loris3 commited on
Commit
db9b4a0
·
verified ·
1 Parent(s): fe16858

Upload folder using huggingface_hub

Browse files
checkpoints/checkpoint-4698/config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "RobertaForMaskedLM"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bos_token_id": 0,
7
+ "classifier_dropout": null,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "layer_norm_eps": 1e-05,
15
+ "max_position_embeddings": 514,
16
+ "model_type": "roberta",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "pad_token_id": 1,
20
+ "position_embedding_type": "absolute",
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.47.0",
23
+ "type_vocab_size": 1,
24
+ "use_cache": true,
25
+ "vocab_size": 52000
26
+ }
checkpoints/checkpoint-4698/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfdb80802bb4487b1bcc80844aed115e957f360d0223e52a96e304b1c7cb3fca
3
+ size 504150808
checkpoints/checkpoint-4698/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a5d86554aab9881798e1baf84ad58ac68af2659fcef3ce83681685c2f995c85
3
+ size 1008422138
checkpoints/checkpoint-4698/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:612575457ce237eed527a43d141019c4023c55c699b964811b6ad45e44a699ed
3
+ size 14244
checkpoints/checkpoint-4698/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af8ba91ad15c249834f7635df590537c2b6081b9bdc64bb9977a548bd26e6874
3
+ size 1064
checkpoints/checkpoint-4698/trainer_state.json ADDED
@@ -0,0 +1,774 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 8.998804113848362,
5
+ "eval_steps": 500,
6
+ "global_step": 4698,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.09567089213106912,
13
+ "grad_norm": 52.15945816040039,
14
+ "learning_rate": 2.5e-06,
15
+ "loss": 10.7426,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.19134178426213824,
20
+ "grad_norm": 38.14910125732422,
21
+ "learning_rate": 5e-06,
22
+ "loss": 10.0104,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.2870126763932074,
27
+ "grad_norm": 39.01994705200195,
28
+ "learning_rate": 7.5e-06,
29
+ "loss": 9.6031,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.3826835685242765,
34
+ "grad_norm": 33.76708984375,
35
+ "learning_rate": 1e-05,
36
+ "loss": 9.1839,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.4783544606553456,
41
+ "grad_norm": 35.43233108520508,
42
+ "learning_rate": 1.25e-05,
43
+ "loss": 8.7209,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.5740253527864148,
48
+ "grad_norm": 29.526994705200195,
49
+ "learning_rate": 1.5e-05,
50
+ "loss": 8.2324,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.6696962449174838,
55
+ "grad_norm": 35.25413131713867,
56
+ "learning_rate": 1.7500000000000002e-05,
57
+ "loss": 7.7545,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.765367137048553,
62
+ "grad_norm": 40.43191909790039,
63
+ "learning_rate": 2e-05,
64
+ "loss": 7.3315,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.8610380291796221,
69
+ "grad_norm": 30.023365020751953,
70
+ "learning_rate": 2.2499999999999998e-05,
71
+ "loss": 6.963,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.9567089213106912,
76
+ "grad_norm": 31.878944396972656,
77
+ "learning_rate": 2.5e-05,
78
+ "loss": 6.7297,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.9988041138483617,
83
+ "eval_accuracy": 0.0,
84
+ "eval_loss": 6.555630207061768,
85
+ "eval_normalizer": 94012.0,
86
+ "eval_runtime": 80.2813,
87
+ "eval_samples_per_second": 665.871,
88
+ "eval_steps_per_second": 1.308,
89
+ "step": 522
90
+ },
91
+ {
92
+ "epoch": 1.0535756995933987,
93
+ "grad_norm": 23.939258575439453,
94
+ "learning_rate": 2.75e-05,
95
+ "loss": 6.5661,
96
+ "step": 550
97
+ },
98
+ {
99
+ "epoch": 1.1492465917244679,
100
+ "grad_norm": 28.73314666748047,
101
+ "learning_rate": 3e-05,
102
+ "loss": 6.4297,
103
+ "step": 600
104
+ },
105
+ {
106
+ "epoch": 1.244917483855537,
107
+ "grad_norm": 25.804365158081055,
108
+ "learning_rate": 3.2500000000000004e-05,
109
+ "loss": 6.3161,
110
+ "step": 650
111
+ },
112
+ {
113
+ "epoch": 1.3405883759866062,
114
+ "grad_norm": 28.974842071533203,
115
+ "learning_rate": 3.5000000000000004e-05,
116
+ "loss": 6.231,
117
+ "step": 700
118
+ },
119
+ {
120
+ "epoch": 1.4362592681176751,
121
+ "grad_norm": 30.538211822509766,
122
+ "learning_rate": 3.75e-05,
123
+ "loss": 6.1787,
124
+ "step": 750
125
+ },
126
+ {
127
+ "epoch": 1.5319301602487443,
128
+ "grad_norm": 22.466585159301758,
129
+ "learning_rate": 4e-05,
130
+ "loss": 6.1109,
131
+ "step": 800
132
+ },
133
+ {
134
+ "epoch": 1.6276010523798135,
135
+ "grad_norm": 25.896324157714844,
136
+ "learning_rate": 4.25e-05,
137
+ "loss": 6.0128,
138
+ "step": 850
139
+ },
140
+ {
141
+ "epoch": 1.7232719445108826,
142
+ "grad_norm": 26.904653549194336,
143
+ "learning_rate": 4.4999999999999996e-05,
144
+ "loss": 5.9551,
145
+ "step": 900
146
+ },
147
+ {
148
+ "epoch": 1.8189428366419516,
149
+ "grad_norm": 26.616758346557617,
150
+ "learning_rate": 4.75e-05,
151
+ "loss": 5.8732,
152
+ "step": 950
153
+ },
154
+ {
155
+ "epoch": 1.9146137287730207,
156
+ "grad_norm": 30.21148109436035,
157
+ "learning_rate": 5e-05,
158
+ "loss": 5.7962,
159
+ "step": 1000
160
+ },
161
+ {
162
+ "epoch": 1.9988041138483617,
163
+ "eval_accuracy": 0.0,
164
+ "eval_loss": 5.6822733879089355,
165
+ "eval_normalizer": 93558.0,
166
+ "eval_runtime": 78.5931,
167
+ "eval_samples_per_second": 680.174,
168
+ "eval_steps_per_second": 1.336,
169
+ "step": 1044
170
+ },
171
+ {
172
+ "epoch": 2.0114805070557282,
173
+ "grad_norm": 29.994068145751953,
174
+ "learning_rate": 5.25e-05,
175
+ "loss": 5.7392,
176
+ "step": 1050
177
+ },
178
+ {
179
+ "epoch": 2.1071513991867974,
180
+ "grad_norm": 35.065006256103516,
181
+ "learning_rate": 5.5e-05,
182
+ "loss": 5.6481,
183
+ "step": 1100
184
+ },
185
+ {
186
+ "epoch": 2.2028222913178666,
187
+ "grad_norm": 37.09373092651367,
188
+ "learning_rate": 5.75e-05,
189
+ "loss": 5.5441,
190
+ "step": 1150
191
+ },
192
+ {
193
+ "epoch": 2.2984931834489357,
194
+ "grad_norm": 32.272159576416016,
195
+ "learning_rate": 6e-05,
196
+ "loss": 5.4931,
197
+ "step": 1200
198
+ },
199
+ {
200
+ "epoch": 2.394164075580005,
201
+ "grad_norm": 27.136716842651367,
202
+ "learning_rate": 6.25e-05,
203
+ "loss": 5.4324,
204
+ "step": 1250
205
+ },
206
+ {
207
+ "epoch": 2.489834967711074,
208
+ "grad_norm": 28.377527236938477,
209
+ "learning_rate": 6.500000000000001e-05,
210
+ "loss": 5.3281,
211
+ "step": 1300
212
+ },
213
+ {
214
+ "epoch": 2.5855058598421428,
215
+ "grad_norm": 25.35126495361328,
216
+ "learning_rate": 6.75e-05,
217
+ "loss": 5.2737,
218
+ "step": 1350
219
+ },
220
+ {
221
+ "epoch": 2.6811767519732124,
222
+ "grad_norm": 27.91370964050293,
223
+ "learning_rate": 7.000000000000001e-05,
224
+ "loss": 5.1901,
225
+ "step": 1400
226
+ },
227
+ {
228
+ "epoch": 2.776847644104281,
229
+ "grad_norm": 24.87859344482422,
230
+ "learning_rate": 7.25e-05,
231
+ "loss": 5.141,
232
+ "step": 1450
233
+ },
234
+ {
235
+ "epoch": 2.8725185362353503,
236
+ "grad_norm": 27.901790618896484,
237
+ "learning_rate": 7.5e-05,
238
+ "loss": 5.0599,
239
+ "step": 1500
240
+ },
241
+ {
242
+ "epoch": 2.9681894283664194,
243
+ "grad_norm": 25.183063507080078,
244
+ "learning_rate": 7.75e-05,
245
+ "loss": 5.034,
246
+ "step": 1550
247
+ },
248
+ {
249
+ "epoch": 2.9988041138483617,
250
+ "eval_accuracy": 0.0,
251
+ "eval_loss": 4.937503337860107,
252
+ "eval_normalizer": 94216.0,
253
+ "eval_runtime": 75.5016,
254
+ "eval_samples_per_second": 708.024,
255
+ "eval_steps_per_second": 1.391,
256
+ "step": 1566
257
+ },
258
+ {
259
+ "epoch": 3.065056206649127,
260
+ "grad_norm": 26.59418487548828,
261
+ "learning_rate": 8e-05,
262
+ "loss": 4.9628,
263
+ "step": 1600
264
+ },
265
+ {
266
+ "epoch": 3.160727098780196,
267
+ "grad_norm": 25.259281158447266,
268
+ "learning_rate": 8.25e-05,
269
+ "loss": 4.9004,
270
+ "step": 1650
271
+ },
272
+ {
273
+ "epoch": 3.2563979909112652,
274
+ "grad_norm": 25.84400749206543,
275
+ "learning_rate": 8.5e-05,
276
+ "loss": 4.8452,
277
+ "step": 1700
278
+ },
279
+ {
280
+ "epoch": 3.3520688830423344,
281
+ "grad_norm": 25.52320671081543,
282
+ "learning_rate": 8.75e-05,
283
+ "loss": 4.7968,
284
+ "step": 1750
285
+ },
286
+ {
287
+ "epoch": 3.4477397751734036,
288
+ "grad_norm": 26.29749298095703,
289
+ "learning_rate": 8.999999999999999e-05,
290
+ "loss": 4.7601,
291
+ "step": 1800
292
+ },
293
+ {
294
+ "epoch": 3.5434106673044727,
295
+ "grad_norm": 27.535959243774414,
296
+ "learning_rate": 9.25e-05,
297
+ "loss": 4.7155,
298
+ "step": 1850
299
+ },
300
+ {
301
+ "epoch": 3.639081559435542,
302
+ "grad_norm": 23.676984786987305,
303
+ "learning_rate": 9.5e-05,
304
+ "loss": 4.6651,
305
+ "step": 1900
306
+ },
307
+ {
308
+ "epoch": 3.734752451566611,
309
+ "grad_norm": 22.59883689880371,
310
+ "learning_rate": 9.750000000000001e-05,
311
+ "loss": 4.6527,
312
+ "step": 1950
313
+ },
314
+ {
315
+ "epoch": 3.83042334369768,
316
+ "grad_norm": 21.733461380004883,
317
+ "learning_rate": 0.0001,
318
+ "loss": 4.616,
319
+ "step": 2000
320
+ },
321
+ {
322
+ "epoch": 3.926094235828749,
323
+ "grad_norm": 25.904829025268555,
324
+ "learning_rate": 0.0001025,
325
+ "loss": 4.5769,
326
+ "step": 2050
327
+ },
328
+ {
329
+ "epoch": 3.9988041138483617,
330
+ "eval_accuracy": 0.0,
331
+ "eval_loss": 4.510604381561279,
332
+ "eval_normalizer": 93665.0,
333
+ "eval_runtime": 76.2941,
334
+ "eval_samples_per_second": 700.67,
335
+ "eval_steps_per_second": 1.376,
336
+ "step": 2088
337
+ },
338
+ {
339
+ "epoch": 4.0229610141114565,
340
+ "grad_norm": 21.005359649658203,
341
+ "learning_rate": 0.000105,
342
+ "loss": 4.5525,
343
+ "step": 2100
344
+ },
345
+ {
346
+ "epoch": 4.118631906242526,
347
+ "grad_norm": 25.174129486083984,
348
+ "learning_rate": 0.0001075,
349
+ "loss": 4.4938,
350
+ "step": 2150
351
+ },
352
+ {
353
+ "epoch": 4.214302798373595,
354
+ "grad_norm": 22.09128189086914,
355
+ "learning_rate": 0.00011,
356
+ "loss": 4.4503,
357
+ "step": 2200
358
+ },
359
+ {
360
+ "epoch": 4.309973690504664,
361
+ "grad_norm": 21.93778419494629,
362
+ "learning_rate": 0.00011250000000000001,
363
+ "loss": 4.4222,
364
+ "step": 2250
365
+ },
366
+ {
367
+ "epoch": 4.405644582635733,
368
+ "grad_norm": 22.780986785888672,
369
+ "learning_rate": 0.000115,
370
+ "loss": 4.388,
371
+ "step": 2300
372
+ },
373
+ {
374
+ "epoch": 4.501315474766802,
375
+ "grad_norm": 24.103681564331055,
376
+ "learning_rate": 0.0001175,
377
+ "loss": 4.3805,
378
+ "step": 2350
379
+ },
380
+ {
381
+ "epoch": 4.596986366897871,
382
+ "grad_norm": 27.335912704467773,
383
+ "learning_rate": 0.00012,
384
+ "loss": 4.3188,
385
+ "step": 2400
386
+ },
387
+ {
388
+ "epoch": 4.69265725902894,
389
+ "grad_norm": 21.6763916015625,
390
+ "learning_rate": 0.0001225,
391
+ "loss": 4.2918,
392
+ "step": 2450
393
+ },
394
+ {
395
+ "epoch": 4.78832815116001,
396
+ "grad_norm": 24.06027603149414,
397
+ "learning_rate": 0.000125,
398
+ "loss": 4.2723,
399
+ "step": 2500
400
+ },
401
+ {
402
+ "epoch": 4.8839990432910785,
403
+ "grad_norm": 23.394180297851562,
404
+ "learning_rate": 0.0001275,
405
+ "loss": 4.2586,
406
+ "step": 2550
407
+ },
408
+ {
409
+ "epoch": 4.979669935422148,
410
+ "grad_norm": 25.303680419921875,
411
+ "learning_rate": 0.00013000000000000002,
412
+ "loss": 4.2298,
413
+ "step": 2600
414
+ },
415
+ {
416
+ "epoch": 4.998804113848362,
417
+ "eval_accuracy": 0.0,
418
+ "eval_loss": 4.226888656616211,
419
+ "eval_normalizer": 94313.0,
420
+ "eval_runtime": 77.1025,
421
+ "eval_samples_per_second": 693.324,
422
+ "eval_steps_per_second": 1.362,
423
+ "step": 2610
424
+ },
425
+ {
426
+ "epoch": 5.076536713704855,
427
+ "grad_norm": 23.319316864013672,
428
+ "learning_rate": 0.00013250000000000002,
429
+ "loss": 4.2049,
430
+ "step": 2650
431
+ },
432
+ {
433
+ "epoch": 5.172207605835925,
434
+ "grad_norm": 27.484468460083008,
435
+ "learning_rate": 0.000135,
436
+ "loss": 4.1758,
437
+ "step": 2700
438
+ },
439
+ {
440
+ "epoch": 5.2678784979669935,
441
+ "grad_norm": 25.021984100341797,
442
+ "learning_rate": 0.0001375,
443
+ "loss": 4.1389,
444
+ "step": 2750
445
+ },
446
+ {
447
+ "epoch": 5.363549390098063,
448
+ "grad_norm": 24.843420028686523,
449
+ "learning_rate": 0.00014000000000000001,
450
+ "loss": 4.123,
451
+ "step": 2800
452
+ },
453
+ {
454
+ "epoch": 5.459220282229132,
455
+ "grad_norm": 22.530752182006836,
456
+ "learning_rate": 0.0001425,
457
+ "loss": 4.1117,
458
+ "step": 2850
459
+ },
460
+ {
461
+ "epoch": 5.5548911743602005,
462
+ "grad_norm": 23.811887741088867,
463
+ "learning_rate": 0.000145,
464
+ "loss": 4.0883,
465
+ "step": 2900
466
+ },
467
+ {
468
+ "epoch": 5.65056206649127,
469
+ "grad_norm": 22.435691833496094,
470
+ "learning_rate": 0.0001475,
471
+ "loss": 4.0598,
472
+ "step": 2950
473
+ },
474
+ {
475
+ "epoch": 5.746232958622339,
476
+ "grad_norm": 22.16675567626953,
477
+ "learning_rate": 0.00015,
478
+ "loss": 4.0562,
479
+ "step": 3000
480
+ },
481
+ {
482
+ "epoch": 5.8419038507534085,
483
+ "grad_norm": 26.84369468688965,
484
+ "learning_rate": 0.0001525,
485
+ "loss": 4.0424,
486
+ "step": 3050
487
+ },
488
+ {
489
+ "epoch": 5.937574742884477,
490
+ "grad_norm": 22.534196853637695,
491
+ "learning_rate": 0.000155,
492
+ "loss": 4.0289,
493
+ "step": 3100
494
+ },
495
+ {
496
+ "epoch": 5.998804113848362,
497
+ "eval_accuracy": 0.0,
498
+ "eval_loss": 4.000634670257568,
499
+ "eval_normalizer": 93901.0,
500
+ "eval_runtime": 76.1749,
501
+ "eval_samples_per_second": 701.767,
502
+ "eval_steps_per_second": 1.378,
503
+ "step": 3132
504
+ },
505
+ {
506
+ "epoch": 6.034441521167185,
507
+ "grad_norm": 22.28015899658203,
508
+ "learning_rate": 0.0001575,
509
+ "loss": 4.0049,
510
+ "step": 3150
511
+ },
512
+ {
513
+ "epoch": 6.130112413298254,
514
+ "grad_norm": 22.31985855102539,
515
+ "learning_rate": 0.00016,
516
+ "loss": 3.9543,
517
+ "step": 3200
518
+ },
519
+ {
520
+ "epoch": 6.2257833054293235,
521
+ "grad_norm": 26.62820053100586,
522
+ "learning_rate": 0.00016250000000000002,
523
+ "loss": 3.96,
524
+ "step": 3250
525
+ },
526
+ {
527
+ "epoch": 6.321454197560392,
528
+ "grad_norm": 22.462852478027344,
529
+ "learning_rate": 0.000165,
530
+ "loss": 3.9324,
531
+ "step": 3300
532
+ },
533
+ {
534
+ "epoch": 6.417125089691462,
535
+ "grad_norm": 21.36258888244629,
536
+ "learning_rate": 0.0001675,
537
+ "loss": 3.9276,
538
+ "step": 3350
539
+ },
540
+ {
541
+ "epoch": 6.5127959818225305,
542
+ "grad_norm": 21.731992721557617,
543
+ "learning_rate": 0.00017,
544
+ "loss": 3.8995,
545
+ "step": 3400
546
+ },
547
+ {
548
+ "epoch": 6.608466873953599,
549
+ "grad_norm": 21.62281608581543,
550
+ "learning_rate": 0.0001725,
551
+ "loss": 3.9031,
552
+ "step": 3450
553
+ },
554
+ {
555
+ "epoch": 6.704137766084669,
556
+ "grad_norm": 22.04605484008789,
557
+ "learning_rate": 0.000175,
558
+ "loss": 3.8636,
559
+ "step": 3500
560
+ },
561
+ {
562
+ "epoch": 6.7998086582157375,
563
+ "grad_norm": 22.352691650390625,
564
+ "learning_rate": 0.0001775,
565
+ "loss": 3.8825,
566
+ "step": 3550
567
+ },
568
+ {
569
+ "epoch": 6.895479550346807,
570
+ "grad_norm": 23.3765811920166,
571
+ "learning_rate": 0.00017999999999999998,
572
+ "loss": 3.842,
573
+ "step": 3600
574
+ },
575
+ {
576
+ "epoch": 6.991150442477876,
577
+ "grad_norm": 21.487628936767578,
578
+ "learning_rate": 0.0001825,
579
+ "loss": 3.8673,
580
+ "step": 3650
581
+ },
582
+ {
583
+ "epoch": 6.998804113848362,
584
+ "eval_accuracy": 0.0,
585
+ "eval_loss": 3.867459774017334,
586
+ "eval_normalizer": 94462.0,
587
+ "eval_runtime": 77.3784,
588
+ "eval_samples_per_second": 690.852,
589
+ "eval_steps_per_second": 1.357,
590
+ "step": 3654
591
+ },
592
+ {
593
+ "epoch": 7.088017220760584,
594
+ "grad_norm": 21.705764770507812,
595
+ "learning_rate": 0.000185,
596
+ "loss": 3.794,
597
+ "step": 3700
598
+ },
599
+ {
600
+ "epoch": 7.1836881128916525,
601
+ "grad_norm": 21.813831329345703,
602
+ "learning_rate": 0.0001875,
603
+ "loss": 3.8044,
604
+ "step": 3750
605
+ },
606
+ {
607
+ "epoch": 7.279359005022722,
608
+ "grad_norm": 23.83123779296875,
609
+ "learning_rate": 0.00019,
610
+ "loss": 3.7978,
611
+ "step": 3800
612
+ },
613
+ {
614
+ "epoch": 7.375029897153791,
615
+ "grad_norm": 22.62957000732422,
616
+ "learning_rate": 0.00019250000000000002,
617
+ "loss": 3.7845,
618
+ "step": 3850
619
+ },
620
+ {
621
+ "epoch": 7.4707007892848605,
622
+ "grad_norm": 23.31735610961914,
623
+ "learning_rate": 0.00019500000000000002,
624
+ "loss": 3.7775,
625
+ "step": 3900
626
+ },
627
+ {
628
+ "epoch": 7.566371681415929,
629
+ "grad_norm": 21.118228912353516,
630
+ "learning_rate": 0.0001975,
631
+ "loss": 3.7552,
632
+ "step": 3950
633
+ },
634
+ {
635
+ "epoch": 7.662042573546998,
636
+ "grad_norm": 21.326833724975586,
637
+ "learning_rate": 0.0002,
638
+ "loss": 3.7476,
639
+ "step": 4000
640
+ },
641
+ {
642
+ "epoch": 7.7577134656780675,
643
+ "grad_norm": 23.456077575683594,
644
+ "learning_rate": 0.00020250000000000002,
645
+ "loss": 3.7478,
646
+ "step": 4050
647
+ },
648
+ {
649
+ "epoch": 7.853384357809136,
650
+ "grad_norm": 23.339223861694336,
651
+ "learning_rate": 0.000205,
652
+ "loss": 3.7394,
653
+ "step": 4100
654
+ },
655
+ {
656
+ "epoch": 7.949055249940206,
657
+ "grad_norm": 22.789596557617188,
658
+ "learning_rate": 0.0002075,
659
+ "loss": 3.7111,
660
+ "step": 4150
661
+ },
662
+ {
663
+ "epoch": 7.998804113848362,
664
+ "eval_accuracy": 0.0,
665
+ "eval_loss": 3.691098928451538,
666
+ "eval_normalizer": 93833.0,
667
+ "eval_runtime": 76.6695,
668
+ "eval_samples_per_second": 697.239,
669
+ "eval_steps_per_second": 1.37,
670
+ "step": 4176
671
+ },
672
+ {
673
+ "epoch": 8.045922028222913,
674
+ "grad_norm": 21.58089256286621,
675
+ "learning_rate": 0.00021,
676
+ "loss": 3.7079,
677
+ "step": 4200
678
+ },
679
+ {
680
+ "epoch": 8.141592920353983,
681
+ "grad_norm": 20.514217376708984,
682
+ "learning_rate": 0.0002125,
683
+ "loss": 3.6871,
684
+ "step": 4250
685
+ },
686
+ {
687
+ "epoch": 8.237263812485052,
688
+ "grad_norm": 23.554931640625,
689
+ "learning_rate": 0.000215,
690
+ "loss": 3.6725,
691
+ "step": 4300
692
+ },
693
+ {
694
+ "epoch": 8.33293470461612,
695
+ "grad_norm": 22.691059112548828,
696
+ "learning_rate": 0.0002175,
697
+ "loss": 3.6517,
698
+ "step": 4350
699
+ },
700
+ {
701
+ "epoch": 8.42860559674719,
702
+ "grad_norm": 22.110071182250977,
703
+ "learning_rate": 0.00022,
704
+ "loss": 3.6511,
705
+ "step": 4400
706
+ },
707
+ {
708
+ "epoch": 8.52427648887826,
709
+ "grad_norm": 19.855382919311523,
710
+ "learning_rate": 0.00022250000000000001,
711
+ "loss": 3.6681,
712
+ "step": 4450
713
+ },
714
+ {
715
+ "epoch": 8.619947381009329,
716
+ "grad_norm": 22.0894832611084,
717
+ "learning_rate": 0.00022500000000000002,
718
+ "loss": 3.6507,
719
+ "step": 4500
720
+ },
721
+ {
722
+ "epoch": 8.715618273140397,
723
+ "grad_norm": 23.990575790405273,
724
+ "learning_rate": 0.0002275,
725
+ "loss": 3.6393,
726
+ "step": 4550
727
+ },
728
+ {
729
+ "epoch": 8.811289165271466,
730
+ "grad_norm": 21.28339385986328,
731
+ "learning_rate": 0.00023,
732
+ "loss": 3.6315,
733
+ "step": 4600
734
+ },
735
+ {
736
+ "epoch": 8.906960057402536,
737
+ "grad_norm": 22.376495361328125,
738
+ "learning_rate": 0.0002325,
739
+ "loss": 3.6439,
740
+ "step": 4650
741
+ },
742
+ {
743
+ "epoch": 8.998804113848362,
744
+ "eval_accuracy": 0.0,
745
+ "eval_loss": 3.6127943992614746,
746
+ "eval_normalizer": 93564.0,
747
+ "eval_runtime": 76.7151,
748
+ "eval_samples_per_second": 696.825,
749
+ "eval_steps_per_second": 1.369,
750
+ "step": 4698
751
+ }
752
+ ],
753
+ "logging_steps": 50,
754
+ "max_steps": 5220,
755
+ "num_input_tokens_seen": 0,
756
+ "num_train_epochs": 10,
757
+ "save_steps": 500,
758
+ "stateful_callbacks": {
759
+ "TrainerControl": {
760
+ "args": {
761
+ "should_epoch_stop": false,
762
+ "should_evaluate": false,
763
+ "should_log": false,
764
+ "should_save": true,
765
+ "should_training_stop": false
766
+ },
767
+ "attributes": {}
768
+ }
769
+ },
770
+ "total_flos": 7.444425289468969e+17,
771
+ "train_batch_size": 128,
772
+ "trial_name": null,
773
+ "trial_params": null
774
+ }
checkpoints/checkpoint-4698/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24a7da9fd672d10200544391c3e51faa9174aebcae93fda1b807e8cae1690006
3
+ size 5368