Hugging Face's logo

loris3
/
stratified_10m_curriculum_roberta_source_difficulty

Fill-Mask
Transformers
Safetensors
roberta
Model card Files
xet
Community
loris3 commited on
Commit
8794235
·
verified ·
1 Parent(s): 13a3817

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. checkpoints/checkpoint-3547/config.json +26 -0
  2. checkpoints/checkpoint-3547/model.safetensors +3 -0
  3. checkpoints/checkpoint-3547/optimizer.pt +3 -0
  4. checkpoints/checkpoint-3547/rng_state.pth +3 -0
  5. checkpoints/checkpoint-3547/scheduler.pt +3 -0
  6. checkpoints/checkpoint-3547/trainer_state.json +613 -0
  7. checkpoints/checkpoint-3547/training_args.bin +3 -0
checkpoints/checkpoint-3547/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-3547/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:522afeba0cb0a838410a288e868f4aed615fab6b1b07ee0d28baec9d7650357d
3
+ size 504150808
checkpoints/checkpoint-3547/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27058c9c3a1f12262c8fe4fe1bc3f934af2306346468f75cb4bd093f0851995c
3
+ size 1008422138
checkpoints/checkpoint-3547/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:456cf69b46a20e993d6228a6d24dcab12f0918f75ad56d48fde80b8519d37780
3
+ size 14244
checkpoints/checkpoint-3547/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3724e2474b0b161c6c99ba411821b3262cedfb3b0ac7a168f5df6ccc62d99804
3
+ size 1064
checkpoints/checkpoint-3547/trainer_state.json ADDED
@@ -0,0 +1,613 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 8.564458263573307,
5
+ "eval_steps": 500,
6
+ "global_step": 3547,
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": 79.64751434326172,
14
+ "learning_rate": 2.5e-06,
15
+ "loss": 10.5359,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.19134178426213824,
20
+ "grad_norm": 57.142730712890625,
21
+ "learning_rate": 5e-06,
22
+ "loss": 9.4534,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.2870126763932074,
27
+ "grad_norm": 56.33840560913086,
28
+ "learning_rate": 7.5e-06,
29
+ "loss": 8.7762,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.3826835685242765,
34
+ "grad_norm": 48.39895248413086,
35
+ "learning_rate": 1e-05,
36
+ "loss": 8.0812,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.4783544606553456,
41
+ "grad_norm": 51.66913604736328,
42
+ "learning_rate": 1.25e-05,
43
+ "loss": 7.4712,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.5740253527864148,
48
+ "grad_norm": 39.3389778137207,
49
+ "learning_rate": 1.5e-05,
50
+ "loss": 6.9249,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.6696962449174838,
55
+ "grad_norm": 42.176029205322266,
56
+ "learning_rate": 1.7500000000000002e-05,
57
+ "loss": 6.3373,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.765367137048553,
62
+ "grad_norm": 27.88520622253418,
63
+ "learning_rate": 2e-05,
64
+ "loss": 5.8156,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.8610380291796221,
69
+ "grad_norm": 27.722829818725586,
70
+ "learning_rate": 2.2499999999999998e-05,
71
+ "loss": 5.362,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.9567089213106912,
76
+ "grad_norm": 28.680429458618164,
77
+ "learning_rate": 2.5e-05,
78
+ "loss": 5.0434,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.9988041138483617,
83
+ "eval_accuracy": 0.0,
84
+ "eval_loss": 8.436083793640137,
85
+ "eval_normalizer": 94012.0,
86
+ "eval_runtime": 81.5272,
87
+ "eval_samples_per_second": 655.695,
88
+ "eval_steps_per_second": 1.288,
89
+ "step": 522
90
+ },
91
+ {
92
+ "epoch": 1.0535756995933987,
93
+ "grad_norm": 25.760271072387695,
94
+ "learning_rate": 2.75e-05,
95
+ "loss": 4.8276,
96
+ "step": 550
97
+ },
98
+ {
99
+ "epoch": 1.1492465917244679,
100
+ "grad_norm": 27.668893814086914,
101
+ "learning_rate": 3e-05,
102
+ "loss": 4.6499,
103
+ "step": 600
104
+ },
105
+ {
106
+ "epoch": 1.244917483855537,
107
+ "grad_norm": 27.842025756835938,
108
+ "learning_rate": 3.2500000000000004e-05,
109
+ "loss": 4.4827,
110
+ "step": 650
111
+ },
112
+ {
113
+ "epoch": 1.3405883759866062,
114
+ "grad_norm": 30.099782943725586,
115
+ "learning_rate": 3.5000000000000004e-05,
116
+ "loss": 4.3643,
117
+ "step": 700
118
+ },
119
+ {
120
+ "epoch": 1.4362592681176751,
121
+ "grad_norm": 33.141204833984375,
122
+ "learning_rate": 3.75e-05,
123
+ "loss": 4.2279,
124
+ "step": 750
125
+ },
126
+ {
127
+ "epoch": 1.5319301602487443,
128
+ "grad_norm": 37.47197723388672,
129
+ "learning_rate": 4e-05,
130
+ "loss": 4.1339,
131
+ "step": 800
132
+ },
133
+ {
134
+ "epoch": 1.6276010523798135,
135
+ "grad_norm": 34.994571685791016,
136
+ "learning_rate": 4.25e-05,
137
+ "loss": 4.0134,
138
+ "step": 850
139
+ },
140
+ {
141
+ "epoch": 1.7232719445108826,
142
+ "grad_norm": 29.075597763061523,
143
+ "learning_rate": 4.4999999999999996e-05,
144
+ "loss": 3.9216,
145
+ "step": 900
146
+ },
147
+ {
148
+ "epoch": 1.8189428366419516,
149
+ "grad_norm": 31.81198501586914,
150
+ "learning_rate": 4.75e-05,
151
+ "loss": 3.8193,
152
+ "step": 950
153
+ },
154
+ {
155
+ "epoch": 1.9146137287730207,
156
+ "grad_norm": 33.48542785644531,
157
+ "learning_rate": 5e-05,
158
+ "loss": 3.7218,
159
+ "step": 1000
160
+ },
161
+ {
162
+ "epoch": 1.9988041138483617,
163
+ "eval_accuracy": 0.0,
164
+ "eval_loss": 8.203332901000977,
165
+ "eval_normalizer": 93558.0,
166
+ "eval_runtime": 83.028,
167
+ "eval_samples_per_second": 643.843,
168
+ "eval_steps_per_second": 1.265,
169
+ "step": 1044
170
+ },
171
+ {
172
+ "epoch": 2.0114805070557282,
173
+ "grad_norm": 84.52313232421875,
174
+ "learning_rate": 5.25e-05,
175
+ "loss": 4.1746,
176
+ "step": 1050
177
+ },
178
+ {
179
+ "epoch": 2.1071513991867974,
180
+ "grad_norm": 32.48819351196289,
181
+ "learning_rate": 5.5e-05,
182
+ "loss": 6.9906,
183
+ "step": 1100
184
+ },
185
+ {
186
+ "epoch": 2.2028222913178666,
187
+ "grad_norm": 20.257213592529297,
188
+ "learning_rate": 5.75e-05,
189
+ "loss": 6.1631,
190
+ "step": 1150
191
+ },
192
+ {
193
+ "epoch": 2.2984931834489357,
194
+ "grad_norm": 20.544626235961914,
195
+ "learning_rate": 6e-05,
196
+ "loss": 5.787,
197
+ "step": 1200
198
+ },
199
+ {
200
+ "epoch": 2.3597225544128198,
201
+ "eval_accuracy": 0.0,
202
+ "eval_loss": 6.199265480041504,
203
+ "eval_normalizer": 94216.0,
204
+ "eval_runtime": 83.3702,
205
+ "eval_samples_per_second": 641.201,
206
+ "eval_steps_per_second": 1.259,
207
+ "step": 1232
208
+ },
209
+ {
210
+ "epoch": 3.0344415211671847,
211
+ "grad_norm": 23.196285247802734,
212
+ "learning_rate": 6.25e-05,
213
+ "loss": 5.59,
214
+ "step": 1250
215
+ },
216
+ {
217
+ "epoch": 3.130112413298254,
218
+ "grad_norm": 19.782928466796875,
219
+ "learning_rate": 6.500000000000001e-05,
220
+ "loss": 5.2666,
221
+ "step": 1300
222
+ },
223
+ {
224
+ "epoch": 3.225783305429323,
225
+ "grad_norm": 20.39004898071289,
226
+ "learning_rate": 6.75e-05,
227
+ "loss": 5.0723,
228
+ "step": 1350
229
+ },
230
+ {
231
+ "epoch": 3.321454197560392,
232
+ "grad_norm": 28.565898895263672,
233
+ "learning_rate": 7.000000000000001e-05,
234
+ "loss": 4.909,
235
+ "step": 1400
236
+ },
237
+ {
238
+ "epoch": 3.3597225544128198,
239
+ "eval_accuracy": 0.0,
240
+ "eval_loss": 5.842504978179932,
241
+ "eval_normalizer": 93665.0,
242
+ "eval_runtime": 82.9992,
243
+ "eval_samples_per_second": 644.067,
244
+ "eval_steps_per_second": 1.265,
245
+ "step": 1420
246
+ },
247
+ {
248
+ "epoch": 4.057402535278642,
249
+ "grad_norm": 27.73941993713379,
250
+ "learning_rate": 7.25e-05,
251
+ "loss": 5.0404,
252
+ "step": 1450
253
+ },
254
+ {
255
+ "epoch": 4.15307342740971,
256
+ "grad_norm": 25.40974998474121,
257
+ "learning_rate": 7.5e-05,
258
+ "loss": 4.6658,
259
+ "step": 1500
260
+ },
261
+ {
262
+ "epoch": 4.24874431954078,
263
+ "grad_norm": 29.910818099975586,
264
+ "learning_rate": 7.75e-05,
265
+ "loss": 4.5082,
266
+ "step": 1550
267
+ },
268
+ {
269
+ "epoch": 4.344415211671849,
270
+ "grad_norm": 26.495317459106445,
271
+ "learning_rate": 8e-05,
272
+ "loss": 4.3851,
273
+ "step": 1600
274
+ },
275
+ {
276
+ "epoch": 4.440086103802918,
277
+ "grad_norm": 28.781644821166992,
278
+ "learning_rate": 8.25e-05,
279
+ "loss": 4.263,
280
+ "step": 1650
281
+ },
282
+ {
283
+ "epoch": 4.535756995933987,
284
+ "grad_norm": 31.134675979614258,
285
+ "learning_rate": 8.5e-05,
286
+ "loss": 4.2074,
287
+ "step": 1700
288
+ },
289
+ {
290
+ "epoch": 4.631427888065057,
291
+ "grad_norm": 24.8262882232666,
292
+ "learning_rate": 8.75e-05,
293
+ "loss": 4.1359,
294
+ "step": 1750
295
+ },
296
+ {
297
+ "epoch": 4.727098780196125,
298
+ "grad_norm": 23.926483154296875,
299
+ "learning_rate": 8.999999999999999e-05,
300
+ "loss": 4.067,
301
+ "step": 1800
302
+ },
303
+ {
304
+ "epoch": 4.822769672327194,
305
+ "grad_norm": 24.742311477661133,
306
+ "learning_rate": 9.25e-05,
307
+ "loss": 4.0219,
308
+ "step": 1850
309
+ },
310
+ {
311
+ "epoch": 4.918440564458264,
312
+ "grad_norm": 27.34615135192871,
313
+ "learning_rate": 9.5e-05,
314
+ "loss": 4.0033,
315
+ "step": 1900
316
+ },
317
+ {
318
+ "epoch": 4.998804113848362,
319
+ "eval_accuracy": 0.0,
320
+ "eval_loss": 5.4360456466674805,
321
+ "eval_normalizer": 94313.0,
322
+ "eval_runtime": 104.459,
323
+ "eval_samples_per_second": 511.751,
324
+ "eval_steps_per_second": 1.005,
325
+ "step": 1942
326
+ },
327
+ {
328
+ "epoch": 5.015307342740971,
329
+ "grad_norm": 24.026140213012695,
330
+ "learning_rate": 9.750000000000001e-05,
331
+ "loss": 3.9511,
332
+ "step": 1950
333
+ },
334
+ {
335
+ "epoch": 5.11097823487204,
336
+ "grad_norm": 26.587495803833008,
337
+ "learning_rate": 0.0001,
338
+ "loss": 3.9194,
339
+ "step": 2000
340
+ },
341
+ {
342
+ "epoch": 5.206649127003109,
343
+ "grad_norm": 23.23008918762207,
344
+ "learning_rate": 0.0001025,
345
+ "loss": 3.8788,
346
+ "step": 2050
347
+ },
348
+ {
349
+ "epoch": 5.302320019134179,
350
+ "grad_norm": 22.73567771911621,
351
+ "learning_rate": 0.000105,
352
+ "loss": 3.8894,
353
+ "step": 2100
354
+ },
355
+ {
356
+ "epoch": 5.397990911265247,
357
+ "grad_norm": 26.093669891357422,
358
+ "learning_rate": 0.0001075,
359
+ "loss": 3.8171,
360
+ "step": 2150
361
+ },
362
+ {
363
+ "epoch": 5.493661803396317,
364
+ "grad_norm": 26.150381088256836,
365
+ "learning_rate": 0.00011,
366
+ "loss": 3.7666,
367
+ "step": 2200
368
+ },
369
+ {
370
+ "epoch": 5.589332695527386,
371
+ "grad_norm": 28.061275482177734,
372
+ "learning_rate": 0.00011250000000000001,
373
+ "loss": 3.7926,
374
+ "step": 2250
375
+ },
376
+ {
377
+ "epoch": 5.685003587658455,
378
+ "grad_norm": 24.63544464111328,
379
+ "learning_rate": 0.000115,
380
+ "loss": 3.7454,
381
+ "step": 2300
382
+ },
383
+ {
384
+ "epoch": 5.780674479789524,
385
+ "grad_norm": 24.13852310180664,
386
+ "learning_rate": 0.0001175,
387
+ "loss": 3.7038,
388
+ "step": 2350
389
+ },
390
+ {
391
+ "epoch": 5.876345371920593,
392
+ "grad_norm": 23.41834259033203,
393
+ "learning_rate": 0.00012,
394
+ "loss": 3.6888,
395
+ "step": 2400
396
+ },
397
+ {
398
+ "epoch": 5.972016264051662,
399
+ "grad_norm": 26.9339656829834,
400
+ "learning_rate": 0.0001225,
401
+ "loss": 3.6689,
402
+ "step": 2450
403
+ },
404
+ {
405
+ "epoch": 5.998804113848362,
406
+ "eval_accuracy": 0.0,
407
+ "eval_loss": 5.285378456115723,
408
+ "eval_normalizer": 93901.0,
409
+ "eval_runtime": 102.6182,
410
+ "eval_samples_per_second": 520.931,
411
+ "eval_steps_per_second": 1.023,
412
+ "step": 2464
413
+ },
414
+ {
415
+ "epoch": 6.068883042334369,
416
+ "grad_norm": 18.920269012451172,
417
+ "learning_rate": 0.000125,
418
+ "loss": 4.4993,
419
+ "step": 2500
420
+ },
421
+ {
422
+ "epoch": 6.164553934465439,
423
+ "grad_norm": 19.811311721801758,
424
+ "learning_rate": 0.0001275,
425
+ "loss": 4.6066,
426
+ "step": 2550
427
+ },
428
+ {
429
+ "epoch": 6.260224826596508,
430
+ "grad_norm": 23.79183578491211,
431
+ "learning_rate": 0.00013000000000000002,
432
+ "loss": 4.4607,
433
+ "step": 2600
434
+ },
435
+ {
436
+ "epoch": 6.355895718727577,
437
+ "grad_norm": 19.32040786743164,
438
+ "learning_rate": 0.00013250000000000002,
439
+ "loss": 4.3629,
440
+ "step": 2650
441
+ },
442
+ {
443
+ "epoch": 6.451566610858646,
444
+ "grad_norm": 18.420616149902344,
445
+ "learning_rate": 0.000135,
446
+ "loss": 4.2841,
447
+ "step": 2700
448
+ },
449
+ {
450
+ "epoch": 6.547237502989716,
451
+ "grad_norm": 24.54909324645996,
452
+ "learning_rate": 0.0001375,
453
+ "loss": 4.2057,
454
+ "step": 2750
455
+ },
456
+ {
457
+ "epoch": 6.642908395120784,
458
+ "grad_norm": 20.581602096557617,
459
+ "learning_rate": 0.00014000000000000001,
460
+ "loss": 4.1175,
461
+ "step": 2800
462
+ },
463
+ {
464
+ "epoch": 6.738579287251854,
465
+ "grad_norm": 19.419635772705078,
466
+ "learning_rate": 0.0001425,
467
+ "loss": 4.0771,
468
+ "step": 2850
469
+ },
470
+ {
471
+ "epoch": 6.753886629992825,
472
+ "eval_accuracy": 0.0,
473
+ "eval_loss": 5.031363487243652,
474
+ "eval_normalizer": 94462.0,
475
+ "eval_runtime": 99.5317,
476
+ "eval_samples_per_second": 537.085,
477
+ "eval_steps_per_second": 1.055,
478
+ "step": 2858
479
+ },
480
+ {
481
+ "epoch": 7.080363549390098,
482
+ "grad_norm": 22.94988250732422,
483
+ "learning_rate": 0.000145,
484
+ "loss": 4.0966,
485
+ "step": 2900
486
+ },
487
+ {
488
+ "epoch": 7.176034441521167,
489
+ "grad_norm": 19.95139503479004,
490
+ "learning_rate": 0.0001475,
491
+ "loss": 3.9988,
492
+ "step": 2950
493
+ },
494
+ {
495
+ "epoch": 7.271705333652236,
496
+ "grad_norm": 20.52120590209961,
497
+ "learning_rate": 0.00015,
498
+ "loss": 3.9693,
499
+ "step": 3000
500
+ },
501
+ {
502
+ "epoch": 7.367376225783305,
503
+ "grad_norm": 20.660249710083008,
504
+ "learning_rate": 0.0001525,
505
+ "loss": 3.8935,
506
+ "step": 3050
507
+ },
508
+ {
509
+ "epoch": 7.463047117914375,
510
+ "grad_norm": 19.104249954223633,
511
+ "learning_rate": 0.000155,
512
+ "loss": 3.8809,
513
+ "step": 3100
514
+ },
515
+ {
516
+ "epoch": 7.558718010045443,
517
+ "grad_norm": 19.59981346130371,
518
+ "learning_rate": 0.0001575,
519
+ "loss": 3.8089,
520
+ "step": 3150
521
+ },
522
+ {
523
+ "epoch": 7.654388902176513,
524
+ "grad_norm": 22.192567825317383,
525
+ "learning_rate": 0.00016,
526
+ "loss": 3.7668,
527
+ "step": 3200
528
+ },
529
+ {
530
+ "epoch": 7.750059794307582,
531
+ "grad_norm": 20.03333282470703,
532
+ "learning_rate": 0.00016250000000000002,
533
+ "loss": 3.7527,
534
+ "step": 3250
535
+ },
536
+ {
537
+ "epoch": 7.753886629992825,
538
+ "eval_accuracy": 0.0,
539
+ "eval_loss": 4.870262145996094,
540
+ "eval_normalizer": 93833.0,
541
+ "eval_runtime": 91.744,
542
+ "eval_samples_per_second": 582.675,
543
+ "eval_steps_per_second": 1.144,
544
+ "step": 3252
545
+ },
546
+ {
547
+ "epoch": 8.091844056445826,
548
+ "grad_norm": 23.52303123474121,
549
+ "learning_rate": 0.000165,
550
+ "loss": 5.2813,
551
+ "step": 3300
552
+ },
553
+ {
554
+ "epoch": 8.187514948576895,
555
+ "grad_norm": 22.477983474731445,
556
+ "learning_rate": 0.0001675,
557
+ "loss": 5.0441,
558
+ "step": 3350
559
+ },
560
+ {
561
+ "epoch": 8.283185840707965,
562
+ "grad_norm": 20.042709350585938,
563
+ "learning_rate": 0.00017,
564
+ "loss": 4.8798,
565
+ "step": 3400
566
+ },
567
+ {
568
+ "epoch": 8.378856732839033,
569
+ "grad_norm": 22.01651954650879,
570
+ "learning_rate": 0.0001725,
571
+ "loss": 4.7684,
572
+ "step": 3450
573
+ },
574
+ {
575
+ "epoch": 8.474527624970102,
576
+ "grad_norm": 23.033382415771484,
577
+ "learning_rate": 0.000175,
578
+ "loss": 4.687,
579
+ "step": 3500
580
+ },
581
+ {
582
+ "epoch": 8.564458263573307,
583
+ "eval_accuracy": 0.0,
584
+ "eval_loss": 4.597339153289795,
585
+ "eval_normalizer": 93564.0,
586
+ "eval_runtime": 80.9177,
587
+ "eval_samples_per_second": 660.634,
588
+ "eval_steps_per_second": 1.298,
589
+ "step": 3547
590
+ }
591
+ ],
592
+ "logging_steps": 50,
593
+ "max_steps": 5220,
594
+ "num_input_tokens_seen": 0,
595
+ "num_train_epochs": 10,
596
+ "save_steps": 500,
597
+ "stateful_callbacks": {
598
+ "TrainerControl": {
599
+ "args": {
600
+ "should_epoch_stop": false,
601
+ "should_evaluate": false,
602
+ "should_log": false,
603
+ "should_save": true,
604
+ "should_training_stop": false
605
+ },
606
+ "attributes": {}
607
+ }
608
+ },
609
+ "total_flos": 3.712812581630738e+17,
610
+ "train_batch_size": 128,
611
+ "trial_name": null,
612
+ "trial_params": null
613
+ }
checkpoints/checkpoint-3547/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe7e211af70ccdcbef5785bd97518ac7bfd13ca661e2594863d6da57a849ebea
3
+ size 5432