e-06,
596
+ "loss": 1.4968,
597
+ "step": 7800
598
+ },
599
+ {
600
+ "epoch": 9.38,
601
+ "learning_rate": 3.0878859857482185e-06,
602
+ "loss": 1.4762,
603
+ "step": 7900
604
+ },
605
+ {
606
+ "epoch": 9.5,
607
+ "learning_rate": 2.494061757719715e-06,
608
+ "loss": 1.496,
609
+ "step": 8000
610
+ },
611
+ {
612
+ "epoch": 9.5,
613
+ "eval_loss": 1.3912383317947388,
614
+ "eval_runtime": 33.9905,
615
+ "eval_samples_per_second": 132.007,
616
+ "eval_steps_per_second": 8.267,
617
+ "step": 8000
618
+ },
619
+ {
620
+ "epoch": 9.62,
621
+ "learning_rate": 1.9002375296912116e-06,
622
+ "loss": 1.4703,
623
+ "step": 8100
624
+ },
625
+ {
626
+ "epoch": 9.74,
627
+ "learning_rate": 1.306413301662708e-06,
628
+ "loss": 1.4722,
629
+ "step": 8200
630
+ },
631
+ {
632
+ "epoch": 9.86,
633
+ "learning_rate": 7.125890736342043e-07,
634
+ "loss": 1.4392,
635
+ "step": 8300
636
+ },
637
+ {
638
+ "epoch": 9.98,
639
+ "learning_rate": 1.1876484560570073e-07,
640
+ "loss": 1.4692,
641
+ "step": 8400
642
+ },
643
+ {
644
+ "epoch": 10.0,
645
+ "step": 8420,
646
+ "total_flos": 1.78429300862976e+16,
647
+ "train_loss": 1.7687041851233982,
648
+ "train_runtime": 3488.0926,
649
+ "train_samples_per_second": 38.588,
650
+ "train_steps_per_second": 2.414
651
+ },
652
+ {
653
+ "epoch": 10.0,
654
+ "eval_loss": 1.3867350816726685,
655
+ "eval_runtime": 34.1096,
656
+ "eval_samples_per_second": 131.547,
657
+ "eval_steps_per_second": 8.238,
658
+ "step": 8420
659
+ },
660
+ {
661
+ "epoch": 10.0,
662
+ "eval_loss": 1.3883991241455078,
663
+ "eval_runtime": 34.15,
664
+ "eval_samples_per_second": 131.391,
665
+ "eval_steps_per_second": 8.228,
666
+ "step": 8420
667
+ }
668
+ ],
669
+ "logging_steps": 100,
670
+ "max_steps": 8420,
671
+ "num_train_epochs": 10,
672
+ "save_steps": 500,
673
+ "total_flos": 1.78429300862976e+16,
674
+ "trial_name": null,
675
+ "trial_params": null
676
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73eed6593a28a17abf2080bc8cf3e704f10c1166262f8ea8f075b90496c0cfd5
3
+ size 4600