Xinging commited on
Commit
687a4f4
·
verified ·
1 Parent(s): 859bc3e

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +483 -0
trainer_state.json ADDED
@@ -0,0 +1,483 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 63,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.047619047619047616,
13
+ "grad_norm": 2.8511476516723633,
14
+ "learning_rate": 2.8571428571428573e-06,
15
+ "loss": 1.0485,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.09523809523809523,
20
+ "grad_norm": 2.6251747608184814,
21
+ "learning_rate": 5.7142857142857145e-06,
22
+ "loss": 1.0247,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.14285714285714285,
27
+ "grad_norm": 2.595073699951172,
28
+ "learning_rate": 8.571428571428571e-06,
29
+ "loss": 1.0005,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.19047619047619047,
34
+ "grad_norm": 2.1916372776031494,
35
+ "learning_rate": 1.1428571428571429e-05,
36
+ "loss": 0.9607,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.23809523809523808,
41
+ "grad_norm": 1.1861807107925415,
42
+ "learning_rate": 1.4285714285714287e-05,
43
+ "loss": 0.9474,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.2857142857142857,
48
+ "grad_norm": 1.0186238288879395,
49
+ "learning_rate": 1.7142857142857142e-05,
50
+ "loss": 0.9091,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.3333333333333333,
55
+ "grad_norm": 1.270335078239441,
56
+ "learning_rate": 2e-05,
57
+ "loss": 0.9359,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.38095238095238093,
62
+ "grad_norm": 1.0229175090789795,
63
+ "learning_rate": 1.998426815017817e-05,
64
+ "loss": 0.9072,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.42857142857142855,
69
+ "grad_norm": 0.9742695093154907,
70
+ "learning_rate": 1.9937122098932428e-05,
71
+ "loss": 0.9051,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.47619047619047616,
76
+ "grad_norm": 0.7967286109924316,
77
+ "learning_rate": 1.985871018518236e-05,
78
+ "loss": 0.8882,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.5238095238095238,
83
+ "grad_norm": 0.759879469871521,
84
+ "learning_rate": 1.9749279121818235e-05,
85
+ "loss": 0.894,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.5714285714285714,
90
+ "grad_norm": 0.6994519829750061,
91
+ "learning_rate": 1.9609173219450998e-05,
92
+ "loss": 0.8808,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.6190476190476191,
97
+ "grad_norm": 0.7012657523155212,
98
+ "learning_rate": 1.9438833303083677e-05,
99
+ "loss": 0.8738,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.6666666666666666,
104
+ "grad_norm": 0.7178092002868652,
105
+ "learning_rate": 1.9238795325112867e-05,
106
+ "loss": 0.9113,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.7142857142857143,
111
+ "grad_norm": 0.6520066261291504,
112
+ "learning_rate": 1.900968867902419e-05,
113
+ "loss": 0.8581,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.7619047619047619,
118
+ "grad_norm": 0.6900270581245422,
119
+ "learning_rate": 1.8752234219087538e-05,
120
+ "loss": 0.8969,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.8095238095238095,
125
+ "grad_norm": 0.6860277056694031,
126
+ "learning_rate": 1.8467241992282842e-05,
127
+ "loss": 0.906,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.8571428571428571,
132
+ "grad_norm": 0.6491706371307373,
133
+ "learning_rate": 1.8155608689592604e-05,
134
+ "loss": 0.8859,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.9047619047619048,
139
+ "grad_norm": 0.6634758710861206,
140
+ "learning_rate": 1.78183148246803e-05,
141
+ "loss": 0.8931,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.9523809523809523,
146
+ "grad_norm": 0.6383916139602661,
147
+ "learning_rate": 1.7456421648831658e-05,
148
+ "loss": 0.8946,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 1.0,
153
+ "grad_norm": 0.7505225539207458,
154
+ "learning_rate": 1.7071067811865477e-05,
155
+ "loss": 0.8441,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 1.0476190476190477,
160
+ "grad_norm": 0.9270818829536438,
161
+ "learning_rate": 1.6663465779520042e-05,
162
+ "loss": 0.6841,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 1.0952380952380953,
167
+ "grad_norm": 0.9149900674819946,
168
+ "learning_rate": 1.6234898018587336e-05,
169
+ "loss": 0.6804,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 1.1428571428571428,
174
+ "grad_norm": 0.8434793949127197,
175
+ "learning_rate": 1.578671296179806e-05,
176
+ "loss": 0.6712,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 1.1904761904761905,
181
+ "grad_norm": 0.8060764670372009,
182
+ "learning_rate": 1.5320320765153367e-05,
183
+ "loss": 0.6592,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 1.2380952380952381,
188
+ "grad_norm": 0.8197386264801025,
189
+ "learning_rate": 1.4837188871052399e-05,
190
+ "loss": 0.6547,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 1.2857142857142856,
195
+ "grad_norm": 0.7730692625045776,
196
+ "learning_rate": 1.4338837391175582e-05,
197
+ "loss": 0.6555,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 1.3333333333333333,
202
+ "grad_norm": 0.7399494647979736,
203
+ "learning_rate": 1.3826834323650899e-05,
204
+ "loss": 0.6326,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 1.380952380952381,
209
+ "grad_norm": 0.7800103425979614,
210
+ "learning_rate": 1.3302790619551673e-05,
211
+ "loss": 0.6659,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 1.4285714285714286,
216
+ "grad_norm": 0.6559262871742249,
217
+ "learning_rate": 1.2768355114248493e-05,
218
+ "loss": 0.6172,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 1.4761904761904763,
223
+ "grad_norm": 0.7337315678596497,
224
+ "learning_rate": 1.2225209339563144e-05,
225
+ "loss": 0.6394,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 1.5238095238095237,
230
+ "grad_norm": 0.6704586744308472,
231
+ "learning_rate": 1.1675062233047365e-05,
232
+ "loss": 0.6278,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 1.5714285714285714,
237
+ "grad_norm": 0.6687719225883484,
238
+ "learning_rate": 1.1119644761033079e-05,
239
+ "loss": 0.6447,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 1.619047619047619,
244
+ "grad_norm": 0.674872100353241,
245
+ "learning_rate": 1.0560704472371919e-05,
246
+ "loss": 0.6293,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 1.6666666666666665,
251
+ "grad_norm": 0.63092041015625,
252
+ "learning_rate": 1e-05,
253
+ "loss": 0.6135,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 1.7142857142857144,
258
+ "grad_norm": 0.6361706256866455,
259
+ "learning_rate": 9.439295527628083e-06,
260
+ "loss": 0.5951,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 1.7619047619047619,
265
+ "grad_norm": 0.6201922297477722,
266
+ "learning_rate": 8.880355238966923e-06,
267
+ "loss": 0.6291,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 1.8095238095238095,
272
+ "grad_norm": 0.6166723370552063,
273
+ "learning_rate": 8.324937766952638e-06,
274
+ "loss": 0.6515,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 1.8571428571428572,
279
+ "grad_norm": 0.6040306687355042,
280
+ "learning_rate": 7.774790660436857e-06,
281
+ "loss": 0.6229,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 1.9047619047619047,
286
+ "grad_norm": 0.5973318815231323,
287
+ "learning_rate": 7.2316448857515076e-06,
288
+ "loss": 0.634,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 1.9523809523809523,
293
+ "grad_norm": 0.6414107084274292,
294
+ "learning_rate": 6.697209380448333e-06,
295
+ "loss": 0.6351,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 2.0,
300
+ "grad_norm": 0.7759326696395874,
301
+ "learning_rate": 6.173165676349103e-06,
302
+ "loss": 0.5002,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 2.0476190476190474,
307
+ "grad_norm": 0.9337152242660522,
308
+ "learning_rate": 5.66116260882442e-06,
309
+ "loss": 0.4895,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 2.0952380952380953,
314
+ "grad_norm": 0.6880698204040527,
315
+ "learning_rate": 5.1628111289476025e-06,
316
+ "loss": 0.4941,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 2.142857142857143,
321
+ "grad_norm": 0.7491559386253357,
322
+ "learning_rate": 4.679679234846636e-06,
323
+ "loss": 0.4848,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 2.1904761904761907,
328
+ "grad_norm": 0.9208627939224243,
329
+ "learning_rate": 4.213287038201943e-06,
330
+ "loss": 0.4795,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 2.238095238095238,
335
+ "grad_norm": 0.8678593039512634,
336
+ "learning_rate": 3.7651019814126656e-06,
337
+ "loss": 0.4889,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 2.2857142857142856,
342
+ "grad_norm": 0.680047869682312,
343
+ "learning_rate": 3.3365342204799613e-06,
344
+ "loss": 0.4689,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 2.3333333333333335,
349
+ "grad_norm": 0.6818625926971436,
350
+ "learning_rate": 2.9289321881345257e-06,
351
+ "loss": 0.4845,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 2.380952380952381,
356
+ "grad_norm": 0.691443145275116,
357
+ "learning_rate": 2.5435783511683444e-06,
358
+ "loss": 0.4718,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 2.4285714285714284,
363
+ "grad_norm": 0.6897311806678772,
364
+ "learning_rate": 2.1816851753197023e-06,
365
+ "loss": 0.4781,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 2.4761904761904763,
370
+ "grad_norm": 0.6840229034423828,
371
+ "learning_rate": 1.8443913104073984e-06,
372
+ "loss": 0.459,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 2.5238095238095237,
377
+ "grad_norm": 0.6606849431991577,
378
+ "learning_rate": 1.5327580077171589e-06,
379
+ "loss": 0.4772,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 2.571428571428571,
384
+ "grad_norm": 0.6099680066108704,
385
+ "learning_rate": 1.2477657809124632e-06,
386
+ "loss": 0.4674,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 2.619047619047619,
391
+ "grad_norm": 0.6196358799934387,
392
+ "learning_rate": 9.903113209758098e-07,
393
+ "loss": 0.4697,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 2.6666666666666665,
398
+ "grad_norm": 0.6188323497772217,
399
+ "learning_rate": 7.612046748871327e-07,
400
+ "loss": 0.4661,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 2.7142857142857144,
405
+ "grad_norm": 0.5998156070709229,
406
+ "learning_rate": 5.611666969163243e-07,
407
+ "loss": 0.4517,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 2.761904761904762,
412
+ "grad_norm": 0.6303263902664185,
413
+ "learning_rate": 3.908267805490051e-07,
414
+ "loss": 0.4688,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 2.8095238095238093,
419
+ "grad_norm": 0.61989825963974,
420
+ "learning_rate": 2.507208781817638e-07,
421
+ "loss": 0.4674,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 2.857142857142857,
426
+ "grad_norm": 0.639975905418396,
427
+ "learning_rate": 1.4128981481764115e-07,
428
+ "loss": 0.4591,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 2.9047619047619047,
433
+ "grad_norm": 0.6036660671234131,
434
+ "learning_rate": 6.287790106757396e-08,
435
+ "loss": 0.4704,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 2.9523809523809526,
440
+ "grad_norm": 0.6197323203086853,
441
+ "learning_rate": 1.5731849821833955e-08,
442
+ "loss": 0.4716,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 3.0,
447
+ "grad_norm": 0.6455105543136597,
448
+ "learning_rate": 0.0,
449
+ "loss": 0.4279,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 3.0,
454
+ "step": 63,
455
+ "total_flos": 3.895140077000458e+17,
456
+ "train_loss": 0.674688827896875,
457
+ "train_runtime": 982.3694,
458
+ "train_samples_per_second": 15.88,
459
+ "train_steps_per_second": 0.064
460
+ }
461
+ ],
462
+ "logging_steps": 1,
463
+ "max_steps": 63,
464
+ "num_input_tokens_seen": 0,
465
+ "num_train_epochs": 3,
466
+ "save_steps": 0,
467
+ "stateful_callbacks": {
468
+ "TrainerControl": {
469
+ "args": {
470
+ "should_epoch_stop": false,
471
+ "should_evaluate": false,
472
+ "should_log": false,
473
+ "should_save": true,
474
+ "should_training_stop": true
475
+ },
476
+ "attributes": {}
477
+ }
478
+ },
479
+ "total_flos": 3.895140077000458e+17,
480
+ "train_batch_size": 64,
481
+ "trial_name": null,
482
+ "trial_params": null
483
+ }