barek2k2 commited on
Commit
1284eab
·
1 Parent(s): 6d299ed

Upload fine-tuned BERT model for HIPAA-sensitive DB schema detection

Browse files
.DS_Store ADDED
Binary file (6.15 kB). View file
 
checkpoint-1044/config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "bert-base-uncased",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "gradient_checkpointing": false,
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-12,
15
+ "max_position_embeddings": 512,
16
+ "model_type": "bert",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "pad_token_id": 0,
20
+ "position_embedding_type": "absolute",
21
+ "problem_type": "single_label_classification",
22
+ "torch_dtype": "float32",
23
+ "transformers_version": "4.48.1",
24
+ "type_vocab_size": 2,
25
+ "use_cache": true,
26
+ "vocab_size": 30522
27
+ }
checkpoint-1044/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9cb61e7eedeb7a38d96c4018e43f99e8716537f87b0b791b0c2a3061fe8ed07
3
+ size 437958648
checkpoint-1044/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:240d1827088543a51edb1df51f1915a6a2568b8362b898c8a8434b3aeeea9468
3
+ size 876032762
checkpoint-1044/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1057e177fad7ca76f4331923531a2db285ec84cb6a578a5354e58ec41f3e61f8
3
+ size 13990
checkpoint-1044/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b045beb5af1ba92789500f83fe272e18919cf39afaba301635688670c80d57f2
3
+ size 1064
checkpoint-1044/special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
checkpoint-1044/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1044/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "pad_token": "[PAD]",
51
+ "sep_token": "[SEP]",
52
+ "strip_accents": null,
53
+ "tokenize_chinese_chars": true,
54
+ "tokenizer_class": "BertTokenizer",
55
+ "unk_token": "[UNK]"
56
+ }
checkpoint-1044/trainer_state.json ADDED
@@ -0,0 +1,793 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.028283841907978058,
3
+ "best_model_checkpoint": "./bert_sensitive_columns/checkpoint-1044",
4
+ "epoch": 4.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1044,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.038314176245210725,
13
+ "grad_norm": 6.37937593460083,
14
+ "learning_rate": 9.923371647509579e-06,
15
+ "loss": 0.7375,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.07662835249042145,
20
+ "grad_norm": 4.292376518249512,
21
+ "learning_rate": 9.846743295019157e-06,
22
+ "loss": 0.6499,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.11494252873563218,
27
+ "grad_norm": 6.317875385284424,
28
+ "learning_rate": 9.770114942528738e-06,
29
+ "loss": 0.6161,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.1532567049808429,
34
+ "grad_norm": 7.314127445220947,
35
+ "learning_rate": 9.693486590038314e-06,
36
+ "loss": 0.5928,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.19157088122605365,
41
+ "grad_norm": 9.533055305480957,
42
+ "learning_rate": 9.616858237547894e-06,
43
+ "loss": 0.5842,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.22988505747126436,
48
+ "grad_norm": 10.419981956481934,
49
+ "learning_rate": 9.54022988505747e-06,
50
+ "loss": 0.539,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.2681992337164751,
55
+ "grad_norm": 12.050772666931152,
56
+ "learning_rate": 9.46360153256705e-06,
57
+ "loss": 0.486,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.3065134099616858,
62
+ "grad_norm": 6.0962958335876465,
63
+ "learning_rate": 9.386973180076629e-06,
64
+ "loss": 0.4488,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.3448275862068966,
69
+ "grad_norm": 9.836843490600586,
70
+ "learning_rate": 9.310344827586207e-06,
71
+ "loss": 0.4442,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.3831417624521073,
76
+ "grad_norm": 10.53395938873291,
77
+ "learning_rate": 9.233716475095786e-06,
78
+ "loss": 0.389,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.421455938697318,
83
+ "grad_norm": 12.318860054016113,
84
+ "learning_rate": 9.157088122605364e-06,
85
+ "loss": 0.3499,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.45977011494252873,
90
+ "grad_norm": 15.045988082885742,
91
+ "learning_rate": 9.080459770114942e-06,
92
+ "loss": 0.3011,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.49808429118773945,
97
+ "grad_norm": 11.415493965148926,
98
+ "learning_rate": 9.003831417624522e-06,
99
+ "loss": 0.3427,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.5363984674329502,
104
+ "grad_norm": 12.154045104980469,
105
+ "learning_rate": 8.9272030651341e-06,
106
+ "loss": 0.2912,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.5747126436781609,
111
+ "grad_norm": 12.373332977294922,
112
+ "learning_rate": 8.85057471264368e-06,
113
+ "loss": 0.291,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.6130268199233716,
118
+ "grad_norm": 11.356131553649902,
119
+ "learning_rate": 8.773946360153257e-06,
120
+ "loss": 0.3067,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.6513409961685823,
125
+ "grad_norm": 14.210739135742188,
126
+ "learning_rate": 8.697318007662836e-06,
127
+ "loss": 0.1868,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.6896551724137931,
132
+ "grad_norm": 14.556158065795898,
133
+ "learning_rate": 8.620689655172414e-06,
134
+ "loss": 0.1947,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.7279693486590039,
139
+ "grad_norm": 6.439550399780273,
140
+ "learning_rate": 8.544061302681992e-06,
141
+ "loss": 0.3156,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.7662835249042146,
146
+ "grad_norm": 17.6083984375,
147
+ "learning_rate": 8.467432950191573e-06,
148
+ "loss": 0.2002,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.8045977011494253,
153
+ "grad_norm": 15.961342811584473,
154
+ "learning_rate": 8.390804597701149e-06,
155
+ "loss": 0.13,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.842911877394636,
160
+ "grad_norm": 31.289339065551758,
161
+ "learning_rate": 8.31417624521073e-06,
162
+ "loss": 0.2278,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.8812260536398467,
167
+ "grad_norm": 15.521924018859863,
168
+ "learning_rate": 8.237547892720307e-06,
169
+ "loss": 0.2059,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.9195402298850575,
174
+ "grad_norm": 15.326404571533203,
175
+ "learning_rate": 8.160919540229886e-06,
176
+ "loss": 0.2234,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.9578544061302682,
181
+ "grad_norm": 19.96432876586914,
182
+ "learning_rate": 8.084291187739464e-06,
183
+ "loss": 0.1941,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.9961685823754789,
188
+ "grad_norm": 18.93885040283203,
189
+ "learning_rate": 8.007662835249042e-06,
190
+ "loss": 0.201,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 1.0,
195
+ "eval_loss": 0.11205583065748215,
196
+ "eval_runtime": 1.2974,
197
+ "eval_samples_per_second": 803.139,
198
+ "eval_steps_per_second": 50.871,
199
+ "step": 261
200
+ },
201
+ {
202
+ "epoch": 1.0344827586206897,
203
+ "grad_norm": 9.7966947555542,
204
+ "learning_rate": 7.93103448275862e-06,
205
+ "loss": 0.1023,
206
+ "step": 270
207
+ },
208
+ {
209
+ "epoch": 1.0727969348659003,
210
+ "grad_norm": 19.185115814208984,
211
+ "learning_rate": 7.854406130268199e-06,
212
+ "loss": 0.1514,
213
+ "step": 280
214
+ },
215
+ {
216
+ "epoch": 1.1111111111111112,
217
+ "grad_norm": 15.477436065673828,
218
+ "learning_rate": 7.77777777777778e-06,
219
+ "loss": 0.1278,
220
+ "step": 290
221
+ },
222
+ {
223
+ "epoch": 1.1494252873563218,
224
+ "grad_norm": 1.358453392982483,
225
+ "learning_rate": 7.701149425287356e-06,
226
+ "loss": 0.1164,
227
+ "step": 300
228
+ },
229
+ {
230
+ "epoch": 1.1877394636015326,
231
+ "grad_norm": 19.57685661315918,
232
+ "learning_rate": 7.624521072796936e-06,
233
+ "loss": 0.1216,
234
+ "step": 310
235
+ },
236
+ {
237
+ "epoch": 1.2260536398467432,
238
+ "grad_norm": 8.200814247131348,
239
+ "learning_rate": 7.547892720306514e-06,
240
+ "loss": 0.1565,
241
+ "step": 320
242
+ },
243
+ {
244
+ "epoch": 1.264367816091954,
245
+ "grad_norm": 2.0913987159729004,
246
+ "learning_rate": 7.4712643678160925e-06,
247
+ "loss": 0.1394,
248
+ "step": 330
249
+ },
250
+ {
251
+ "epoch": 1.3026819923371646,
252
+ "grad_norm": 10.067784309387207,
253
+ "learning_rate": 7.394636015325672e-06,
254
+ "loss": 0.133,
255
+ "step": 340
256
+ },
257
+ {
258
+ "epoch": 1.3409961685823755,
259
+ "grad_norm": 27.741924285888672,
260
+ "learning_rate": 7.318007662835249e-06,
261
+ "loss": 0.1483,
262
+ "step": 350
263
+ },
264
+ {
265
+ "epoch": 1.3793103448275863,
266
+ "grad_norm": 56.61354064941406,
267
+ "learning_rate": 7.241379310344828e-06,
268
+ "loss": 0.1156,
269
+ "step": 360
270
+ },
271
+ {
272
+ "epoch": 1.417624521072797,
273
+ "grad_norm": 19.167062759399414,
274
+ "learning_rate": 7.1647509578544075e-06,
275
+ "loss": 0.1101,
276
+ "step": 370
277
+ },
278
+ {
279
+ "epoch": 1.4559386973180077,
280
+ "grad_norm": 24.54031753540039,
281
+ "learning_rate": 7.088122605363985e-06,
282
+ "loss": 0.0848,
283
+ "step": 380
284
+ },
285
+ {
286
+ "epoch": 1.4942528735632183,
287
+ "grad_norm": 0.2951218783855438,
288
+ "learning_rate": 7.011494252873564e-06,
289
+ "loss": 0.062,
290
+ "step": 390
291
+ },
292
+ {
293
+ "epoch": 1.5325670498084292,
294
+ "grad_norm": 0.5242842435836792,
295
+ "learning_rate": 6.934865900383142e-06,
296
+ "loss": 0.1586,
297
+ "step": 400
298
+ },
299
+ {
300
+ "epoch": 1.5708812260536398,
301
+ "grad_norm": 1.8837841749191284,
302
+ "learning_rate": 6.858237547892721e-06,
303
+ "loss": 0.0764,
304
+ "step": 410
305
+ },
306
+ {
307
+ "epoch": 1.6091954022988506,
308
+ "grad_norm": 8.144768714904785,
309
+ "learning_rate": 6.781609195402299e-06,
310
+ "loss": 0.1034,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 1.6475095785440614,
315
+ "grad_norm": 25.173917770385742,
316
+ "learning_rate": 6.7049808429118775e-06,
317
+ "loss": 0.1133,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.685823754789272,
322
+ "grad_norm": 36.27584457397461,
323
+ "learning_rate": 6.628352490421457e-06,
324
+ "loss": 0.1259,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.7241379310344827,
329
+ "grad_norm": 44.28847122192383,
330
+ "learning_rate": 6.551724137931035e-06,
331
+ "loss": 0.0782,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.7624521072796935,
336
+ "grad_norm": 30.58072280883789,
337
+ "learning_rate": 6.475095785440614e-06,
338
+ "loss": 0.1196,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.8007662835249043,
343
+ "grad_norm": 1.082352876663208,
344
+ "learning_rate": 6.398467432950192e-06,
345
+ "loss": 0.1099,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.839080459770115,
350
+ "grad_norm": 0.34228336811065674,
351
+ "learning_rate": 6.321839080459771e-06,
352
+ "loss": 0.0765,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.8773946360153255,
357
+ "grad_norm": 53.12428283691406,
358
+ "learning_rate": 6.24521072796935e-06,
359
+ "loss": 0.1151,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.9157088122605364,
364
+ "grad_norm": 0.43705159425735474,
365
+ "learning_rate": 6.1685823754789275e-06,
366
+ "loss": 0.0485,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.9540229885057472,
371
+ "grad_norm": 1.2276843786239624,
372
+ "learning_rate": 6.091954022988507e-06,
373
+ "loss": 0.0962,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.9923371647509578,
378
+ "grad_norm": 13.329333305358887,
379
+ "learning_rate": 6.015325670498084e-06,
380
+ "loss": 0.0608,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 2.0,
385
+ "eval_loss": 0.06104712933301926,
386
+ "eval_runtime": 1.1775,
387
+ "eval_samples_per_second": 884.96,
388
+ "eval_steps_per_second": 56.053,
389
+ "step": 522
390
+ },
391
+ {
392
+ "epoch": 2.0306513409961684,
393
+ "grad_norm": 1.5714704990386963,
394
+ "learning_rate": 5.938697318007663e-06,
395
+ "loss": 0.0464,
396
+ "step": 530
397
+ },
398
+ {
399
+ "epoch": 2.0689655172413794,
400
+ "grad_norm": 10.750398635864258,
401
+ "learning_rate": 5.862068965517242e-06,
402
+ "loss": 0.0518,
403
+ "step": 540
404
+ },
405
+ {
406
+ "epoch": 2.10727969348659,
407
+ "grad_norm": 26.733074188232422,
408
+ "learning_rate": 5.78544061302682e-06,
409
+ "loss": 0.1253,
410
+ "step": 550
411
+ },
412
+ {
413
+ "epoch": 2.1455938697318007,
414
+ "grad_norm": 7.556675434112549,
415
+ "learning_rate": 5.708812260536399e-06,
416
+ "loss": 0.0313,
417
+ "step": 560
418
+ },
419
+ {
420
+ "epoch": 2.1839080459770113,
421
+ "grad_norm": 24.295198440551758,
422
+ "learning_rate": 5.6321839080459775e-06,
423
+ "loss": 0.0909,
424
+ "step": 570
425
+ },
426
+ {
427
+ "epoch": 2.2222222222222223,
428
+ "grad_norm": 19.714115142822266,
429
+ "learning_rate": 5.555555555555557e-06,
430
+ "loss": 0.0387,
431
+ "step": 580
432
+ },
433
+ {
434
+ "epoch": 2.260536398467433,
435
+ "grad_norm": 0.8200851082801819,
436
+ "learning_rate": 5.478927203065134e-06,
437
+ "loss": 0.0455,
438
+ "step": 590
439
+ },
440
+ {
441
+ "epoch": 2.2988505747126435,
442
+ "grad_norm": 0.183212548494339,
443
+ "learning_rate": 5.402298850574713e-06,
444
+ "loss": 0.0327,
445
+ "step": 600
446
+ },
447
+ {
448
+ "epoch": 2.3371647509578546,
449
+ "grad_norm": 61.08492660522461,
450
+ "learning_rate": 5.3256704980842925e-06,
451
+ "loss": 0.0621,
452
+ "step": 610
453
+ },
454
+ {
455
+ "epoch": 2.375478927203065,
456
+ "grad_norm": 0.16310882568359375,
457
+ "learning_rate": 5.24904214559387e-06,
458
+ "loss": 0.0545,
459
+ "step": 620
460
+ },
461
+ {
462
+ "epoch": 2.413793103448276,
463
+ "grad_norm": 39.58172607421875,
464
+ "learning_rate": 5.172413793103449e-06,
465
+ "loss": 0.0539,
466
+ "step": 630
467
+ },
468
+ {
469
+ "epoch": 2.4521072796934864,
470
+ "grad_norm": 23.265289306640625,
471
+ "learning_rate": 5.095785440613027e-06,
472
+ "loss": 0.1148,
473
+ "step": 640
474
+ },
475
+ {
476
+ "epoch": 2.4904214559386975,
477
+ "grad_norm": 0.07729102671146393,
478
+ "learning_rate": 5.019157088122606e-06,
479
+ "loss": 0.0281,
480
+ "step": 650
481
+ },
482
+ {
483
+ "epoch": 2.528735632183908,
484
+ "grad_norm": 15.417061805725098,
485
+ "learning_rate": 4.942528735632184e-06,
486
+ "loss": 0.0086,
487
+ "step": 660
488
+ },
489
+ {
490
+ "epoch": 2.5670498084291187,
491
+ "grad_norm": 0.08401647210121155,
492
+ "learning_rate": 4.8659003831417625e-06,
493
+ "loss": 0.0334,
494
+ "step": 670
495
+ },
496
+ {
497
+ "epoch": 2.6053639846743293,
498
+ "grad_norm": 2.524700164794922,
499
+ "learning_rate": 4.789272030651342e-06,
500
+ "loss": 0.0206,
501
+ "step": 680
502
+ },
503
+ {
504
+ "epoch": 2.6436781609195403,
505
+ "grad_norm": 0.24694575369358063,
506
+ "learning_rate": 4.71264367816092e-06,
507
+ "loss": 0.0038,
508
+ "step": 690
509
+ },
510
+ {
511
+ "epoch": 2.681992337164751,
512
+ "grad_norm": 0.28223150968551636,
513
+ "learning_rate": 4.636015325670498e-06,
514
+ "loss": 0.0411,
515
+ "step": 700
516
+ },
517
+ {
518
+ "epoch": 2.7203065134099615,
519
+ "grad_norm": 0.08240451663732529,
520
+ "learning_rate": 4.5593869731800775e-06,
521
+ "loss": 0.0744,
522
+ "step": 710
523
+ },
524
+ {
525
+ "epoch": 2.7586206896551726,
526
+ "grad_norm": 10.965692520141602,
527
+ "learning_rate": 4.482758620689656e-06,
528
+ "loss": 0.0412,
529
+ "step": 720
530
+ },
531
+ {
532
+ "epoch": 2.796934865900383,
533
+ "grad_norm": 0.4931705892086029,
534
+ "learning_rate": 4.406130268199234e-06,
535
+ "loss": 0.014,
536
+ "step": 730
537
+ },
538
+ {
539
+ "epoch": 2.835249042145594,
540
+ "grad_norm": 0.066756471991539,
541
+ "learning_rate": 4.3295019157088125e-06,
542
+ "loss": 0.0271,
543
+ "step": 740
544
+ },
545
+ {
546
+ "epoch": 2.873563218390805,
547
+ "grad_norm": 0.06784966588020325,
548
+ "learning_rate": 4.252873563218391e-06,
549
+ "loss": 0.0026,
550
+ "step": 750
551
+ },
552
+ {
553
+ "epoch": 2.9118773946360155,
554
+ "grad_norm": 0.06324368715286255,
555
+ "learning_rate": 4.17624521072797e-06,
556
+ "loss": 0.0961,
557
+ "step": 760
558
+ },
559
+ {
560
+ "epoch": 2.950191570881226,
561
+ "grad_norm": 28.81197166442871,
562
+ "learning_rate": 4.099616858237548e-06,
563
+ "loss": 0.0472,
564
+ "step": 770
565
+ },
566
+ {
567
+ "epoch": 2.9885057471264367,
568
+ "grad_norm": 0.05656365305185318,
569
+ "learning_rate": 4.022988505747127e-06,
570
+ "loss": 0.0078,
571
+ "step": 780
572
+ },
573
+ {
574
+ "epoch": 3.0,
575
+ "eval_loss": 0.028292344883084297,
576
+ "eval_runtime": 1.1733,
577
+ "eval_samples_per_second": 888.059,
578
+ "eval_steps_per_second": 56.249,
579
+ "step": 783
580
+ },
581
+ {
582
+ "epoch": 3.0268199233716473,
583
+ "grad_norm": 42.40280532836914,
584
+ "learning_rate": 3.946360153256705e-06,
585
+ "loss": 0.0335,
586
+ "step": 790
587
+ },
588
+ {
589
+ "epoch": 3.0651340996168583,
590
+ "grad_norm": 0.05205749720335007,
591
+ "learning_rate": 3.869731800766283e-06,
592
+ "loss": 0.0405,
593
+ "step": 800
594
+ },
595
+ {
596
+ "epoch": 3.103448275862069,
597
+ "grad_norm": 0.09792916476726532,
598
+ "learning_rate": 3.793103448275862e-06,
599
+ "loss": 0.0278,
600
+ "step": 810
601
+ },
602
+ {
603
+ "epoch": 3.1417624521072796,
604
+ "grad_norm": 0.05216526985168457,
605
+ "learning_rate": 3.7164750957854412e-06,
606
+ "loss": 0.0041,
607
+ "step": 820
608
+ },
609
+ {
610
+ "epoch": 3.1800766283524906,
611
+ "grad_norm": 0.1886385679244995,
612
+ "learning_rate": 3.6398467432950196e-06,
613
+ "loss": 0.0354,
614
+ "step": 830
615
+ },
616
+ {
617
+ "epoch": 3.218390804597701,
618
+ "grad_norm": 0.0512104369699955,
619
+ "learning_rate": 3.563218390804598e-06,
620
+ "loss": 0.0609,
621
+ "step": 840
622
+ },
623
+ {
624
+ "epoch": 3.256704980842912,
625
+ "grad_norm": 0.34635305404663086,
626
+ "learning_rate": 3.4865900383141767e-06,
627
+ "loss": 0.0506,
628
+ "step": 850
629
+ },
630
+ {
631
+ "epoch": 3.2950191570881224,
632
+ "grad_norm": 11.31212329864502,
633
+ "learning_rate": 3.409961685823755e-06,
634
+ "loss": 0.0429,
635
+ "step": 860
636
+ },
637
+ {
638
+ "epoch": 3.3333333333333335,
639
+ "grad_norm": 0.20496389269828796,
640
+ "learning_rate": 3.3333333333333333e-06,
641
+ "loss": 0.041,
642
+ "step": 870
643
+ },
644
+ {
645
+ "epoch": 3.371647509578544,
646
+ "grad_norm": 9.11811351776123,
647
+ "learning_rate": 3.256704980842912e-06,
648
+ "loss": 0.003,
649
+ "step": 880
650
+ },
651
+ {
652
+ "epoch": 3.4099616858237547,
653
+ "grad_norm": 39.507572174072266,
654
+ "learning_rate": 3.180076628352491e-06,
655
+ "loss": 0.1216,
656
+ "step": 890
657
+ },
658
+ {
659
+ "epoch": 3.4482758620689653,
660
+ "grad_norm": 0.24006924033164978,
661
+ "learning_rate": 3.103448275862069e-06,
662
+ "loss": 0.0022,
663
+ "step": 900
664
+ },
665
+ {
666
+ "epoch": 3.4865900383141764,
667
+ "grad_norm": 0.1950913518667221,
668
+ "learning_rate": 3.026819923371648e-06,
669
+ "loss": 0.0233,
670
+ "step": 910
671
+ },
672
+ {
673
+ "epoch": 3.524904214559387,
674
+ "grad_norm": 0.059875085949897766,
675
+ "learning_rate": 2.9501915708812262e-06,
676
+ "loss": 0.0182,
677
+ "step": 920
678
+ },
679
+ {
680
+ "epoch": 3.5632183908045976,
681
+ "grad_norm": 65.01390838623047,
682
+ "learning_rate": 2.8735632183908046e-06,
683
+ "loss": 0.0522,
684
+ "step": 930
685
+ },
686
+ {
687
+ "epoch": 3.6015325670498086,
688
+ "grad_norm": 0.30775925517082214,
689
+ "learning_rate": 2.796934865900383e-06,
690
+ "loss": 0.0019,
691
+ "step": 940
692
+ },
693
+ {
694
+ "epoch": 3.6398467432950192,
695
+ "grad_norm": 6.678956985473633,
696
+ "learning_rate": 2.720306513409962e-06,
697
+ "loss": 0.0395,
698
+ "step": 950
699
+ },
700
+ {
701
+ "epoch": 3.67816091954023,
702
+ "grad_norm": 0.08484911918640137,
703
+ "learning_rate": 2.6436781609195404e-06,
704
+ "loss": 0.0071,
705
+ "step": 960
706
+ },
707
+ {
708
+ "epoch": 3.716475095785441,
709
+ "grad_norm": 0.0435795895755291,
710
+ "learning_rate": 2.567049808429119e-06,
711
+ "loss": 0.01,
712
+ "step": 970
713
+ },
714
+ {
715
+ "epoch": 3.7547892720306515,
716
+ "grad_norm": 62.425113677978516,
717
+ "learning_rate": 2.4904214559386975e-06,
718
+ "loss": 0.0811,
719
+ "step": 980
720
+ },
721
+ {
722
+ "epoch": 3.793103448275862,
723
+ "grad_norm": 0.05430278554558754,
724
+ "learning_rate": 2.4137931034482762e-06,
725
+ "loss": 0.0021,
726
+ "step": 990
727
+ },
728
+ {
729
+ "epoch": 3.8314176245210727,
730
+ "grad_norm": 0.05314672365784645,
731
+ "learning_rate": 2.3371647509578546e-06,
732
+ "loss": 0.0015,
733
+ "step": 1000
734
+ },
735
+ {
736
+ "epoch": 3.8697318007662833,
737
+ "grad_norm": 0.04625044763088226,
738
+ "learning_rate": 2.260536398467433e-06,
739
+ "loss": 0.0287,
740
+ "step": 1010
741
+ },
742
+ {
743
+ "epoch": 3.9080459770114944,
744
+ "grad_norm": 0.07921384274959564,
745
+ "learning_rate": 2.1839080459770117e-06,
746
+ "loss": 0.0402,
747
+ "step": 1020
748
+ },
749
+ {
750
+ "epoch": 3.946360153256705,
751
+ "grad_norm": 0.07240170985460281,
752
+ "learning_rate": 2.1072796934865904e-06,
753
+ "loss": 0.03,
754
+ "step": 1030
755
+ },
756
+ {
757
+ "epoch": 3.9846743295019156,
758
+ "grad_norm": 0.04859640449285507,
759
+ "learning_rate": 2.0306513409961687e-06,
760
+ "loss": 0.0063,
761
+ "step": 1040
762
+ },
763
+ {
764
+ "epoch": 4.0,
765
+ "eval_loss": 0.028283841907978058,
766
+ "eval_runtime": 1.1847,
767
+ "eval_samples_per_second": 879.559,
768
+ "eval_steps_per_second": 55.711,
769
+ "step": 1044
770
+ }
771
+ ],
772
+ "logging_steps": 10,
773
+ "max_steps": 1305,
774
+ "num_input_tokens_seen": 0,
775
+ "num_train_epochs": 5,
776
+ "save_steps": 500,
777
+ "stateful_callbacks": {
778
+ "TrainerControl": {
779
+ "args": {
780
+ "should_epoch_stop": false,
781
+ "should_evaluate": false,
782
+ "should_log": false,
783
+ "should_save": true,
784
+ "should_training_stop": false
785
+ },
786
+ "attributes": {}
787
+ }
788
+ },
789
+ "total_flos": 111269254200720.0,
790
+ "train_batch_size": 16,
791
+ "trial_name": null,
792
+ "trial_params": null
793
+ }
checkpoint-1044/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:13876956a2e3f1bd8274ebcf3ca3531994346aa809549469b3b4c24335252e18
3
+ size 5240
checkpoint-1044/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2200/config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "bert-base-uncased",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "gradient_checkpointing": false,
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-12,
15
+ "max_position_embeddings": 512,
16
+ "model_type": "bert",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "pad_token_id": 0,
20
+ "position_embedding_type": "absolute",
21
+ "problem_type": "single_label_classification",
22
+ "torch_dtype": "float32",
23
+ "transformers_version": "4.48.1",
24
+ "type_vocab_size": 2,
25
+ "use_cache": true,
26
+ "vocab_size": 30522
27
+ }
checkpoint-2200/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56d5c134de8f55553e1ef47b7e5d9c65c1194ae349cadb381530d2a7cd3ee1a8
3
+ size 437958648
checkpoint-2200/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5ef9c70e3fb3e7f4450dc40e9a69d820727b27f562f94a44c6e96621adb75f0
3
+ size 876032762
checkpoint-2200/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8e39748b1efa39a5c7c195bcbfa712350a4a33bb6044081048e2ce4d4454a96
3
+ size 13990
checkpoint-2200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4188fc5da0c534cf341529f18e31e6ae20321e43fd244c78faadf7adb7115841
3
+ size 1064
checkpoint-2200/special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
checkpoint-2200/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2200/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "pad_token": "[PAD]",
51
+ "sep_token": "[SEP]",
52
+ "strip_accents": null,
53
+ "tokenize_chinese_chars": true,
54
+ "tokenizer_class": "BertTokenizer",
55
+ "unk_token": "[UNK]"
56
+ }
checkpoint-2200/trainer_state.json ADDED
@@ -0,0 +1,1653 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 3.619581912062131e-05,
3
+ "best_model_checkpoint": "./bert_sensitive_columns/checkpoint-2200",
4
+ "epoch": 10.0,
5
+ "eval_steps": 500,
6
+ "global_step": 2200,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.045454545454545456,
13
+ "grad_norm": 8.354640007019043,
14
+ "learning_rate": 2.9863636363636365e-05,
15
+ "loss": 0.6464,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.09090909090909091,
20
+ "grad_norm": 4.68574857711792,
21
+ "learning_rate": 2.972727272727273e-05,
22
+ "loss": 0.6782,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.13636363636363635,
27
+ "grad_norm": 18.99250602722168,
28
+ "learning_rate": 2.959090909090909e-05,
29
+ "loss": 0.5558,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.18181818181818182,
34
+ "grad_norm": 6.562917709350586,
35
+ "learning_rate": 2.9454545454545456e-05,
36
+ "loss": 0.4202,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.22727272727272727,
41
+ "grad_norm": 23.24289321899414,
42
+ "learning_rate": 2.931818181818182e-05,
43
+ "loss": 0.3803,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.2727272727272727,
48
+ "grad_norm": 6.368680953979492,
49
+ "learning_rate": 2.9181818181818185e-05,
50
+ "loss": 0.436,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.3181818181818182,
55
+ "grad_norm": 7.490790367126465,
56
+ "learning_rate": 2.9045454545454546e-05,
57
+ "loss": 0.359,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.36363636363636365,
62
+ "grad_norm": 14.990336418151855,
63
+ "learning_rate": 2.890909090909091e-05,
64
+ "loss": 0.351,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.4090909090909091,
69
+ "grad_norm": 16.408206939697266,
70
+ "learning_rate": 2.8772727272727272e-05,
71
+ "loss": 0.2892,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.45454545454545453,
76
+ "grad_norm": 4.340272903442383,
77
+ "learning_rate": 2.8636363636363637e-05,
78
+ "loss": 0.3178,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.5,
83
+ "grad_norm": 1.4607642889022827,
84
+ "learning_rate": 2.8499999999999998e-05,
85
+ "loss": 0.1817,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.5454545454545454,
90
+ "grad_norm": 70.05758666992188,
91
+ "learning_rate": 2.8363636363636363e-05,
92
+ "loss": 0.1249,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.5909090909090909,
97
+ "grad_norm": 3.1135473251342773,
98
+ "learning_rate": 2.8227272727272727e-05,
99
+ "loss": 0.2626,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.6363636363636364,
104
+ "grad_norm": 11.004677772521973,
105
+ "learning_rate": 2.8090909090909092e-05,
106
+ "loss": 0.2237,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.6818181818181818,
111
+ "grad_norm": 3.861924886703491,
112
+ "learning_rate": 2.7954545454545453e-05,
113
+ "loss": 0.1652,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.7272727272727273,
118
+ "grad_norm": 0.520849883556366,
119
+ "learning_rate": 2.7818181818181818e-05,
120
+ "loss": 0.11,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.7727272727272727,
125
+ "grad_norm": 0.3427947163581848,
126
+ "learning_rate": 2.7681818181818183e-05,
127
+ "loss": 0.0663,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.8181818181818182,
132
+ "grad_norm": 2.3007407188415527,
133
+ "learning_rate": 2.7545454545454547e-05,
134
+ "loss": 0.302,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.8636363636363636,
139
+ "grad_norm": 0.3552773594856262,
140
+ "learning_rate": 2.7409090909090912e-05,
141
+ "loss": 0.1245,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.9090909090909091,
146
+ "grad_norm": 0.14678223431110382,
147
+ "learning_rate": 2.7272727272727273e-05,
148
+ "loss": 0.1086,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.9545454545454546,
153
+ "grad_norm": 0.7700904607772827,
154
+ "learning_rate": 2.7136363636363638e-05,
155
+ "loss": 0.1189,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 1.0,
160
+ "grad_norm": 1.1396052837371826,
161
+ "learning_rate": 2.7000000000000002e-05,
162
+ "loss": 0.1262,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 1.0,
167
+ "eval_loss": 0.08161866664886475,
168
+ "eval_runtime": 1.4104,
169
+ "eval_samples_per_second": 622.523,
170
+ "eval_steps_per_second": 38.996,
171
+ "step": 220
172
+ },
173
+ {
174
+ "epoch": 1.0454545454545454,
175
+ "grad_norm": 5.328164577484131,
176
+ "learning_rate": 2.6863636363636367e-05,
177
+ "loss": 0.0575,
178
+ "step": 230
179
+ },
180
+ {
181
+ "epoch": 1.0909090909090908,
182
+ "grad_norm": 0.3553941547870636,
183
+ "learning_rate": 2.6727272727272728e-05,
184
+ "loss": 0.1262,
185
+ "step": 240
186
+ },
187
+ {
188
+ "epoch": 1.1363636363636362,
189
+ "grad_norm": 0.1090177446603775,
190
+ "learning_rate": 2.6590909090909093e-05,
191
+ "loss": 0.0168,
192
+ "step": 250
193
+ },
194
+ {
195
+ "epoch": 1.1818181818181819,
196
+ "grad_norm": 3.0335617065429688,
197
+ "learning_rate": 2.6454545454545454e-05,
198
+ "loss": 0.0829,
199
+ "step": 260
200
+ },
201
+ {
202
+ "epoch": 1.2272727272727273,
203
+ "grad_norm": 40.902191162109375,
204
+ "learning_rate": 2.631818181818182e-05,
205
+ "loss": 0.0295,
206
+ "step": 270
207
+ },
208
+ {
209
+ "epoch": 1.2727272727272727,
210
+ "grad_norm": 0.07629093527793884,
211
+ "learning_rate": 2.618181818181818e-05,
212
+ "loss": 0.0022,
213
+ "step": 280
214
+ },
215
+ {
216
+ "epoch": 1.3181818181818181,
217
+ "grad_norm": 9.37717056274414,
218
+ "learning_rate": 2.6045454545454545e-05,
219
+ "loss": 0.1626,
220
+ "step": 290
221
+ },
222
+ {
223
+ "epoch": 1.3636363636363638,
224
+ "grad_norm": 0.09757604449987411,
225
+ "learning_rate": 2.590909090909091e-05,
226
+ "loss": 0.0302,
227
+ "step": 300
228
+ },
229
+ {
230
+ "epoch": 1.4090909090909092,
231
+ "grad_norm": 2.172060966491699,
232
+ "learning_rate": 2.5772727272727274e-05,
233
+ "loss": 0.0301,
234
+ "step": 310
235
+ },
236
+ {
237
+ "epoch": 1.4545454545454546,
238
+ "grad_norm": 0.06304845958948135,
239
+ "learning_rate": 2.5636363636363635e-05,
240
+ "loss": 0.0411,
241
+ "step": 320
242
+ },
243
+ {
244
+ "epoch": 1.5,
245
+ "grad_norm": 0.07031694054603577,
246
+ "learning_rate": 2.55e-05,
247
+ "loss": 0.0139,
248
+ "step": 330
249
+ },
250
+ {
251
+ "epoch": 1.5454545454545454,
252
+ "grad_norm": 3.3001744747161865,
253
+ "learning_rate": 2.5363636363636364e-05,
254
+ "loss": 0.1424,
255
+ "step": 340
256
+ },
257
+ {
258
+ "epoch": 1.5909090909090908,
259
+ "grad_norm": 0.08083586394786835,
260
+ "learning_rate": 2.522727272727273e-05,
261
+ "loss": 0.0291,
262
+ "step": 350
263
+ },
264
+ {
265
+ "epoch": 1.6363636363636362,
266
+ "grad_norm": 34.89284133911133,
267
+ "learning_rate": 2.509090909090909e-05,
268
+ "loss": 0.0757,
269
+ "step": 360
270
+ },
271
+ {
272
+ "epoch": 1.6818181818181817,
273
+ "grad_norm": 0.082089863717556,
274
+ "learning_rate": 2.4954545454545455e-05,
275
+ "loss": 0.0143,
276
+ "step": 370
277
+ },
278
+ {
279
+ "epoch": 1.7272727272727273,
280
+ "grad_norm": 0.2883528172969818,
281
+ "learning_rate": 2.481818181818182e-05,
282
+ "loss": 0.0769,
283
+ "step": 380
284
+ },
285
+ {
286
+ "epoch": 1.7727272727272727,
287
+ "grad_norm": 11.552708625793457,
288
+ "learning_rate": 2.4681818181818184e-05,
289
+ "loss": 0.0487,
290
+ "step": 390
291
+ },
292
+ {
293
+ "epoch": 1.8181818181818183,
294
+ "grad_norm": 0.09678918868303299,
295
+ "learning_rate": 2.454545454545455e-05,
296
+ "loss": 0.0612,
297
+ "step": 400
298
+ },
299
+ {
300
+ "epoch": 1.8636363636363638,
301
+ "grad_norm": 0.08423396944999695,
302
+ "learning_rate": 2.440909090909091e-05,
303
+ "loss": 0.0214,
304
+ "step": 410
305
+ },
306
+ {
307
+ "epoch": 1.9090909090909092,
308
+ "grad_norm": 97.11893463134766,
309
+ "learning_rate": 2.4272727272727275e-05,
310
+ "loss": 0.1287,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 1.9545454545454546,
315
+ "grad_norm": 0.04146511107683182,
316
+ "learning_rate": 2.4136363636363636e-05,
317
+ "loss": 0.0202,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 2.0,
322
+ "grad_norm": 0.03929423540830612,
323
+ "learning_rate": 2.4e-05,
324
+ "loss": 0.0047,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 2.0,
329
+ "eval_loss": 0.010260566137731075,
330
+ "eval_runtime": 0.9845,
331
+ "eval_samples_per_second": 891.815,
332
+ "eval_steps_per_second": 55.865,
333
+ "step": 440
334
+ },
335
+ {
336
+ "epoch": 2.0454545454545454,
337
+ "grad_norm": 0.10950633883476257,
338
+ "learning_rate": 2.3863636363636362e-05,
339
+ "loss": 0.0509,
340
+ "step": 450
341
+ },
342
+ {
343
+ "epoch": 2.090909090909091,
344
+ "grad_norm": 0.02540852129459381,
345
+ "learning_rate": 2.3727272727272726e-05,
346
+ "loss": 0.0476,
347
+ "step": 460
348
+ },
349
+ {
350
+ "epoch": 2.1363636363636362,
351
+ "grad_norm": 0.024377569556236267,
352
+ "learning_rate": 2.359090909090909e-05,
353
+ "loss": 0.0356,
354
+ "step": 470
355
+ },
356
+ {
357
+ "epoch": 2.1818181818181817,
358
+ "grad_norm": 0.025098495185375214,
359
+ "learning_rate": 2.3454545454545456e-05,
360
+ "loss": 0.0009,
361
+ "step": 480
362
+ },
363
+ {
364
+ "epoch": 2.227272727272727,
365
+ "grad_norm": 0.06944375485181808,
366
+ "learning_rate": 2.3318181818181817e-05,
367
+ "loss": 0.0193,
368
+ "step": 490
369
+ },
370
+ {
371
+ "epoch": 2.2727272727272725,
372
+ "grad_norm": 28.133596420288086,
373
+ "learning_rate": 2.318181818181818e-05,
374
+ "loss": 0.0339,
375
+ "step": 500
376
+ },
377
+ {
378
+ "epoch": 2.3181818181818183,
379
+ "grad_norm": 0.025236543267965317,
380
+ "learning_rate": 2.3045454545454546e-05,
381
+ "loss": 0.0462,
382
+ "step": 510
383
+ },
384
+ {
385
+ "epoch": 2.3636363636363638,
386
+ "grad_norm": 0.022095683962106705,
387
+ "learning_rate": 2.290909090909091e-05,
388
+ "loss": 0.0248,
389
+ "step": 520
390
+ },
391
+ {
392
+ "epoch": 2.409090909090909,
393
+ "grad_norm": 0.02965674363076687,
394
+ "learning_rate": 2.2772727272727272e-05,
395
+ "loss": 0.0023,
396
+ "step": 530
397
+ },
398
+ {
399
+ "epoch": 2.4545454545454546,
400
+ "grad_norm": 0.02505609020590782,
401
+ "learning_rate": 2.2636363636363637e-05,
402
+ "loss": 0.031,
403
+ "step": 540
404
+ },
405
+ {
406
+ "epoch": 2.5,
407
+ "grad_norm": 0.0254733357578516,
408
+ "learning_rate": 2.25e-05,
409
+ "loss": 0.0019,
410
+ "step": 550
411
+ },
412
+ {
413
+ "epoch": 2.5454545454545454,
414
+ "grad_norm": 0.018397022038698196,
415
+ "learning_rate": 2.2363636363636366e-05,
416
+ "loss": 0.0006,
417
+ "step": 560
418
+ },
419
+ {
420
+ "epoch": 2.590909090909091,
421
+ "grad_norm": 0.01716865971684456,
422
+ "learning_rate": 2.222727272727273e-05,
423
+ "loss": 0.0517,
424
+ "step": 570
425
+ },
426
+ {
427
+ "epoch": 2.6363636363636362,
428
+ "grad_norm": 0.016052110120654106,
429
+ "learning_rate": 2.2090909090909092e-05,
430
+ "loss": 0.0005,
431
+ "step": 580
432
+ },
433
+ {
434
+ "epoch": 2.6818181818181817,
435
+ "grad_norm": 0.025269588455557823,
436
+ "learning_rate": 2.1954545454545457e-05,
437
+ "loss": 0.002,
438
+ "step": 590
439
+ },
440
+ {
441
+ "epoch": 2.7272727272727275,
442
+ "grad_norm": 0.04568961635231972,
443
+ "learning_rate": 2.1818181818181818e-05,
444
+ "loss": 0.036,
445
+ "step": 600
446
+ },
447
+ {
448
+ "epoch": 2.7727272727272725,
449
+ "grad_norm": 0.01690821908414364,
450
+ "learning_rate": 2.1681818181818182e-05,
451
+ "loss": 0.0239,
452
+ "step": 610
453
+ },
454
+ {
455
+ "epoch": 2.8181818181818183,
456
+ "grad_norm": 0.02976076677441597,
457
+ "learning_rate": 2.1545454545454544e-05,
458
+ "loss": 0.0005,
459
+ "step": 620
460
+ },
461
+ {
462
+ "epoch": 2.8636363636363638,
463
+ "grad_norm": 0.15770655870437622,
464
+ "learning_rate": 2.140909090909091e-05,
465
+ "loss": 0.0006,
466
+ "step": 630
467
+ },
468
+ {
469
+ "epoch": 2.909090909090909,
470
+ "grad_norm": 0.011741632595658302,
471
+ "learning_rate": 2.1272727272727273e-05,
472
+ "loss": 0.0004,
473
+ "step": 640
474
+ },
475
+ {
476
+ "epoch": 2.9545454545454546,
477
+ "grad_norm": 0.012714399956166744,
478
+ "learning_rate": 2.1136363636363638e-05,
479
+ "loss": 0.0004,
480
+ "step": 650
481
+ },
482
+ {
483
+ "epoch": 3.0,
484
+ "grad_norm": 0.016626961529254913,
485
+ "learning_rate": 2.1e-05,
486
+ "loss": 0.0383,
487
+ "step": 660
488
+ },
489
+ {
490
+ "epoch": 3.0,
491
+ "eval_loss": 0.0007602364639751613,
492
+ "eval_runtime": 1.0081,
493
+ "eval_samples_per_second": 870.902,
494
+ "eval_steps_per_second": 54.555,
495
+ "step": 660
496
+ },
497
+ {
498
+ "epoch": 3.0454545454545454,
499
+ "grad_norm": 0.010502061806619167,
500
+ "learning_rate": 2.0863636363636363e-05,
501
+ "loss": 0.0004,
502
+ "step": 670
503
+ },
504
+ {
505
+ "epoch": 3.090909090909091,
506
+ "grad_norm": 0.011059875600039959,
507
+ "learning_rate": 2.0727272727272728e-05,
508
+ "loss": 0.0003,
509
+ "step": 680
510
+ },
511
+ {
512
+ "epoch": 3.1363636363636362,
513
+ "grad_norm": 0.013830793090164661,
514
+ "learning_rate": 2.0590909090909093e-05,
515
+ "loss": 0.0005,
516
+ "step": 690
517
+ },
518
+ {
519
+ "epoch": 3.1818181818181817,
520
+ "grad_norm": 0.016489654779434204,
521
+ "learning_rate": 2.0454545454545454e-05,
522
+ "loss": 0.0014,
523
+ "step": 700
524
+ },
525
+ {
526
+ "epoch": 3.227272727272727,
527
+ "grad_norm": 0.011767825111746788,
528
+ "learning_rate": 2.031818181818182e-05,
529
+ "loss": 0.0004,
530
+ "step": 710
531
+ },
532
+ {
533
+ "epoch": 3.2727272727272725,
534
+ "grad_norm": 0.01013511698693037,
535
+ "learning_rate": 2.0181818181818183e-05,
536
+ "loss": 0.0003,
537
+ "step": 720
538
+ },
539
+ {
540
+ "epoch": 3.3181818181818183,
541
+ "grad_norm": 0.016494890674948692,
542
+ "learning_rate": 2.0045454545454548e-05,
543
+ "loss": 0.0941,
544
+ "step": 730
545
+ },
546
+ {
547
+ "epoch": 3.3636363636363638,
548
+ "grad_norm": 0.028399920091032982,
549
+ "learning_rate": 1.9909090909090913e-05,
550
+ "loss": 0.0003,
551
+ "step": 740
552
+ },
553
+ {
554
+ "epoch": 3.409090909090909,
555
+ "grad_norm": 0.006911724805831909,
556
+ "learning_rate": 1.9772727272727274e-05,
557
+ "loss": 0.0003,
558
+ "step": 750
559
+ },
560
+ {
561
+ "epoch": 3.4545454545454546,
562
+ "grad_norm": 0.009757892228662968,
563
+ "learning_rate": 1.963636363636364e-05,
564
+ "loss": 0.0003,
565
+ "step": 760
566
+ },
567
+ {
568
+ "epoch": 3.5,
569
+ "grad_norm": 0.014759697020053864,
570
+ "learning_rate": 1.95e-05,
571
+ "loss": 0.0702,
572
+ "step": 770
573
+ },
574
+ {
575
+ "epoch": 3.5454545454545454,
576
+ "grad_norm": 0.018820617347955704,
577
+ "learning_rate": 1.9363636363636364e-05,
578
+ "loss": 0.0023,
579
+ "step": 780
580
+ },
581
+ {
582
+ "epoch": 3.590909090909091,
583
+ "grad_norm": 0.008335668593645096,
584
+ "learning_rate": 1.9227272727272726e-05,
585
+ "loss": 0.099,
586
+ "step": 790
587
+ },
588
+ {
589
+ "epoch": 3.6363636363636362,
590
+ "grad_norm": 0.007782892789691687,
591
+ "learning_rate": 1.909090909090909e-05,
592
+ "loss": 0.0005,
593
+ "step": 800
594
+ },
595
+ {
596
+ "epoch": 3.6818181818181817,
597
+ "grad_norm": 0.007866962812840939,
598
+ "learning_rate": 1.8954545454545455e-05,
599
+ "loss": 0.0007,
600
+ "step": 810
601
+ },
602
+ {
603
+ "epoch": 3.7272727272727275,
604
+ "grad_norm": 0.00743032805621624,
605
+ "learning_rate": 1.881818181818182e-05,
606
+ "loss": 0.0003,
607
+ "step": 820
608
+ },
609
+ {
610
+ "epoch": 3.7727272727272725,
611
+ "grad_norm": 0.007409967016428709,
612
+ "learning_rate": 1.868181818181818e-05,
613
+ "loss": 0.0128,
614
+ "step": 830
615
+ },
616
+ {
617
+ "epoch": 3.8181818181818183,
618
+ "grad_norm": 0.008645043708384037,
619
+ "learning_rate": 1.8545454545454545e-05,
620
+ "loss": 0.0002,
621
+ "step": 840
622
+ },
623
+ {
624
+ "epoch": 3.8636363636363638,
625
+ "grad_norm": 0.006393834948539734,
626
+ "learning_rate": 1.840909090909091e-05,
627
+ "loss": 0.0002,
628
+ "step": 850
629
+ },
630
+ {
631
+ "epoch": 3.909090909090909,
632
+ "grad_norm": 0.008199839852750301,
633
+ "learning_rate": 1.8272727272727275e-05,
634
+ "loss": 0.0003,
635
+ "step": 860
636
+ },
637
+ {
638
+ "epoch": 3.9545454545454546,
639
+ "grad_norm": 0.007581517565995455,
640
+ "learning_rate": 1.8136363636363636e-05,
641
+ "loss": 0.0002,
642
+ "step": 870
643
+ },
644
+ {
645
+ "epoch": 4.0,
646
+ "grad_norm": 0.007007090840488672,
647
+ "learning_rate": 1.8e-05,
648
+ "loss": 0.0002,
649
+ "step": 880
650
+ },
651
+ {
652
+ "epoch": 4.0,
653
+ "eval_loss": 0.00015143574273679405,
654
+ "eval_runtime": 0.9926,
655
+ "eval_samples_per_second": 884.547,
656
+ "eval_steps_per_second": 55.41,
657
+ "step": 880
658
+ },
659
+ {
660
+ "epoch": 4.045454545454546,
661
+ "grad_norm": 81.15371704101562,
662
+ "learning_rate": 1.7863636363636365e-05,
663
+ "loss": 0.006,
664
+ "step": 890
665
+ },
666
+ {
667
+ "epoch": 4.090909090909091,
668
+ "grad_norm": 0.01149928942322731,
669
+ "learning_rate": 1.772727272727273e-05,
670
+ "loss": 0.0002,
671
+ "step": 900
672
+ },
673
+ {
674
+ "epoch": 4.136363636363637,
675
+ "grad_norm": 0.0050843264907598495,
676
+ "learning_rate": 1.759090909090909e-05,
677
+ "loss": 0.0002,
678
+ "step": 910
679
+ },
680
+ {
681
+ "epoch": 4.181818181818182,
682
+ "grad_norm": 0.005241791717708111,
683
+ "learning_rate": 1.7454545454545456e-05,
684
+ "loss": 0.0002,
685
+ "step": 920
686
+ },
687
+ {
688
+ "epoch": 4.2272727272727275,
689
+ "grad_norm": 0.004743785131722689,
690
+ "learning_rate": 1.731818181818182e-05,
691
+ "loss": 0.0002,
692
+ "step": 930
693
+ },
694
+ {
695
+ "epoch": 4.2727272727272725,
696
+ "grad_norm": 0.004982436075806618,
697
+ "learning_rate": 1.718181818181818e-05,
698
+ "loss": 0.0002,
699
+ "step": 940
700
+ },
701
+ {
702
+ "epoch": 4.318181818181818,
703
+ "grad_norm": 0.005252942908555269,
704
+ "learning_rate": 1.7045454545454546e-05,
705
+ "loss": 0.0002,
706
+ "step": 950
707
+ },
708
+ {
709
+ "epoch": 4.363636363636363,
710
+ "grad_norm": 0.00490264967083931,
711
+ "learning_rate": 1.6909090909090907e-05,
712
+ "loss": 0.0002,
713
+ "step": 960
714
+ },
715
+ {
716
+ "epoch": 4.409090909090909,
717
+ "grad_norm": 0.003635741537436843,
718
+ "learning_rate": 1.6772727272727272e-05,
719
+ "loss": 0.0001,
720
+ "step": 970
721
+ },
722
+ {
723
+ "epoch": 4.454545454545454,
724
+ "grad_norm": 0.005204927641898394,
725
+ "learning_rate": 1.6636363636363637e-05,
726
+ "loss": 0.0001,
727
+ "step": 980
728
+ },
729
+ {
730
+ "epoch": 4.5,
731
+ "grad_norm": 0.009996837005019188,
732
+ "learning_rate": 1.65e-05,
733
+ "loss": 0.0002,
734
+ "step": 990
735
+ },
736
+ {
737
+ "epoch": 4.545454545454545,
738
+ "grad_norm": 0.017751624807715416,
739
+ "learning_rate": 1.6363636363636363e-05,
740
+ "loss": 0.0315,
741
+ "step": 1000
742
+ },
743
+ {
744
+ "epoch": 4.590909090909091,
745
+ "grad_norm": 0.00822280440479517,
746
+ "learning_rate": 1.6227272727272727e-05,
747
+ "loss": 0.0001,
748
+ "step": 1010
749
+ },
750
+ {
751
+ "epoch": 4.636363636363637,
752
+ "grad_norm": 0.005263584200292826,
753
+ "learning_rate": 1.6090909090909092e-05,
754
+ "loss": 0.0001,
755
+ "step": 1020
756
+ },
757
+ {
758
+ "epoch": 4.681818181818182,
759
+ "grad_norm": 0.004647717345505953,
760
+ "learning_rate": 1.5954545454545456e-05,
761
+ "loss": 0.0002,
762
+ "step": 1030
763
+ },
764
+ {
765
+ "epoch": 4.7272727272727275,
766
+ "grad_norm": 0.0030076594557613134,
767
+ "learning_rate": 1.5818181818181818e-05,
768
+ "loss": 0.0001,
769
+ "step": 1040
770
+ },
771
+ {
772
+ "epoch": 4.7727272727272725,
773
+ "grad_norm": 0.0052589308470487595,
774
+ "learning_rate": 1.5681818181818182e-05,
775
+ "loss": 0.0001,
776
+ "step": 1050
777
+ },
778
+ {
779
+ "epoch": 4.818181818181818,
780
+ "grad_norm": 0.003165638307109475,
781
+ "learning_rate": 1.5545454545454547e-05,
782
+ "loss": 0.0001,
783
+ "step": 1060
784
+ },
785
+ {
786
+ "epoch": 4.863636363636363,
787
+ "grad_norm": 0.0037285718135535717,
788
+ "learning_rate": 1.540909090909091e-05,
789
+ "loss": 0.0001,
790
+ "step": 1070
791
+ },
792
+ {
793
+ "epoch": 4.909090909090909,
794
+ "grad_norm": 0.004402661230415106,
795
+ "learning_rate": 1.5272727272727273e-05,
796
+ "loss": 0.0001,
797
+ "step": 1080
798
+ },
799
+ {
800
+ "epoch": 4.954545454545455,
801
+ "grad_norm": 0.003948619589209557,
802
+ "learning_rate": 1.5136363636363636e-05,
803
+ "loss": 0.0001,
804
+ "step": 1090
805
+ },
806
+ {
807
+ "epoch": 5.0,
808
+ "grad_norm": 0.0029597911052405834,
809
+ "learning_rate": 1.5e-05,
810
+ "loss": 0.0001,
811
+ "step": 1100
812
+ },
813
+ {
814
+ "epoch": 5.0,
815
+ "eval_loss": 8.946753951022401e-05,
816
+ "eval_runtime": 0.9926,
817
+ "eval_samples_per_second": 884.51,
818
+ "eval_steps_per_second": 55.408,
819
+ "step": 1100
820
+ },
821
+ {
822
+ "epoch": 5.045454545454546,
823
+ "grad_norm": 0.00478377053514123,
824
+ "learning_rate": 1.4863636363636365e-05,
825
+ "loss": 0.0001,
826
+ "step": 1110
827
+ },
828
+ {
829
+ "epoch": 5.090909090909091,
830
+ "grad_norm": 0.0036518580745905638,
831
+ "learning_rate": 1.4727272727272728e-05,
832
+ "loss": 0.0001,
833
+ "step": 1120
834
+ },
835
+ {
836
+ "epoch": 5.136363636363637,
837
+ "grad_norm": 0.004087444860488176,
838
+ "learning_rate": 1.4590909090909093e-05,
839
+ "loss": 0.0001,
840
+ "step": 1130
841
+ },
842
+ {
843
+ "epoch": 5.181818181818182,
844
+ "grad_norm": 0.004169765394181013,
845
+ "learning_rate": 1.4454545454545456e-05,
846
+ "loss": 0.0001,
847
+ "step": 1140
848
+ },
849
+ {
850
+ "epoch": 5.2272727272727275,
851
+ "grad_norm": 0.00350973685272038,
852
+ "learning_rate": 1.4318181818181818e-05,
853
+ "loss": 0.0001,
854
+ "step": 1150
855
+ },
856
+ {
857
+ "epoch": 5.2727272727272725,
858
+ "grad_norm": 0.0037286856677383184,
859
+ "learning_rate": 1.4181818181818181e-05,
860
+ "loss": 0.0001,
861
+ "step": 1160
862
+ },
863
+ {
864
+ "epoch": 5.318181818181818,
865
+ "grad_norm": 0.003630703780800104,
866
+ "learning_rate": 1.4045454545454546e-05,
867
+ "loss": 0.0001,
868
+ "step": 1170
869
+ },
870
+ {
871
+ "epoch": 5.363636363636363,
872
+ "grad_norm": 0.0036072884686291218,
873
+ "learning_rate": 1.3909090909090909e-05,
874
+ "loss": 0.0001,
875
+ "step": 1180
876
+ },
877
+ {
878
+ "epoch": 5.409090909090909,
879
+ "grad_norm": 0.004187653306871653,
880
+ "learning_rate": 1.3772727272727274e-05,
881
+ "loss": 0.0001,
882
+ "step": 1190
883
+ },
884
+ {
885
+ "epoch": 5.454545454545454,
886
+ "grad_norm": 0.00530035886913538,
887
+ "learning_rate": 1.3636363636363637e-05,
888
+ "loss": 0.0001,
889
+ "step": 1200
890
+ },
891
+ {
892
+ "epoch": 5.5,
893
+ "grad_norm": 0.0027139252051711082,
894
+ "learning_rate": 1.3500000000000001e-05,
895
+ "loss": 0.0001,
896
+ "step": 1210
897
+ },
898
+ {
899
+ "epoch": 5.545454545454545,
900
+ "grad_norm": 0.0034846195485442877,
901
+ "learning_rate": 1.3363636363636364e-05,
902
+ "loss": 0.0001,
903
+ "step": 1220
904
+ },
905
+ {
906
+ "epoch": 5.590909090909091,
907
+ "grad_norm": 0.003408952383324504,
908
+ "learning_rate": 1.3227272727272727e-05,
909
+ "loss": 0.0001,
910
+ "step": 1230
911
+ },
912
+ {
913
+ "epoch": 5.636363636363637,
914
+ "grad_norm": 0.0027936245314776897,
915
+ "learning_rate": 1.309090909090909e-05,
916
+ "loss": 0.0001,
917
+ "step": 1240
918
+ },
919
+ {
920
+ "epoch": 5.681818181818182,
921
+ "grad_norm": 0.002841574139893055,
922
+ "learning_rate": 1.2954545454545455e-05,
923
+ "loss": 0.0001,
924
+ "step": 1250
925
+ },
926
+ {
927
+ "epoch": 5.7272727272727275,
928
+ "grad_norm": 0.26475799083709717,
929
+ "learning_rate": 1.2818181818181818e-05,
930
+ "loss": 0.0001,
931
+ "step": 1260
932
+ },
933
+ {
934
+ "epoch": 5.7727272727272725,
935
+ "grad_norm": 0.0034859974402934313,
936
+ "learning_rate": 1.2681818181818182e-05,
937
+ "loss": 0.0001,
938
+ "step": 1270
939
+ },
940
+ {
941
+ "epoch": 5.818181818181818,
942
+ "grad_norm": 0.0028127585537731647,
943
+ "learning_rate": 1.2545454545454545e-05,
944
+ "loss": 0.0001,
945
+ "step": 1280
946
+ },
947
+ {
948
+ "epoch": 5.863636363636363,
949
+ "grad_norm": 0.002384282648563385,
950
+ "learning_rate": 1.240909090909091e-05,
951
+ "loss": 0.0001,
952
+ "step": 1290
953
+ },
954
+ {
955
+ "epoch": 5.909090909090909,
956
+ "grad_norm": 0.0030948910862207413,
957
+ "learning_rate": 1.2272727272727274e-05,
958
+ "loss": 0.0001,
959
+ "step": 1300
960
+ },
961
+ {
962
+ "epoch": 5.954545454545455,
963
+ "grad_norm": 0.002442040015012026,
964
+ "learning_rate": 1.2136363636363637e-05,
965
+ "loss": 0.0001,
966
+ "step": 1310
967
+ },
968
+ {
969
+ "epoch": 6.0,
970
+ "grad_norm": 0.0018502968596294522,
971
+ "learning_rate": 1.2e-05,
972
+ "loss": 0.0001,
973
+ "step": 1320
974
+ },
975
+ {
976
+ "epoch": 6.0,
977
+ "eval_loss": 6.235863111214712e-05,
978
+ "eval_runtime": 0.9927,
979
+ "eval_samples_per_second": 884.48,
980
+ "eval_steps_per_second": 55.406,
981
+ "step": 1320
982
+ },
983
+ {
984
+ "epoch": 6.045454545454546,
985
+ "grad_norm": 0.002548688091337681,
986
+ "learning_rate": 1.1863636363636363e-05,
987
+ "loss": 0.0001,
988
+ "step": 1330
989
+ },
990
+ {
991
+ "epoch": 6.090909090909091,
992
+ "grad_norm": 0.002689023967832327,
993
+ "learning_rate": 1.1727272727272728e-05,
994
+ "loss": 0.0001,
995
+ "step": 1340
996
+ },
997
+ {
998
+ "epoch": 6.136363636363637,
999
+ "grad_norm": 0.002400546334683895,
1000
+ "learning_rate": 1.159090909090909e-05,
1001
+ "loss": 0.0001,
1002
+ "step": 1350
1003
+ },
1004
+ {
1005
+ "epoch": 6.181818181818182,
1006
+ "grad_norm": 0.0029753774870187044,
1007
+ "learning_rate": 1.1454545454545455e-05,
1008
+ "loss": 0.0001,
1009
+ "step": 1360
1010
+ },
1011
+ {
1012
+ "epoch": 6.2272727272727275,
1013
+ "grad_norm": 0.0020764051005244255,
1014
+ "learning_rate": 1.1318181818181818e-05,
1015
+ "loss": 0.0001,
1016
+ "step": 1370
1017
+ },
1018
+ {
1019
+ "epoch": 6.2727272727272725,
1020
+ "grad_norm": 0.00242880592122674,
1021
+ "learning_rate": 1.1181818181818183e-05,
1022
+ "loss": 0.0004,
1023
+ "step": 1380
1024
+ },
1025
+ {
1026
+ "epoch": 6.318181818181818,
1027
+ "grad_norm": 0.013365722261369228,
1028
+ "learning_rate": 1.1045454545454546e-05,
1029
+ "loss": 0.0001,
1030
+ "step": 1390
1031
+ },
1032
+ {
1033
+ "epoch": 6.363636363636363,
1034
+ "grad_norm": 0.0019247201271355152,
1035
+ "learning_rate": 1.0909090909090909e-05,
1036
+ "loss": 0.0001,
1037
+ "step": 1400
1038
+ },
1039
+ {
1040
+ "epoch": 6.409090909090909,
1041
+ "grad_norm": 0.0019971681758761406,
1042
+ "learning_rate": 1.0772727272727272e-05,
1043
+ "loss": 0.0001,
1044
+ "step": 1410
1045
+ },
1046
+ {
1047
+ "epoch": 6.454545454545454,
1048
+ "grad_norm": 0.002282345201820135,
1049
+ "learning_rate": 1.0636363636363636e-05,
1050
+ "loss": 0.0001,
1051
+ "step": 1420
1052
+ },
1053
+ {
1054
+ "epoch": 6.5,
1055
+ "grad_norm": 0.0025554117746651173,
1056
+ "learning_rate": 1.05e-05,
1057
+ "loss": 0.0001,
1058
+ "step": 1430
1059
+ },
1060
+ {
1061
+ "epoch": 6.545454545454545,
1062
+ "grad_norm": 0.003260772442445159,
1063
+ "learning_rate": 1.0363636363636364e-05,
1064
+ "loss": 0.0001,
1065
+ "step": 1440
1066
+ },
1067
+ {
1068
+ "epoch": 6.590909090909091,
1069
+ "grad_norm": 0.0019251167541369796,
1070
+ "learning_rate": 1.0227272727272727e-05,
1071
+ "loss": 0.0001,
1072
+ "step": 1450
1073
+ },
1074
+ {
1075
+ "epoch": 6.636363636363637,
1076
+ "grad_norm": 0.002389734610915184,
1077
+ "learning_rate": 1.0090909090909092e-05,
1078
+ "loss": 0.0001,
1079
+ "step": 1460
1080
+ },
1081
+ {
1082
+ "epoch": 6.681818181818182,
1083
+ "grad_norm": 0.00212781666778028,
1084
+ "learning_rate": 9.954545454545456e-06,
1085
+ "loss": 0.0001,
1086
+ "step": 1470
1087
+ },
1088
+ {
1089
+ "epoch": 6.7272727272727275,
1090
+ "grad_norm": 0.003619167488068342,
1091
+ "learning_rate": 9.81818181818182e-06,
1092
+ "loss": 0.0001,
1093
+ "step": 1480
1094
+ },
1095
+ {
1096
+ "epoch": 6.7727272727272725,
1097
+ "grad_norm": 0.002498344052582979,
1098
+ "learning_rate": 9.681818181818182e-06,
1099
+ "loss": 0.0001,
1100
+ "step": 1490
1101
+ },
1102
+ {
1103
+ "epoch": 6.818181818181818,
1104
+ "grad_norm": 0.0023231736849993467,
1105
+ "learning_rate": 9.545454545454545e-06,
1106
+ "loss": 0.0001,
1107
+ "step": 1500
1108
+ },
1109
+ {
1110
+ "epoch": 6.863636363636363,
1111
+ "grad_norm": 0.0021667128894478083,
1112
+ "learning_rate": 9.40909090909091e-06,
1113
+ "loss": 0.003,
1114
+ "step": 1510
1115
+ },
1116
+ {
1117
+ "epoch": 6.909090909090909,
1118
+ "grad_norm": 0.0022279066033661366,
1119
+ "learning_rate": 9.272727272727273e-06,
1120
+ "loss": 0.0001,
1121
+ "step": 1520
1122
+ },
1123
+ {
1124
+ "epoch": 6.954545454545455,
1125
+ "grad_norm": 0.00177993334364146,
1126
+ "learning_rate": 9.136363636363637e-06,
1127
+ "loss": 0.0001,
1128
+ "step": 1530
1129
+ },
1130
+ {
1131
+ "epoch": 7.0,
1132
+ "grad_norm": 0.0025669343303889036,
1133
+ "learning_rate": 9e-06,
1134
+ "loss": 0.0001,
1135
+ "step": 1540
1136
+ },
1137
+ {
1138
+ "epoch": 7.0,
1139
+ "eval_loss": 4.8654597776476294e-05,
1140
+ "eval_runtime": 0.981,
1141
+ "eval_samples_per_second": 895.028,
1142
+ "eval_steps_per_second": 56.067,
1143
+ "step": 1540
1144
+ },
1145
+ {
1146
+ "epoch": 7.045454545454546,
1147
+ "grad_norm": 0.0023670855443924665,
1148
+ "learning_rate": 8.863636363636365e-06,
1149
+ "loss": 0.0001,
1150
+ "step": 1550
1151
+ },
1152
+ {
1153
+ "epoch": 7.090909090909091,
1154
+ "grad_norm": 0.0023604007437825203,
1155
+ "learning_rate": 8.727272727272728e-06,
1156
+ "loss": 0.0001,
1157
+ "step": 1560
1158
+ },
1159
+ {
1160
+ "epoch": 7.136363636363637,
1161
+ "grad_norm": 0.0018116602441295981,
1162
+ "learning_rate": 8.59090909090909e-06,
1163
+ "loss": 0.0001,
1164
+ "step": 1570
1165
+ },
1166
+ {
1167
+ "epoch": 7.181818181818182,
1168
+ "grad_norm": 0.00249605649150908,
1169
+ "learning_rate": 8.454545454545454e-06,
1170
+ "loss": 0.0001,
1171
+ "step": 1580
1172
+ },
1173
+ {
1174
+ "epoch": 7.2272727272727275,
1175
+ "grad_norm": 0.001895196153782308,
1176
+ "learning_rate": 8.318181818181818e-06,
1177
+ "loss": 0.0001,
1178
+ "step": 1590
1179
+ },
1180
+ {
1181
+ "epoch": 7.2727272727272725,
1182
+ "grad_norm": 0.0017933849012479186,
1183
+ "learning_rate": 8.181818181818181e-06,
1184
+ "loss": 0.0429,
1185
+ "step": 1600
1186
+ },
1187
+ {
1188
+ "epoch": 7.318181818181818,
1189
+ "grad_norm": 0.0016959038330242038,
1190
+ "learning_rate": 8.045454545454546e-06,
1191
+ "loss": 0.0001,
1192
+ "step": 1610
1193
+ },
1194
+ {
1195
+ "epoch": 7.363636363636363,
1196
+ "grad_norm": 0.0016535187605768442,
1197
+ "learning_rate": 7.909090909090909e-06,
1198
+ "loss": 0.0001,
1199
+ "step": 1620
1200
+ },
1201
+ {
1202
+ "epoch": 7.409090909090909,
1203
+ "grad_norm": 0.0020366287790238857,
1204
+ "learning_rate": 7.772727272727273e-06,
1205
+ "loss": 0.0001,
1206
+ "step": 1630
1207
+ },
1208
+ {
1209
+ "epoch": 7.454545454545454,
1210
+ "grad_norm": 0.0017039361409842968,
1211
+ "learning_rate": 7.636363636363636e-06,
1212
+ "loss": 0.0001,
1213
+ "step": 1640
1214
+ },
1215
+ {
1216
+ "epoch": 7.5,
1217
+ "grad_norm": 0.0018439262639731169,
1218
+ "learning_rate": 7.5e-06,
1219
+ "loss": 0.0001,
1220
+ "step": 1650
1221
+ },
1222
+ {
1223
+ "epoch": 7.545454545454545,
1224
+ "grad_norm": 0.0017576288664713502,
1225
+ "learning_rate": 7.363636363636364e-06,
1226
+ "loss": 0.0001,
1227
+ "step": 1660
1228
+ },
1229
+ {
1230
+ "epoch": 7.590909090909091,
1231
+ "grad_norm": 0.0015827094903215766,
1232
+ "learning_rate": 7.227272727272728e-06,
1233
+ "loss": 0.0001,
1234
+ "step": 1670
1235
+ },
1236
+ {
1237
+ "epoch": 7.636363636363637,
1238
+ "grad_norm": 0.0021957652643322945,
1239
+ "learning_rate": 7.090909090909091e-06,
1240
+ "loss": 0.0001,
1241
+ "step": 1680
1242
+ },
1243
+ {
1244
+ "epoch": 7.681818181818182,
1245
+ "grad_norm": 0.0018625753000378609,
1246
+ "learning_rate": 6.9545454545454545e-06,
1247
+ "loss": 0.0001,
1248
+ "step": 1690
1249
+ },
1250
+ {
1251
+ "epoch": 7.7272727272727275,
1252
+ "grad_norm": 0.003254745388403535,
1253
+ "learning_rate": 6.818181818181818e-06,
1254
+ "loss": 0.0001,
1255
+ "step": 1700
1256
+ },
1257
+ {
1258
+ "epoch": 7.7727272727272725,
1259
+ "grad_norm": 0.0016077288892120123,
1260
+ "learning_rate": 6.681818181818182e-06,
1261
+ "loss": 0.0001,
1262
+ "step": 1710
1263
+ },
1264
+ {
1265
+ "epoch": 7.818181818181818,
1266
+ "grad_norm": 0.0023993789218366146,
1267
+ "learning_rate": 6.545454545454545e-06,
1268
+ "loss": 0.0001,
1269
+ "step": 1720
1270
+ },
1271
+ {
1272
+ "epoch": 7.863636363636363,
1273
+ "grad_norm": 0.001893221982754767,
1274
+ "learning_rate": 6.409090909090909e-06,
1275
+ "loss": 0.0001,
1276
+ "step": 1730
1277
+ },
1278
+ {
1279
+ "epoch": 7.909090909090909,
1280
+ "grad_norm": 0.0018822376150637865,
1281
+ "learning_rate": 6.272727272727273e-06,
1282
+ "loss": 0.0001,
1283
+ "step": 1740
1284
+ },
1285
+ {
1286
+ "epoch": 7.954545454545455,
1287
+ "grad_norm": 0.009597906842827797,
1288
+ "learning_rate": 6.136363636363637e-06,
1289
+ "loss": 0.0001,
1290
+ "step": 1750
1291
+ },
1292
+ {
1293
+ "epoch": 8.0,
1294
+ "grad_norm": 0.0014578086556866765,
1295
+ "learning_rate": 6e-06,
1296
+ "loss": 0.0001,
1297
+ "step": 1760
1298
+ },
1299
+ {
1300
+ "epoch": 8.0,
1301
+ "eval_loss": 4.155210262979381e-05,
1302
+ "eval_runtime": 1.017,
1303
+ "eval_samples_per_second": 863.319,
1304
+ "eval_steps_per_second": 54.08,
1305
+ "step": 1760
1306
+ },
1307
+ {
1308
+ "epoch": 8.045454545454545,
1309
+ "grad_norm": 0.001562977209687233,
1310
+ "learning_rate": 5.863636363636364e-06,
1311
+ "loss": 0.0001,
1312
+ "step": 1770
1313
+ },
1314
+ {
1315
+ "epoch": 8.090909090909092,
1316
+ "grad_norm": 0.0019722983706742525,
1317
+ "learning_rate": 5.727272727272728e-06,
1318
+ "loss": 0.0001,
1319
+ "step": 1780
1320
+ },
1321
+ {
1322
+ "epoch": 8.136363636363637,
1323
+ "grad_norm": 0.0016468315152451396,
1324
+ "learning_rate": 5.5909090909090915e-06,
1325
+ "loss": 0.0001,
1326
+ "step": 1790
1327
+ },
1328
+ {
1329
+ "epoch": 8.181818181818182,
1330
+ "grad_norm": 0.0018590294057503343,
1331
+ "learning_rate": 5.4545454545454545e-06,
1332
+ "loss": 0.0001,
1333
+ "step": 1800
1334
+ },
1335
+ {
1336
+ "epoch": 8.227272727272727,
1337
+ "grad_norm": 0.002041436033323407,
1338
+ "learning_rate": 5.318181818181818e-06,
1339
+ "loss": 0.0001,
1340
+ "step": 1810
1341
+ },
1342
+ {
1343
+ "epoch": 8.272727272727273,
1344
+ "grad_norm": 0.0021510140504688025,
1345
+ "learning_rate": 5.181818181818182e-06,
1346
+ "loss": 0.0001,
1347
+ "step": 1820
1348
+ },
1349
+ {
1350
+ "epoch": 8.318181818181818,
1351
+ "grad_norm": 0.0015175098087638617,
1352
+ "learning_rate": 5.045454545454546e-06,
1353
+ "loss": 0.0001,
1354
+ "step": 1830
1355
+ },
1356
+ {
1357
+ "epoch": 8.363636363636363,
1358
+ "grad_norm": 0.001754813943989575,
1359
+ "learning_rate": 4.90909090909091e-06,
1360
+ "loss": 0.0001,
1361
+ "step": 1840
1362
+ },
1363
+ {
1364
+ "epoch": 8.409090909090908,
1365
+ "grad_norm": 0.001608902239240706,
1366
+ "learning_rate": 4.7727272727272725e-06,
1367
+ "loss": 0.0001,
1368
+ "step": 1850
1369
+ },
1370
+ {
1371
+ "epoch": 8.454545454545455,
1372
+ "grad_norm": 0.002168968552723527,
1373
+ "learning_rate": 4.636363636363636e-06,
1374
+ "loss": 0.0001,
1375
+ "step": 1860
1376
+ },
1377
+ {
1378
+ "epoch": 8.5,
1379
+ "grad_norm": 0.001456632511690259,
1380
+ "learning_rate": 4.5e-06,
1381
+ "loss": 0.0,
1382
+ "step": 1870
1383
+ },
1384
+ {
1385
+ "epoch": 8.545454545454545,
1386
+ "grad_norm": 0.0017024242551997304,
1387
+ "learning_rate": 4.363636363636364e-06,
1388
+ "loss": 0.0001,
1389
+ "step": 1880
1390
+ },
1391
+ {
1392
+ "epoch": 8.590909090909092,
1393
+ "grad_norm": 0.00176154519431293,
1394
+ "learning_rate": 4.227272727272727e-06,
1395
+ "loss": 0.0001,
1396
+ "step": 1890
1397
+ },
1398
+ {
1399
+ "epoch": 8.636363636363637,
1400
+ "grad_norm": 0.0019339303253218532,
1401
+ "learning_rate": 4.090909090909091e-06,
1402
+ "loss": 0.0001,
1403
+ "step": 1900
1404
+ },
1405
+ {
1406
+ "epoch": 8.681818181818182,
1407
+ "grad_norm": 0.0019142951350659132,
1408
+ "learning_rate": 3.954545454545454e-06,
1409
+ "loss": 0.0001,
1410
+ "step": 1910
1411
+ },
1412
+ {
1413
+ "epoch": 8.727272727272727,
1414
+ "grad_norm": 0.0015304730040952563,
1415
+ "learning_rate": 3.818181818181818e-06,
1416
+ "loss": 0.0001,
1417
+ "step": 1920
1418
+ },
1419
+ {
1420
+ "epoch": 8.772727272727273,
1421
+ "grad_norm": 0.0021663594525307417,
1422
+ "learning_rate": 3.681818181818182e-06,
1423
+ "loss": 0.0001,
1424
+ "step": 1930
1425
+ },
1426
+ {
1427
+ "epoch": 8.818181818181818,
1428
+ "grad_norm": 0.001586704864166677,
1429
+ "learning_rate": 3.5454545454545454e-06,
1430
+ "loss": 0.0001,
1431
+ "step": 1940
1432
+ },
1433
+ {
1434
+ "epoch": 8.863636363636363,
1435
+ "grad_norm": 0.001986338524147868,
1436
+ "learning_rate": 3.409090909090909e-06,
1437
+ "loss": 0.0,
1438
+ "step": 1950
1439
+ },
1440
+ {
1441
+ "epoch": 8.909090909090908,
1442
+ "grad_norm": 0.001552366535179317,
1443
+ "learning_rate": 3.2727272727272725e-06,
1444
+ "loss": 0.0033,
1445
+ "step": 1960
1446
+ },
1447
+ {
1448
+ "epoch": 8.954545454545455,
1449
+ "grad_norm": 0.0018597301095724106,
1450
+ "learning_rate": 3.1363636363636363e-06,
1451
+ "loss": 0.0001,
1452
+ "step": 1970
1453
+ },
1454
+ {
1455
+ "epoch": 9.0,
1456
+ "grad_norm": 0.002290609758347273,
1457
+ "learning_rate": 3e-06,
1458
+ "loss": 0.0,
1459
+ "step": 1980
1460
+ },
1461
+ {
1462
+ "epoch": 9.0,
1463
+ "eval_loss": 3.7486017390619963e-05,
1464
+ "eval_runtime": 1.0378,
1465
+ "eval_samples_per_second": 846.047,
1466
+ "eval_steps_per_second": 52.998,
1467
+ "step": 1980
1468
+ },
1469
+ {
1470
+ "epoch": 9.045454545454545,
1471
+ "grad_norm": 0.0014905119314789772,
1472
+ "learning_rate": 2.863636363636364e-06,
1473
+ "loss": 0.0,
1474
+ "step": 1990
1475
+ },
1476
+ {
1477
+ "epoch": 9.090909090909092,
1478
+ "grad_norm": 0.0013228630414232612,
1479
+ "learning_rate": 2.7272727272727272e-06,
1480
+ "loss": 0.0001,
1481
+ "step": 2000
1482
+ },
1483
+ {
1484
+ "epoch": 9.136363636363637,
1485
+ "grad_norm": 0.00189464061986655,
1486
+ "learning_rate": 2.590909090909091e-06,
1487
+ "loss": 0.0,
1488
+ "step": 2010
1489
+ },
1490
+ {
1491
+ "epoch": 9.181818181818182,
1492
+ "grad_norm": 0.0020432292949408293,
1493
+ "learning_rate": 2.454545454545455e-06,
1494
+ "loss": 0.0001,
1495
+ "step": 2020
1496
+ },
1497
+ {
1498
+ "epoch": 9.227272727272727,
1499
+ "grad_norm": 0.0014936975203454494,
1500
+ "learning_rate": 2.318181818181818e-06,
1501
+ "loss": 0.0001,
1502
+ "step": 2030
1503
+ },
1504
+ {
1505
+ "epoch": 9.272727272727273,
1506
+ "grad_norm": 0.0020053344778716564,
1507
+ "learning_rate": 2.181818181818182e-06,
1508
+ "loss": 0.0,
1509
+ "step": 2040
1510
+ },
1511
+ {
1512
+ "epoch": 9.318181818181818,
1513
+ "grad_norm": 0.0015693982131779194,
1514
+ "learning_rate": 2.0454545454545453e-06,
1515
+ "loss": 0.0,
1516
+ "step": 2050
1517
+ },
1518
+ {
1519
+ "epoch": 9.363636363636363,
1520
+ "grad_norm": 0.0016571198357269168,
1521
+ "learning_rate": 1.909090909090909e-06,
1522
+ "loss": 0.0001,
1523
+ "step": 2060
1524
+ },
1525
+ {
1526
+ "epoch": 9.409090909090908,
1527
+ "grad_norm": 0.0013354700058698654,
1528
+ "learning_rate": 1.7727272727272727e-06,
1529
+ "loss": 0.0,
1530
+ "step": 2070
1531
+ },
1532
+ {
1533
+ "epoch": 9.454545454545455,
1534
+ "grad_norm": 0.0013917312026023865,
1535
+ "learning_rate": 1.6363636363636363e-06,
1536
+ "loss": 0.0,
1537
+ "step": 2080
1538
+ },
1539
+ {
1540
+ "epoch": 9.5,
1541
+ "grad_norm": 0.0014988429611548781,
1542
+ "learning_rate": 1.5e-06,
1543
+ "loss": 0.0,
1544
+ "step": 2090
1545
+ },
1546
+ {
1547
+ "epoch": 9.545454545454545,
1548
+ "grad_norm": 0.0014679876621812582,
1549
+ "learning_rate": 1.3636363636363636e-06,
1550
+ "loss": 0.0,
1551
+ "step": 2100
1552
+ },
1553
+ {
1554
+ "epoch": 9.590909090909092,
1555
+ "grad_norm": 0.0018639364279806614,
1556
+ "learning_rate": 1.2272727272727274e-06,
1557
+ "loss": 0.0001,
1558
+ "step": 2110
1559
+ },
1560
+ {
1561
+ "epoch": 9.636363636363637,
1562
+ "grad_norm": 0.001416134531609714,
1563
+ "learning_rate": 1.090909090909091e-06,
1564
+ "loss": 0.0,
1565
+ "step": 2120
1566
+ },
1567
+ {
1568
+ "epoch": 9.681818181818182,
1569
+ "grad_norm": 0.0018406022572889924,
1570
+ "learning_rate": 9.545454545454546e-07,
1571
+ "loss": 0.0001,
1572
+ "step": 2130
1573
+ },
1574
+ {
1575
+ "epoch": 9.727272727272727,
1576
+ "grad_norm": 0.0014891604660078883,
1577
+ "learning_rate": 8.181818181818181e-07,
1578
+ "loss": 0.0001,
1579
+ "step": 2140
1580
+ },
1581
+ {
1582
+ "epoch": 9.772727272727273,
1583
+ "grad_norm": 0.0017427564598619938,
1584
+ "learning_rate": 6.818181818181818e-07,
1585
+ "loss": 0.0,
1586
+ "step": 2150
1587
+ },
1588
+ {
1589
+ "epoch": 9.818181818181818,
1590
+ "grad_norm": 0.0014797528274357319,
1591
+ "learning_rate": 5.454545454545455e-07,
1592
+ "loss": 0.0,
1593
+ "step": 2160
1594
+ },
1595
+ {
1596
+ "epoch": 9.863636363636363,
1597
+ "grad_norm": 0.0015343551058322191,
1598
+ "learning_rate": 4.0909090909090906e-07,
1599
+ "loss": 0.0001,
1600
+ "step": 2170
1601
+ },
1602
+ {
1603
+ "epoch": 9.909090909090908,
1604
+ "grad_norm": 0.0014587711775675416,
1605
+ "learning_rate": 2.7272727272727274e-07,
1606
+ "loss": 0.0,
1607
+ "step": 2180
1608
+ },
1609
+ {
1610
+ "epoch": 9.954545454545455,
1611
+ "grad_norm": 0.0013577837962657213,
1612
+ "learning_rate": 1.3636363636363637e-07,
1613
+ "loss": 0.0353,
1614
+ "step": 2190
1615
+ },
1616
+ {
1617
+ "epoch": 10.0,
1618
+ "grad_norm": 0.0017256715800613165,
1619
+ "learning_rate": 0.0,
1620
+ "loss": 0.0001,
1621
+ "step": 2200
1622
+ },
1623
+ {
1624
+ "epoch": 10.0,
1625
+ "eval_loss": 3.619581912062131e-05,
1626
+ "eval_runtime": 1.0359,
1627
+ "eval_samples_per_second": 847.586,
1628
+ "eval_steps_per_second": 53.095,
1629
+ "step": 2200
1630
+ }
1631
+ ],
1632
+ "logging_steps": 10,
1633
+ "max_steps": 2200,
1634
+ "num_input_tokens_seen": 0,
1635
+ "num_train_epochs": 10,
1636
+ "save_steps": 500,
1637
+ "stateful_callbacks": {
1638
+ "TrainerControl": {
1639
+ "args": {
1640
+ "should_epoch_stop": false,
1641
+ "should_evaluate": false,
1642
+ "should_log": false,
1643
+ "should_save": true,
1644
+ "should_training_stop": true
1645
+ },
1646
+ "attributes": {}
1647
+ }
1648
+ },
1649
+ "total_flos": 234554255855400.0,
1650
+ "train_batch_size": 16,
1651
+ "trial_name": null,
1652
+ "trial_params": null
1653
+ }
checkpoint-2200/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:daffa8511fa7248b90c5171ae9aa4650082ed9d1a452d2503ddb3bfc15e73d3b
3
+ size 5240
checkpoint-2200/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "bert-base-uncased",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "gradient_checkpointing": false,
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-12,
15
+ "max_position_embeddings": 512,
16
+ "model_type": "bert",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "pad_token_id": 0,
20
+ "position_embedding_type": "absolute",
21
+ "problem_type": "single_label_classification",
22
+ "torch_dtype": "float32",
23
+ "transformers_version": "4.48.1",
24
+ "type_vocab_size": 2,
25
+ "use_cache": true,
26
+ "vocab_size": 30522
27
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9cb61e7eedeb7a38d96c4018e43f99e8716537f87b0b791b0c2a3061fe8ed07
3
+ size 437958648
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "pad_token": "[PAD]",
51
+ "sep_token": "[SEP]",
52
+ "strip_accents": null,
53
+ "tokenize_chinese_chars": true,
54
+ "tokenizer_class": "BertTokenizer",
55
+ "unk_token": "[UNK]"
56
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
vocab.txt ADDED
The diff for this file is too large to render. See raw diff