adamadam111 commited on
Commit
16145b8
·
verified ·
1 Parent(s): 229b5aa

Upload folder using huggingface_hub

Browse files
checkpoint-4500/config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BrosForDocumentClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bbox_scale": 100.0,
7
+ "classifier_dropout_prob": 0.1,
8
+ "dim_bbox": 8,
9
+ "dim_bbox_projection": 64,
10
+ "dim_bbox_sinusoid_emb_1d": 24,
11
+ "dim_bbox_sinusoid_emb_2d": 192,
12
+ "hidden_act": "gelu",
13
+ "hidden_dropout_prob": 0.1,
14
+ "hidden_size": 768,
15
+ "id2label": {
16
+ "0": "form",
17
+ "1": "invoice",
18
+ "2": "budget",
19
+ "3": "file folder",
20
+ "4": "questionnaire"
21
+ },
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 3072,
24
+ "label2id": {
25
+ "budget": 2,
26
+ "file folder": 3,
27
+ "form": 0,
28
+ "invoice": 1,
29
+ "questionnaire": 4
30
+ },
31
+ "layer_norm_eps": 1e-12,
32
+ "max_position_embeddings": 512,
33
+ "model_type": "bros",
34
+ "n_relations": 1,
35
+ "num_attention_heads": 12,
36
+ "num_hidden_layers": 12,
37
+ "pad_token_id": 0,
38
+ "pe_type": "crel",
39
+ "torch_dtype": "float32",
40
+ "transformers_version": "4.53.0",
41
+ "type_vocab_size": 2,
42
+ "vocab_size": 30522
43
+ }
checkpoint-4500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c734282224483402874fe0d8b9670e9d06e66a03724d1e68498e9b7eca98951a
3
+ size 438021652
checkpoint-4500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9331b0d8e134e282ce5c64d7df4cbe52f7237165270812e5b8d742b7db105a5
3
+ size 871429498
checkpoint-4500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f81594be2e47b64f681f892d4e630e3eb0a96022bd503a0e9ed224a3d9fe2bd
3
+ size 14244
checkpoint-4500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:803ff710c7cc7ca3c1db1a96a251f3adc9a7434ab9c0d55acd8afc98758c4b4f
3
+ size 1064
checkpoint-4500/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-4500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-4500/tokenizer_config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "mask_token": "[MASK]",
50
+ "model_max_length": 1000000000000000019884624838656,
51
+ "never_split": null,
52
+ "pad_token": "[PAD]",
53
+ "sep_token": "[SEP]",
54
+ "strip_accents": null,
55
+ "tokenize_chinese_chars": true,
56
+ "tokenizer_class": "BertTokenizer",
57
+ "unk_token": "[UNK]"
58
+ }
checkpoint-4500/trainer_state.json ADDED
@@ -0,0 +1,781 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 4500,
3
+ "best_metric": 0.852,
4
+ "best_model_checkpoint": "./bros-docclass-finetuned/checkpoint-4500",
5
+ "epoch": 9.0,
6
+ "eval_steps": 500,
7
+ "global_step": 4500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.1,
14
+ "grad_norm": 9.038439750671387,
15
+ "learning_rate": 1.9804000000000002e-05,
16
+ "loss": 1.2156,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.2,
21
+ "grad_norm": 12.505393028259277,
22
+ "learning_rate": 1.9604e-05,
23
+ "loss": 0.8366,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.3,
28
+ "grad_norm": 17.0165958404541,
29
+ "learning_rate": 1.9404e-05,
30
+ "loss": 0.7993,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.4,
35
+ "grad_norm": 25.100793838500977,
36
+ "learning_rate": 1.9204e-05,
37
+ "loss": 0.6683,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 0.5,
42
+ "grad_norm": 12.89208698272705,
43
+ "learning_rate": 1.9004000000000003e-05,
44
+ "loss": 0.7392,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 0.6,
49
+ "grad_norm": 19.72749137878418,
50
+ "learning_rate": 1.8804e-05,
51
+ "loss": 0.666,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.7,
56
+ "grad_norm": 30.814998626708984,
57
+ "learning_rate": 1.8604000000000003e-05,
58
+ "loss": 0.6453,
59
+ "step": 350
60
+ },
61
+ {
62
+ "epoch": 0.8,
63
+ "grad_norm": 1.3695790767669678,
64
+ "learning_rate": 1.8404000000000002e-05,
65
+ "loss": 0.6402,
66
+ "step": 400
67
+ },
68
+ {
69
+ "epoch": 0.9,
70
+ "grad_norm": 3.071685791015625,
71
+ "learning_rate": 1.8204e-05,
72
+ "loss": 0.6646,
73
+ "step": 450
74
+ },
75
+ {
76
+ "epoch": 1.0,
77
+ "grad_norm": 19.854291915893555,
78
+ "learning_rate": 1.8004000000000002e-05,
79
+ "loss": 0.6606,
80
+ "step": 500
81
+ },
82
+ {
83
+ "epoch": 1.0,
84
+ "eval_accuracy": 0.798,
85
+ "eval_f1": 0.7962131024443857,
86
+ "eval_loss": 0.6697192788124084,
87
+ "eval_precision": 0.8016786327842085,
88
+ "eval_recall": 0.798,
89
+ "eval_runtime": 16.5436,
90
+ "eval_samples_per_second": 60.446,
91
+ "eval_steps_per_second": 7.556,
92
+ "step": 500
93
+ },
94
+ {
95
+ "epoch": 1.1,
96
+ "grad_norm": 3.369779109954834,
97
+ "learning_rate": 1.7804e-05,
98
+ "loss": 0.5355,
99
+ "step": 550
100
+ },
101
+ {
102
+ "epoch": 1.2,
103
+ "grad_norm": 20.186811447143555,
104
+ "learning_rate": 1.7604e-05,
105
+ "loss": 0.5781,
106
+ "step": 600
107
+ },
108
+ {
109
+ "epoch": 1.3,
110
+ "grad_norm": 19.78652000427246,
111
+ "learning_rate": 1.7404e-05,
112
+ "loss": 0.5731,
113
+ "step": 650
114
+ },
115
+ {
116
+ "epoch": 1.4,
117
+ "grad_norm": 0.9622361063957214,
118
+ "learning_rate": 1.7204e-05,
119
+ "loss": 0.5907,
120
+ "step": 700
121
+ },
122
+ {
123
+ "epoch": 1.5,
124
+ "grad_norm": 22.184574127197266,
125
+ "learning_rate": 1.7004000000000002e-05,
126
+ "loss": 0.4993,
127
+ "step": 750
128
+ },
129
+ {
130
+ "epoch": 1.6,
131
+ "grad_norm": 37.49524688720703,
132
+ "learning_rate": 1.6804e-05,
133
+ "loss": 0.6767,
134
+ "step": 800
135
+ },
136
+ {
137
+ "epoch": 1.7,
138
+ "grad_norm": 0.731647789478302,
139
+ "learning_rate": 1.6604000000000002e-05,
140
+ "loss": 0.6219,
141
+ "step": 850
142
+ },
143
+ {
144
+ "epoch": 1.8,
145
+ "grad_norm": 5.991142272949219,
146
+ "learning_rate": 1.6404e-05,
147
+ "loss": 0.5425,
148
+ "step": 900
149
+ },
150
+ {
151
+ "epoch": 1.9,
152
+ "grad_norm": 35.003902435302734,
153
+ "learning_rate": 1.6204000000000003e-05,
154
+ "loss": 0.5822,
155
+ "step": 950
156
+ },
157
+ {
158
+ "epoch": 2.0,
159
+ "grad_norm": 7.839473247528076,
160
+ "learning_rate": 1.6004e-05,
161
+ "loss": 0.5845,
162
+ "step": 1000
163
+ },
164
+ {
165
+ "epoch": 2.0,
166
+ "eval_accuracy": 0.828,
167
+ "eval_f1": 0.8261256801910406,
168
+ "eval_loss": 0.6260756254196167,
169
+ "eval_precision": 0.8353381322045145,
170
+ "eval_recall": 0.828,
171
+ "eval_runtime": 16.5173,
172
+ "eval_samples_per_second": 60.543,
173
+ "eval_steps_per_second": 7.568,
174
+ "step": 1000
175
+ },
176
+ {
177
+ "epoch": 2.1,
178
+ "grad_norm": 22.77824592590332,
179
+ "learning_rate": 1.5804000000000003e-05,
180
+ "loss": 0.526,
181
+ "step": 1050
182
+ },
183
+ {
184
+ "epoch": 2.2,
185
+ "grad_norm": 28.53985023498535,
186
+ "learning_rate": 1.5604000000000002e-05,
187
+ "loss": 0.3672,
188
+ "step": 1100
189
+ },
190
+ {
191
+ "epoch": 2.3,
192
+ "grad_norm": 13.204597473144531,
193
+ "learning_rate": 1.5404e-05,
194
+ "loss": 0.521,
195
+ "step": 1150
196
+ },
197
+ {
198
+ "epoch": 2.4,
199
+ "grad_norm": 33.86636734008789,
200
+ "learning_rate": 1.5204e-05,
201
+ "loss": 0.4376,
202
+ "step": 1200
203
+ },
204
+ {
205
+ "epoch": 2.5,
206
+ "grad_norm": 3.819549798965454,
207
+ "learning_rate": 1.5004e-05,
208
+ "loss": 0.5127,
209
+ "step": 1250
210
+ },
211
+ {
212
+ "epoch": 2.6,
213
+ "grad_norm": 9.592472076416016,
214
+ "learning_rate": 1.4804000000000001e-05,
215
+ "loss": 0.4313,
216
+ "step": 1300
217
+ },
218
+ {
219
+ "epoch": 2.7,
220
+ "grad_norm": 2.242670774459839,
221
+ "learning_rate": 1.4604000000000001e-05,
222
+ "loss": 0.4445,
223
+ "step": 1350
224
+ },
225
+ {
226
+ "epoch": 2.8,
227
+ "grad_norm": 61.427616119384766,
228
+ "learning_rate": 1.4404e-05,
229
+ "loss": 0.4974,
230
+ "step": 1400
231
+ },
232
+ {
233
+ "epoch": 2.9,
234
+ "grad_norm": 0.4456544816493988,
235
+ "learning_rate": 1.4204000000000002e-05,
236
+ "loss": 0.4999,
237
+ "step": 1450
238
+ },
239
+ {
240
+ "epoch": 3.0,
241
+ "grad_norm": 33.89624786376953,
242
+ "learning_rate": 1.4004000000000002e-05,
243
+ "loss": 0.4223,
244
+ "step": 1500
245
+ },
246
+ {
247
+ "epoch": 3.0,
248
+ "eval_accuracy": 0.818,
249
+ "eval_f1": 0.8172688666919534,
250
+ "eval_loss": 0.7300030589103699,
251
+ "eval_precision": 0.8328778365646615,
252
+ "eval_recall": 0.818,
253
+ "eval_runtime": 16.4984,
254
+ "eval_samples_per_second": 60.612,
255
+ "eval_steps_per_second": 7.576,
256
+ "step": 1500
257
+ },
258
+ {
259
+ "epoch": 3.1,
260
+ "grad_norm": 18.57257652282715,
261
+ "learning_rate": 1.3804000000000002e-05,
262
+ "loss": 0.4925,
263
+ "step": 1550
264
+ },
265
+ {
266
+ "epoch": 3.2,
267
+ "grad_norm": 40.097713470458984,
268
+ "learning_rate": 1.3604000000000002e-05,
269
+ "loss": 0.2834,
270
+ "step": 1600
271
+ },
272
+ {
273
+ "epoch": 3.3,
274
+ "grad_norm": 191.4404754638672,
275
+ "learning_rate": 1.3404e-05,
276
+ "loss": 0.4128,
277
+ "step": 1650
278
+ },
279
+ {
280
+ "epoch": 3.4,
281
+ "grad_norm": 67.55642700195312,
282
+ "learning_rate": 1.3204000000000001e-05,
283
+ "loss": 0.3483,
284
+ "step": 1700
285
+ },
286
+ {
287
+ "epoch": 3.5,
288
+ "grad_norm": 26.829570770263672,
289
+ "learning_rate": 1.3004000000000001e-05,
290
+ "loss": 0.3372,
291
+ "step": 1750
292
+ },
293
+ {
294
+ "epoch": 3.6,
295
+ "grad_norm": 12.817995071411133,
296
+ "learning_rate": 1.2804000000000001e-05,
297
+ "loss": 0.4087,
298
+ "step": 1800
299
+ },
300
+ {
301
+ "epoch": 3.7,
302
+ "grad_norm": 1.6152303218841553,
303
+ "learning_rate": 1.2604e-05,
304
+ "loss": 0.4033,
305
+ "step": 1850
306
+ },
307
+ {
308
+ "epoch": 3.8,
309
+ "grad_norm": 0.07653416693210602,
310
+ "learning_rate": 1.2404e-05,
311
+ "loss": 0.3495,
312
+ "step": 1900
313
+ },
314
+ {
315
+ "epoch": 3.9,
316
+ "grad_norm": 2.4822909832000732,
317
+ "learning_rate": 1.2204e-05,
318
+ "loss": 0.4195,
319
+ "step": 1950
320
+ },
321
+ {
322
+ "epoch": 4.0,
323
+ "grad_norm": 91.26575469970703,
324
+ "learning_rate": 1.2004e-05,
325
+ "loss": 0.3644,
326
+ "step": 2000
327
+ },
328
+ {
329
+ "epoch": 4.0,
330
+ "eval_accuracy": 0.839,
331
+ "eval_f1": 0.8384490586281527,
332
+ "eval_loss": 0.7187097668647766,
333
+ "eval_precision": 0.8450004509095641,
334
+ "eval_recall": 0.839,
335
+ "eval_runtime": 16.8707,
336
+ "eval_samples_per_second": 59.274,
337
+ "eval_steps_per_second": 7.409,
338
+ "step": 2000
339
+ },
340
+ {
341
+ "epoch": 4.1,
342
+ "grad_norm": 4.009641170501709,
343
+ "learning_rate": 1.1803999999999999e-05,
344
+ "loss": 0.3486,
345
+ "step": 2050
346
+ },
347
+ {
348
+ "epoch": 4.2,
349
+ "grad_norm": 16.93126106262207,
350
+ "learning_rate": 1.1604000000000003e-05,
351
+ "loss": 0.3628,
352
+ "step": 2100
353
+ },
354
+ {
355
+ "epoch": 4.3,
356
+ "grad_norm": 65.41966247558594,
357
+ "learning_rate": 1.1404000000000001e-05,
358
+ "loss": 0.2832,
359
+ "step": 2150
360
+ },
361
+ {
362
+ "epoch": 4.4,
363
+ "grad_norm": 36.100013732910156,
364
+ "learning_rate": 1.1204000000000001e-05,
365
+ "loss": 0.3515,
366
+ "step": 2200
367
+ },
368
+ {
369
+ "epoch": 4.5,
370
+ "grad_norm": 13.878293991088867,
371
+ "learning_rate": 1.1004000000000002e-05,
372
+ "loss": 0.282,
373
+ "step": 2250
374
+ },
375
+ {
376
+ "epoch": 4.6,
377
+ "grad_norm": 82.44599914550781,
378
+ "learning_rate": 1.0804000000000002e-05,
379
+ "loss": 0.2804,
380
+ "step": 2300
381
+ },
382
+ {
383
+ "epoch": 4.7,
384
+ "grad_norm": 3.182461977005005,
385
+ "learning_rate": 1.0604e-05,
386
+ "loss": 0.2802,
387
+ "step": 2350
388
+ },
389
+ {
390
+ "epoch": 4.8,
391
+ "grad_norm": 0.41410791873931885,
392
+ "learning_rate": 1.0404e-05,
393
+ "loss": 0.309,
394
+ "step": 2400
395
+ },
396
+ {
397
+ "epoch": 4.9,
398
+ "grad_norm": 37.641319274902344,
399
+ "learning_rate": 1.0204000000000001e-05,
400
+ "loss": 0.284,
401
+ "step": 2450
402
+ },
403
+ {
404
+ "epoch": 5.0,
405
+ "grad_norm": 3.98016619682312,
406
+ "learning_rate": 1.0004000000000001e-05,
407
+ "loss": 0.3447,
408
+ "step": 2500
409
+ },
410
+ {
411
+ "epoch": 5.0,
412
+ "eval_accuracy": 0.844,
413
+ "eval_f1": 0.844281271446929,
414
+ "eval_loss": 0.7856075167655945,
415
+ "eval_precision": 0.8486261672220536,
416
+ "eval_recall": 0.844,
417
+ "eval_runtime": 16.8405,
418
+ "eval_samples_per_second": 59.381,
419
+ "eval_steps_per_second": 7.423,
420
+ "step": 2500
421
+ },
422
+ {
423
+ "epoch": 5.1,
424
+ "grad_norm": 0.1525699645280838,
425
+ "learning_rate": 9.804000000000001e-06,
426
+ "loss": 0.2387,
427
+ "step": 2550
428
+ },
429
+ {
430
+ "epoch": 5.2,
431
+ "grad_norm": 0.08033642917871475,
432
+ "learning_rate": 9.604000000000002e-06,
433
+ "loss": 0.2718,
434
+ "step": 2600
435
+ },
436
+ {
437
+ "epoch": 5.3,
438
+ "grad_norm": 0.4260154366493225,
439
+ "learning_rate": 9.404e-06,
440
+ "loss": 0.3076,
441
+ "step": 2650
442
+ },
443
+ {
444
+ "epoch": 5.4,
445
+ "grad_norm": 13.845452308654785,
446
+ "learning_rate": 9.204e-06,
447
+ "loss": 0.3018,
448
+ "step": 2700
449
+ },
450
+ {
451
+ "epoch": 5.5,
452
+ "grad_norm": 0.18619368970394135,
453
+ "learning_rate": 9.004e-06,
454
+ "loss": 0.1924,
455
+ "step": 2750
456
+ },
457
+ {
458
+ "epoch": 5.6,
459
+ "grad_norm": 0.2769290506839752,
460
+ "learning_rate": 8.804e-06,
461
+ "loss": 0.2033,
462
+ "step": 2800
463
+ },
464
+ {
465
+ "epoch": 5.7,
466
+ "grad_norm": 2.3528101444244385,
467
+ "learning_rate": 8.604000000000001e-06,
468
+ "loss": 0.2645,
469
+ "step": 2850
470
+ },
471
+ {
472
+ "epoch": 5.8,
473
+ "grad_norm": 0.7793205380439758,
474
+ "learning_rate": 8.404000000000001e-06,
475
+ "loss": 0.243,
476
+ "step": 2900
477
+ },
478
+ {
479
+ "epoch": 5.9,
480
+ "grad_norm": 5.553439617156982,
481
+ "learning_rate": 8.204000000000001e-06,
482
+ "loss": 0.3189,
483
+ "step": 2950
484
+ },
485
+ {
486
+ "epoch": 6.0,
487
+ "grad_norm": 0.482384592294693,
488
+ "learning_rate": 8.004e-06,
489
+ "loss": 0.2533,
490
+ "step": 3000
491
+ },
492
+ {
493
+ "epoch": 6.0,
494
+ "eval_accuracy": 0.832,
495
+ "eval_f1": 0.8305888522667962,
496
+ "eval_loss": 0.8593474626541138,
497
+ "eval_precision": 0.8398522097237431,
498
+ "eval_recall": 0.832,
499
+ "eval_runtime": 16.8439,
500
+ "eval_samples_per_second": 59.369,
501
+ "eval_steps_per_second": 7.421,
502
+ "step": 3000
503
+ },
504
+ {
505
+ "epoch": 6.1,
506
+ "grad_norm": 7.8804826736450195,
507
+ "learning_rate": 7.804e-06,
508
+ "loss": 0.2671,
509
+ "step": 3050
510
+ },
511
+ {
512
+ "epoch": 6.2,
513
+ "grad_norm": 0.4820297658443451,
514
+ "learning_rate": 7.604e-06,
515
+ "loss": 0.1255,
516
+ "step": 3100
517
+ },
518
+ {
519
+ "epoch": 6.3,
520
+ "grad_norm": 23.434314727783203,
521
+ "learning_rate": 7.404e-06,
522
+ "loss": 0.2414,
523
+ "step": 3150
524
+ },
525
+ {
526
+ "epoch": 6.4,
527
+ "grad_norm": 0.35348212718963623,
528
+ "learning_rate": 7.204000000000001e-06,
529
+ "loss": 0.2153,
530
+ "step": 3200
531
+ },
532
+ {
533
+ "epoch": 6.5,
534
+ "grad_norm": 33.563053131103516,
535
+ "learning_rate": 7.004000000000001e-06,
536
+ "loss": 0.2589,
537
+ "step": 3250
538
+ },
539
+ {
540
+ "epoch": 6.6,
541
+ "grad_norm": 0.13904573023319244,
542
+ "learning_rate": 6.804e-06,
543
+ "loss": 0.253,
544
+ "step": 3300
545
+ },
546
+ {
547
+ "epoch": 6.7,
548
+ "grad_norm": 0.17627717554569244,
549
+ "learning_rate": 6.604000000000001e-06,
550
+ "loss": 0.3281,
551
+ "step": 3350
552
+ },
553
+ {
554
+ "epoch": 6.8,
555
+ "grad_norm": 0.2751530110836029,
556
+ "learning_rate": 6.404e-06,
557
+ "loss": 0.1926,
558
+ "step": 3400
559
+ },
560
+ {
561
+ "epoch": 6.9,
562
+ "grad_norm": 0.1987292319536209,
563
+ "learning_rate": 6.204e-06,
564
+ "loss": 0.232,
565
+ "step": 3450
566
+ },
567
+ {
568
+ "epoch": 7.0,
569
+ "grad_norm": 0.8847119808197021,
570
+ "learning_rate": 6.004000000000001e-06,
571
+ "loss": 0.1605,
572
+ "step": 3500
573
+ },
574
+ {
575
+ "epoch": 7.0,
576
+ "eval_accuracy": 0.847,
577
+ "eval_f1": 0.8464652675343645,
578
+ "eval_loss": 0.8528432846069336,
579
+ "eval_precision": 0.8499221841477009,
580
+ "eval_recall": 0.847,
581
+ "eval_runtime": 16.8832,
582
+ "eval_samples_per_second": 59.231,
583
+ "eval_steps_per_second": 7.404,
584
+ "step": 3500
585
+ },
586
+ {
587
+ "epoch": 7.1,
588
+ "grad_norm": 5.153850555419922,
589
+ "learning_rate": 5.804000000000001e-06,
590
+ "loss": 0.1543,
591
+ "step": 3550
592
+ },
593
+ {
594
+ "epoch": 7.2,
595
+ "grad_norm": 0.14595018327236176,
596
+ "learning_rate": 5.604000000000001e-06,
597
+ "loss": 0.1685,
598
+ "step": 3600
599
+ },
600
+ {
601
+ "epoch": 7.3,
602
+ "grad_norm": 0.2887927293777466,
603
+ "learning_rate": 5.404e-06,
604
+ "loss": 0.1567,
605
+ "step": 3650
606
+ },
607
+ {
608
+ "epoch": 7.4,
609
+ "grad_norm": 0.2623150646686554,
610
+ "learning_rate": 5.2040000000000005e-06,
611
+ "loss": 0.2714,
612
+ "step": 3700
613
+ },
614
+ {
615
+ "epoch": 7.5,
616
+ "grad_norm": 0.04045703634619713,
617
+ "learning_rate": 5.004e-06,
618
+ "loss": 0.1672,
619
+ "step": 3750
620
+ },
621
+ {
622
+ "epoch": 7.6,
623
+ "grad_norm": 2.894289970397949,
624
+ "learning_rate": 4.804e-06,
625
+ "loss": 0.1748,
626
+ "step": 3800
627
+ },
628
+ {
629
+ "epoch": 7.7,
630
+ "grad_norm": 0.3380846679210663,
631
+ "learning_rate": 4.604e-06,
632
+ "loss": 0.1617,
633
+ "step": 3850
634
+ },
635
+ {
636
+ "epoch": 7.8,
637
+ "grad_norm": 38.41429138183594,
638
+ "learning_rate": 4.4040000000000005e-06,
639
+ "loss": 0.3048,
640
+ "step": 3900
641
+ },
642
+ {
643
+ "epoch": 7.9,
644
+ "grad_norm": 0.3114001154899597,
645
+ "learning_rate": 4.204e-06,
646
+ "loss": 0.1535,
647
+ "step": 3950
648
+ },
649
+ {
650
+ "epoch": 8.0,
651
+ "grad_norm": 0.10832878202199936,
652
+ "learning_rate": 4.004e-06,
653
+ "loss": 0.1855,
654
+ "step": 4000
655
+ },
656
+ {
657
+ "epoch": 8.0,
658
+ "eval_accuracy": 0.847,
659
+ "eval_f1": 0.8465633748668708,
660
+ "eval_loss": 0.8490852117538452,
661
+ "eval_precision": 0.8483482604799019,
662
+ "eval_recall": 0.847,
663
+ "eval_runtime": 16.5292,
664
+ "eval_samples_per_second": 60.499,
665
+ "eval_steps_per_second": 7.562,
666
+ "step": 4000
667
+ },
668
+ {
669
+ "epoch": 8.1,
670
+ "grad_norm": 0.09421584755182266,
671
+ "learning_rate": 3.8040000000000003e-06,
672
+ "loss": 0.1689,
673
+ "step": 4050
674
+ },
675
+ {
676
+ "epoch": 8.2,
677
+ "grad_norm": 1.5713932514190674,
678
+ "learning_rate": 3.604e-06,
679
+ "loss": 0.1557,
680
+ "step": 4100
681
+ },
682
+ {
683
+ "epoch": 8.3,
684
+ "grad_norm": 0.2470749169588089,
685
+ "learning_rate": 3.404e-06,
686
+ "loss": 0.1846,
687
+ "step": 4150
688
+ },
689
+ {
690
+ "epoch": 8.4,
691
+ "grad_norm": 1.1986793279647827,
692
+ "learning_rate": 3.2040000000000006e-06,
693
+ "loss": 0.1269,
694
+ "step": 4200
695
+ },
696
+ {
697
+ "epoch": 8.5,
698
+ "grad_norm": 4.091843605041504,
699
+ "learning_rate": 3.0040000000000004e-06,
700
+ "loss": 0.1492,
701
+ "step": 4250
702
+ },
703
+ {
704
+ "epoch": 8.6,
705
+ "grad_norm": 0.12018954753875732,
706
+ "learning_rate": 2.804e-06,
707
+ "loss": 0.1986,
708
+ "step": 4300
709
+ },
710
+ {
711
+ "epoch": 8.7,
712
+ "grad_norm": 0.3240673243999481,
713
+ "learning_rate": 2.6040000000000004e-06,
714
+ "loss": 0.1442,
715
+ "step": 4350
716
+ },
717
+ {
718
+ "epoch": 8.8,
719
+ "grad_norm": 3.653069496154785,
720
+ "learning_rate": 2.404e-06,
721
+ "loss": 0.1531,
722
+ "step": 4400
723
+ },
724
+ {
725
+ "epoch": 8.9,
726
+ "grad_norm": 5.728586196899414,
727
+ "learning_rate": 2.2040000000000004e-06,
728
+ "loss": 0.2052,
729
+ "step": 4450
730
+ },
731
+ {
732
+ "epoch": 9.0,
733
+ "grad_norm": 7.657519340515137,
734
+ "learning_rate": 2.004e-06,
735
+ "loss": 0.1738,
736
+ "step": 4500
737
+ },
738
+ {
739
+ "epoch": 9.0,
740
+ "eval_accuracy": 0.852,
741
+ "eval_f1": 0.8512761651471602,
742
+ "eval_loss": 0.8877223134040833,
743
+ "eval_precision": 0.8546560428736699,
744
+ "eval_recall": 0.852,
745
+ "eval_runtime": 16.5529,
746
+ "eval_samples_per_second": 60.412,
747
+ "eval_steps_per_second": 7.552,
748
+ "step": 4500
749
+ }
750
+ ],
751
+ "logging_steps": 50,
752
+ "max_steps": 5000,
753
+ "num_input_tokens_seen": 0,
754
+ "num_train_epochs": 10,
755
+ "save_steps": 500,
756
+ "stateful_callbacks": {
757
+ "EarlyStoppingCallback": {
758
+ "args": {
759
+ "early_stopping_patience": 2,
760
+ "early_stopping_threshold": 0.0
761
+ },
762
+ "attributes": {
763
+ "early_stopping_patience_counter": 0
764
+ }
765
+ },
766
+ "TrainerControl": {
767
+ "args": {
768
+ "should_epoch_stop": false,
769
+ "should_evaluate": false,
770
+ "should_log": false,
771
+ "should_save": true,
772
+ "should_training_stop": false
773
+ },
774
+ "attributes": {}
775
+ }
776
+ },
777
+ "total_flos": 9473612083200000.0,
778
+ "train_batch_size": 8,
779
+ "trial_name": null,
780
+ "trial_params": null
781
+ }
checkpoint-4500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89b560840c2c90d1414d71d66ff5e4a2e615b3668acd2084f3457d01ca8c7238
3
+ size 5304
checkpoint-4500/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-5000/config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BrosForDocumentClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bbox_scale": 100.0,
7
+ "classifier_dropout_prob": 0.1,
8
+ "dim_bbox": 8,
9
+ "dim_bbox_projection": 64,
10
+ "dim_bbox_sinusoid_emb_1d": 24,
11
+ "dim_bbox_sinusoid_emb_2d": 192,
12
+ "hidden_act": "gelu",
13
+ "hidden_dropout_prob": 0.1,
14
+ "hidden_size": 768,
15
+ "id2label": {
16
+ "0": "form",
17
+ "1": "invoice",
18
+ "2": "budget",
19
+ "3": "file folder",
20
+ "4": "questionnaire"
21
+ },
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 3072,
24
+ "label2id": {
25
+ "budget": 2,
26
+ "file folder": 3,
27
+ "form": 0,
28
+ "invoice": 1,
29
+ "questionnaire": 4
30
+ },
31
+ "layer_norm_eps": 1e-12,
32
+ "max_position_embeddings": 512,
33
+ "model_type": "bros",
34
+ "n_relations": 1,
35
+ "num_attention_heads": 12,
36
+ "num_hidden_layers": 12,
37
+ "pad_token_id": 0,
38
+ "pe_type": "crel",
39
+ "torch_dtype": "float32",
40
+ "transformers_version": "4.53.0",
41
+ "type_vocab_size": 2,
42
+ "vocab_size": 30522
43
+ }
checkpoint-5000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2eaa63b6e01524ef3b252d1870b11dd3a8d4af77eb549b42111575188fc50a04
3
+ size 438021652
checkpoint-5000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:576cefa562bc053a8f0e5d49c9dd6d2a1de824d51706ba28454fc843689323e9
3
+ size 871429498
checkpoint-5000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02963f477e31738ebaabc79695fb55551b27b4f9145cf72586be3594d7af3397
3
+ size 14244
checkpoint-5000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2be67f6aac8e482bb2022409709d8774ffb125292c0c9cf025c0ae747f3a6d57
3
+ size 1064
checkpoint-5000/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-5000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-5000/tokenizer_config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "mask_token": "[MASK]",
50
+ "model_max_length": 1000000000000000019884624838656,
51
+ "never_split": null,
52
+ "pad_token": "[PAD]",
53
+ "sep_token": "[SEP]",
54
+ "strip_accents": null,
55
+ "tokenize_chinese_chars": true,
56
+ "tokenizer_class": "BertTokenizer",
57
+ "unk_token": "[UNK]"
58
+ }
checkpoint-5000/trainer_state.json ADDED
@@ -0,0 +1,863 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 4500,
3
+ "best_metric": 0.852,
4
+ "best_model_checkpoint": "./bros-docclass-finetuned/checkpoint-4500",
5
+ "epoch": 10.0,
6
+ "eval_steps": 500,
7
+ "global_step": 5000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.1,
14
+ "grad_norm": 9.038439750671387,
15
+ "learning_rate": 1.9804000000000002e-05,
16
+ "loss": 1.2156,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.2,
21
+ "grad_norm": 12.505393028259277,
22
+ "learning_rate": 1.9604e-05,
23
+ "loss": 0.8366,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.3,
28
+ "grad_norm": 17.0165958404541,
29
+ "learning_rate": 1.9404e-05,
30
+ "loss": 0.7993,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.4,
35
+ "grad_norm": 25.100793838500977,
36
+ "learning_rate": 1.9204e-05,
37
+ "loss": 0.6683,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 0.5,
42
+ "grad_norm": 12.89208698272705,
43
+ "learning_rate": 1.9004000000000003e-05,
44
+ "loss": 0.7392,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 0.6,
49
+ "grad_norm": 19.72749137878418,
50
+ "learning_rate": 1.8804e-05,
51
+ "loss": 0.666,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.7,
56
+ "grad_norm": 30.814998626708984,
57
+ "learning_rate": 1.8604000000000003e-05,
58
+ "loss": 0.6453,
59
+ "step": 350
60
+ },
61
+ {
62
+ "epoch": 0.8,
63
+ "grad_norm": 1.3695790767669678,
64
+ "learning_rate": 1.8404000000000002e-05,
65
+ "loss": 0.6402,
66
+ "step": 400
67
+ },
68
+ {
69
+ "epoch": 0.9,
70
+ "grad_norm": 3.071685791015625,
71
+ "learning_rate": 1.8204e-05,
72
+ "loss": 0.6646,
73
+ "step": 450
74
+ },
75
+ {
76
+ "epoch": 1.0,
77
+ "grad_norm": 19.854291915893555,
78
+ "learning_rate": 1.8004000000000002e-05,
79
+ "loss": 0.6606,
80
+ "step": 500
81
+ },
82
+ {
83
+ "epoch": 1.0,
84
+ "eval_accuracy": 0.798,
85
+ "eval_f1": 0.7962131024443857,
86
+ "eval_loss": 0.6697192788124084,
87
+ "eval_precision": 0.8016786327842085,
88
+ "eval_recall": 0.798,
89
+ "eval_runtime": 16.5436,
90
+ "eval_samples_per_second": 60.446,
91
+ "eval_steps_per_second": 7.556,
92
+ "step": 500
93
+ },
94
+ {
95
+ "epoch": 1.1,
96
+ "grad_norm": 3.369779109954834,
97
+ "learning_rate": 1.7804e-05,
98
+ "loss": 0.5355,
99
+ "step": 550
100
+ },
101
+ {
102
+ "epoch": 1.2,
103
+ "grad_norm": 20.186811447143555,
104
+ "learning_rate": 1.7604e-05,
105
+ "loss": 0.5781,
106
+ "step": 600
107
+ },
108
+ {
109
+ "epoch": 1.3,
110
+ "grad_norm": 19.78652000427246,
111
+ "learning_rate": 1.7404e-05,
112
+ "loss": 0.5731,
113
+ "step": 650
114
+ },
115
+ {
116
+ "epoch": 1.4,
117
+ "grad_norm": 0.9622361063957214,
118
+ "learning_rate": 1.7204e-05,
119
+ "loss": 0.5907,
120
+ "step": 700
121
+ },
122
+ {
123
+ "epoch": 1.5,
124
+ "grad_norm": 22.184574127197266,
125
+ "learning_rate": 1.7004000000000002e-05,
126
+ "loss": 0.4993,
127
+ "step": 750
128
+ },
129
+ {
130
+ "epoch": 1.6,
131
+ "grad_norm": 37.49524688720703,
132
+ "learning_rate": 1.6804e-05,
133
+ "loss": 0.6767,
134
+ "step": 800
135
+ },
136
+ {
137
+ "epoch": 1.7,
138
+ "grad_norm": 0.731647789478302,
139
+ "learning_rate": 1.6604000000000002e-05,
140
+ "loss": 0.6219,
141
+ "step": 850
142
+ },
143
+ {
144
+ "epoch": 1.8,
145
+ "grad_norm": 5.991142272949219,
146
+ "learning_rate": 1.6404e-05,
147
+ "loss": 0.5425,
148
+ "step": 900
149
+ },
150
+ {
151
+ "epoch": 1.9,
152
+ "grad_norm": 35.003902435302734,
153
+ "learning_rate": 1.6204000000000003e-05,
154
+ "loss": 0.5822,
155
+ "step": 950
156
+ },
157
+ {
158
+ "epoch": 2.0,
159
+ "grad_norm": 7.839473247528076,
160
+ "learning_rate": 1.6004e-05,
161
+ "loss": 0.5845,
162
+ "step": 1000
163
+ },
164
+ {
165
+ "epoch": 2.0,
166
+ "eval_accuracy": 0.828,
167
+ "eval_f1": 0.8261256801910406,
168
+ "eval_loss": 0.6260756254196167,
169
+ "eval_precision": 0.8353381322045145,
170
+ "eval_recall": 0.828,
171
+ "eval_runtime": 16.5173,
172
+ "eval_samples_per_second": 60.543,
173
+ "eval_steps_per_second": 7.568,
174
+ "step": 1000
175
+ },
176
+ {
177
+ "epoch": 2.1,
178
+ "grad_norm": 22.77824592590332,
179
+ "learning_rate": 1.5804000000000003e-05,
180
+ "loss": 0.526,
181
+ "step": 1050
182
+ },
183
+ {
184
+ "epoch": 2.2,
185
+ "grad_norm": 28.53985023498535,
186
+ "learning_rate": 1.5604000000000002e-05,
187
+ "loss": 0.3672,
188
+ "step": 1100
189
+ },
190
+ {
191
+ "epoch": 2.3,
192
+ "grad_norm": 13.204597473144531,
193
+ "learning_rate": 1.5404e-05,
194
+ "loss": 0.521,
195
+ "step": 1150
196
+ },
197
+ {
198
+ "epoch": 2.4,
199
+ "grad_norm": 33.86636734008789,
200
+ "learning_rate": 1.5204e-05,
201
+ "loss": 0.4376,
202
+ "step": 1200
203
+ },
204
+ {
205
+ "epoch": 2.5,
206
+ "grad_norm": 3.819549798965454,
207
+ "learning_rate": 1.5004e-05,
208
+ "loss": 0.5127,
209
+ "step": 1250
210
+ },
211
+ {
212
+ "epoch": 2.6,
213
+ "grad_norm": 9.592472076416016,
214
+ "learning_rate": 1.4804000000000001e-05,
215
+ "loss": 0.4313,
216
+ "step": 1300
217
+ },
218
+ {
219
+ "epoch": 2.7,
220
+ "grad_norm": 2.242670774459839,
221
+ "learning_rate": 1.4604000000000001e-05,
222
+ "loss": 0.4445,
223
+ "step": 1350
224
+ },
225
+ {
226
+ "epoch": 2.8,
227
+ "grad_norm": 61.427616119384766,
228
+ "learning_rate": 1.4404e-05,
229
+ "loss": 0.4974,
230
+ "step": 1400
231
+ },
232
+ {
233
+ "epoch": 2.9,
234
+ "grad_norm": 0.4456544816493988,
235
+ "learning_rate": 1.4204000000000002e-05,
236
+ "loss": 0.4999,
237
+ "step": 1450
238
+ },
239
+ {
240
+ "epoch": 3.0,
241
+ "grad_norm": 33.89624786376953,
242
+ "learning_rate": 1.4004000000000002e-05,
243
+ "loss": 0.4223,
244
+ "step": 1500
245
+ },
246
+ {
247
+ "epoch": 3.0,
248
+ "eval_accuracy": 0.818,
249
+ "eval_f1": 0.8172688666919534,
250
+ "eval_loss": 0.7300030589103699,
251
+ "eval_precision": 0.8328778365646615,
252
+ "eval_recall": 0.818,
253
+ "eval_runtime": 16.4984,
254
+ "eval_samples_per_second": 60.612,
255
+ "eval_steps_per_second": 7.576,
256
+ "step": 1500
257
+ },
258
+ {
259
+ "epoch": 3.1,
260
+ "grad_norm": 18.57257652282715,
261
+ "learning_rate": 1.3804000000000002e-05,
262
+ "loss": 0.4925,
263
+ "step": 1550
264
+ },
265
+ {
266
+ "epoch": 3.2,
267
+ "grad_norm": 40.097713470458984,
268
+ "learning_rate": 1.3604000000000002e-05,
269
+ "loss": 0.2834,
270
+ "step": 1600
271
+ },
272
+ {
273
+ "epoch": 3.3,
274
+ "grad_norm": 191.4404754638672,
275
+ "learning_rate": 1.3404e-05,
276
+ "loss": 0.4128,
277
+ "step": 1650
278
+ },
279
+ {
280
+ "epoch": 3.4,
281
+ "grad_norm": 67.55642700195312,
282
+ "learning_rate": 1.3204000000000001e-05,
283
+ "loss": 0.3483,
284
+ "step": 1700
285
+ },
286
+ {
287
+ "epoch": 3.5,
288
+ "grad_norm": 26.829570770263672,
289
+ "learning_rate": 1.3004000000000001e-05,
290
+ "loss": 0.3372,
291
+ "step": 1750
292
+ },
293
+ {
294
+ "epoch": 3.6,
295
+ "grad_norm": 12.817995071411133,
296
+ "learning_rate": 1.2804000000000001e-05,
297
+ "loss": 0.4087,
298
+ "step": 1800
299
+ },
300
+ {
301
+ "epoch": 3.7,
302
+ "grad_norm": 1.6152303218841553,
303
+ "learning_rate": 1.2604e-05,
304
+ "loss": 0.4033,
305
+ "step": 1850
306
+ },
307
+ {
308
+ "epoch": 3.8,
309
+ "grad_norm": 0.07653416693210602,
310
+ "learning_rate": 1.2404e-05,
311
+ "loss": 0.3495,
312
+ "step": 1900
313
+ },
314
+ {
315
+ "epoch": 3.9,
316
+ "grad_norm": 2.4822909832000732,
317
+ "learning_rate": 1.2204e-05,
318
+ "loss": 0.4195,
319
+ "step": 1950
320
+ },
321
+ {
322
+ "epoch": 4.0,
323
+ "grad_norm": 91.26575469970703,
324
+ "learning_rate": 1.2004e-05,
325
+ "loss": 0.3644,
326
+ "step": 2000
327
+ },
328
+ {
329
+ "epoch": 4.0,
330
+ "eval_accuracy": 0.839,
331
+ "eval_f1": 0.8384490586281527,
332
+ "eval_loss": 0.7187097668647766,
333
+ "eval_precision": 0.8450004509095641,
334
+ "eval_recall": 0.839,
335
+ "eval_runtime": 16.8707,
336
+ "eval_samples_per_second": 59.274,
337
+ "eval_steps_per_second": 7.409,
338
+ "step": 2000
339
+ },
340
+ {
341
+ "epoch": 4.1,
342
+ "grad_norm": 4.009641170501709,
343
+ "learning_rate": 1.1803999999999999e-05,
344
+ "loss": 0.3486,
345
+ "step": 2050
346
+ },
347
+ {
348
+ "epoch": 4.2,
349
+ "grad_norm": 16.93126106262207,
350
+ "learning_rate": 1.1604000000000003e-05,
351
+ "loss": 0.3628,
352
+ "step": 2100
353
+ },
354
+ {
355
+ "epoch": 4.3,
356
+ "grad_norm": 65.41966247558594,
357
+ "learning_rate": 1.1404000000000001e-05,
358
+ "loss": 0.2832,
359
+ "step": 2150
360
+ },
361
+ {
362
+ "epoch": 4.4,
363
+ "grad_norm": 36.100013732910156,
364
+ "learning_rate": 1.1204000000000001e-05,
365
+ "loss": 0.3515,
366
+ "step": 2200
367
+ },
368
+ {
369
+ "epoch": 4.5,
370
+ "grad_norm": 13.878293991088867,
371
+ "learning_rate": 1.1004000000000002e-05,
372
+ "loss": 0.282,
373
+ "step": 2250
374
+ },
375
+ {
376
+ "epoch": 4.6,
377
+ "grad_norm": 82.44599914550781,
378
+ "learning_rate": 1.0804000000000002e-05,
379
+ "loss": 0.2804,
380
+ "step": 2300
381
+ },
382
+ {
383
+ "epoch": 4.7,
384
+ "grad_norm": 3.182461977005005,
385
+ "learning_rate": 1.0604e-05,
386
+ "loss": 0.2802,
387
+ "step": 2350
388
+ },
389
+ {
390
+ "epoch": 4.8,
391
+ "grad_norm": 0.41410791873931885,
392
+ "learning_rate": 1.0404e-05,
393
+ "loss": 0.309,
394
+ "step": 2400
395
+ },
396
+ {
397
+ "epoch": 4.9,
398
+ "grad_norm": 37.641319274902344,
399
+ "learning_rate": 1.0204000000000001e-05,
400
+ "loss": 0.284,
401
+ "step": 2450
402
+ },
403
+ {
404
+ "epoch": 5.0,
405
+ "grad_norm": 3.98016619682312,
406
+ "learning_rate": 1.0004000000000001e-05,
407
+ "loss": 0.3447,
408
+ "step": 2500
409
+ },
410
+ {
411
+ "epoch": 5.0,
412
+ "eval_accuracy": 0.844,
413
+ "eval_f1": 0.844281271446929,
414
+ "eval_loss": 0.7856075167655945,
415
+ "eval_precision": 0.8486261672220536,
416
+ "eval_recall": 0.844,
417
+ "eval_runtime": 16.8405,
418
+ "eval_samples_per_second": 59.381,
419
+ "eval_steps_per_second": 7.423,
420
+ "step": 2500
421
+ },
422
+ {
423
+ "epoch": 5.1,
424
+ "grad_norm": 0.1525699645280838,
425
+ "learning_rate": 9.804000000000001e-06,
426
+ "loss": 0.2387,
427
+ "step": 2550
428
+ },
429
+ {
430
+ "epoch": 5.2,
431
+ "grad_norm": 0.08033642917871475,
432
+ "learning_rate": 9.604000000000002e-06,
433
+ "loss": 0.2718,
434
+ "step": 2600
435
+ },
436
+ {
437
+ "epoch": 5.3,
438
+ "grad_norm": 0.4260154366493225,
439
+ "learning_rate": 9.404e-06,
440
+ "loss": 0.3076,
441
+ "step": 2650
442
+ },
443
+ {
444
+ "epoch": 5.4,
445
+ "grad_norm": 13.845452308654785,
446
+ "learning_rate": 9.204e-06,
447
+ "loss": 0.3018,
448
+ "step": 2700
449
+ },
450
+ {
451
+ "epoch": 5.5,
452
+ "grad_norm": 0.18619368970394135,
453
+ "learning_rate": 9.004e-06,
454
+ "loss": 0.1924,
455
+ "step": 2750
456
+ },
457
+ {
458
+ "epoch": 5.6,
459
+ "grad_norm": 0.2769290506839752,
460
+ "learning_rate": 8.804e-06,
461
+ "loss": 0.2033,
462
+ "step": 2800
463
+ },
464
+ {
465
+ "epoch": 5.7,
466
+ "grad_norm": 2.3528101444244385,
467
+ "learning_rate": 8.604000000000001e-06,
468
+ "loss": 0.2645,
469
+ "step": 2850
470
+ },
471
+ {
472
+ "epoch": 5.8,
473
+ "grad_norm": 0.7793205380439758,
474
+ "learning_rate": 8.404000000000001e-06,
475
+ "loss": 0.243,
476
+ "step": 2900
477
+ },
478
+ {
479
+ "epoch": 5.9,
480
+ "grad_norm": 5.553439617156982,
481
+ "learning_rate": 8.204000000000001e-06,
482
+ "loss": 0.3189,
483
+ "step": 2950
484
+ },
485
+ {
486
+ "epoch": 6.0,
487
+ "grad_norm": 0.482384592294693,
488
+ "learning_rate": 8.004e-06,
489
+ "loss": 0.2533,
490
+ "step": 3000
491
+ },
492
+ {
493
+ "epoch": 6.0,
494
+ "eval_accuracy": 0.832,
495
+ "eval_f1": 0.8305888522667962,
496
+ "eval_loss": 0.8593474626541138,
497
+ "eval_precision": 0.8398522097237431,
498
+ "eval_recall": 0.832,
499
+ "eval_runtime": 16.8439,
500
+ "eval_samples_per_second": 59.369,
501
+ "eval_steps_per_second": 7.421,
502
+ "step": 3000
503
+ },
504
+ {
505
+ "epoch": 6.1,
506
+ "grad_norm": 7.8804826736450195,
507
+ "learning_rate": 7.804e-06,
508
+ "loss": 0.2671,
509
+ "step": 3050
510
+ },
511
+ {
512
+ "epoch": 6.2,
513
+ "grad_norm": 0.4820297658443451,
514
+ "learning_rate": 7.604e-06,
515
+ "loss": 0.1255,
516
+ "step": 3100
517
+ },
518
+ {
519
+ "epoch": 6.3,
520
+ "grad_norm": 23.434314727783203,
521
+ "learning_rate": 7.404e-06,
522
+ "loss": 0.2414,
523
+ "step": 3150
524
+ },
525
+ {
526
+ "epoch": 6.4,
527
+ "grad_norm": 0.35348212718963623,
528
+ "learning_rate": 7.204000000000001e-06,
529
+ "loss": 0.2153,
530
+ "step": 3200
531
+ },
532
+ {
533
+ "epoch": 6.5,
534
+ "grad_norm": 33.563053131103516,
535
+ "learning_rate": 7.004000000000001e-06,
536
+ "loss": 0.2589,
537
+ "step": 3250
538
+ },
539
+ {
540
+ "epoch": 6.6,
541
+ "grad_norm": 0.13904573023319244,
542
+ "learning_rate": 6.804e-06,
543
+ "loss": 0.253,
544
+ "step": 3300
545
+ },
546
+ {
547
+ "epoch": 6.7,
548
+ "grad_norm": 0.17627717554569244,
549
+ "learning_rate": 6.604000000000001e-06,
550
+ "loss": 0.3281,
551
+ "step": 3350
552
+ },
553
+ {
554
+ "epoch": 6.8,
555
+ "grad_norm": 0.2751530110836029,
556
+ "learning_rate": 6.404e-06,
557
+ "loss": 0.1926,
558
+ "step": 3400
559
+ },
560
+ {
561
+ "epoch": 6.9,
562
+ "grad_norm": 0.1987292319536209,
563
+ "learning_rate": 6.204e-06,
564
+ "loss": 0.232,
565
+ "step": 3450
566
+ },
567
+ {
568
+ "epoch": 7.0,
569
+ "grad_norm": 0.8847119808197021,
570
+ "learning_rate": 6.004000000000001e-06,
571
+ "loss": 0.1605,
572
+ "step": 3500
573
+ },
574
+ {
575
+ "epoch": 7.0,
576
+ "eval_accuracy": 0.847,
577
+ "eval_f1": 0.8464652675343645,
578
+ "eval_loss": 0.8528432846069336,
579
+ "eval_precision": 0.8499221841477009,
580
+ "eval_recall": 0.847,
581
+ "eval_runtime": 16.8832,
582
+ "eval_samples_per_second": 59.231,
583
+ "eval_steps_per_second": 7.404,
584
+ "step": 3500
585
+ },
586
+ {
587
+ "epoch": 7.1,
588
+ "grad_norm": 5.153850555419922,
589
+ "learning_rate": 5.804000000000001e-06,
590
+ "loss": 0.1543,
591
+ "step": 3550
592
+ },
593
+ {
594
+ "epoch": 7.2,
595
+ "grad_norm": 0.14595018327236176,
596
+ "learning_rate": 5.604000000000001e-06,
597
+ "loss": 0.1685,
598
+ "step": 3600
599
+ },
600
+ {
601
+ "epoch": 7.3,
602
+ "grad_norm": 0.2887927293777466,
603
+ "learning_rate": 5.404e-06,
604
+ "loss": 0.1567,
605
+ "step": 3650
606
+ },
607
+ {
608
+ "epoch": 7.4,
609
+ "grad_norm": 0.2623150646686554,
610
+ "learning_rate": 5.2040000000000005e-06,
611
+ "loss": 0.2714,
612
+ "step": 3700
613
+ },
614
+ {
615
+ "epoch": 7.5,
616
+ "grad_norm": 0.04045703634619713,
617
+ "learning_rate": 5.004e-06,
618
+ "loss": 0.1672,
619
+ "step": 3750
620
+ },
621
+ {
622
+ "epoch": 7.6,
623
+ "grad_norm": 2.894289970397949,
624
+ "learning_rate": 4.804e-06,
625
+ "loss": 0.1748,
626
+ "step": 3800
627
+ },
628
+ {
629
+ "epoch": 7.7,
630
+ "grad_norm": 0.3380846679210663,
631
+ "learning_rate": 4.604e-06,
632
+ "loss": 0.1617,
633
+ "step": 3850
634
+ },
635
+ {
636
+ "epoch": 7.8,
637
+ "grad_norm": 38.41429138183594,
638
+ "learning_rate": 4.4040000000000005e-06,
639
+ "loss": 0.3048,
640
+ "step": 3900
641
+ },
642
+ {
643
+ "epoch": 7.9,
644
+ "grad_norm": 0.3114001154899597,
645
+ "learning_rate": 4.204e-06,
646
+ "loss": 0.1535,
647
+ "step": 3950
648
+ },
649
+ {
650
+ "epoch": 8.0,
651
+ "grad_norm": 0.10832878202199936,
652
+ "learning_rate": 4.004e-06,
653
+ "loss": 0.1855,
654
+ "step": 4000
655
+ },
656
+ {
657
+ "epoch": 8.0,
658
+ "eval_accuracy": 0.847,
659
+ "eval_f1": 0.8465633748668708,
660
+ "eval_loss": 0.8490852117538452,
661
+ "eval_precision": 0.8483482604799019,
662
+ "eval_recall": 0.847,
663
+ "eval_runtime": 16.5292,
664
+ "eval_samples_per_second": 60.499,
665
+ "eval_steps_per_second": 7.562,
666
+ "step": 4000
667
+ },
668
+ {
669
+ "epoch": 8.1,
670
+ "grad_norm": 0.09421584755182266,
671
+ "learning_rate": 3.8040000000000003e-06,
672
+ "loss": 0.1689,
673
+ "step": 4050
674
+ },
675
+ {
676
+ "epoch": 8.2,
677
+ "grad_norm": 1.5713932514190674,
678
+ "learning_rate": 3.604e-06,
679
+ "loss": 0.1557,
680
+ "step": 4100
681
+ },
682
+ {
683
+ "epoch": 8.3,
684
+ "grad_norm": 0.2470749169588089,
685
+ "learning_rate": 3.404e-06,
686
+ "loss": 0.1846,
687
+ "step": 4150
688
+ },
689
+ {
690
+ "epoch": 8.4,
691
+ "grad_norm": 1.1986793279647827,
692
+ "learning_rate": 3.2040000000000006e-06,
693
+ "loss": 0.1269,
694
+ "step": 4200
695
+ },
696
+ {
697
+ "epoch": 8.5,
698
+ "grad_norm": 4.091843605041504,
699
+ "learning_rate": 3.0040000000000004e-06,
700
+ "loss": 0.1492,
701
+ "step": 4250
702
+ },
703
+ {
704
+ "epoch": 8.6,
705
+ "grad_norm": 0.12018954753875732,
706
+ "learning_rate": 2.804e-06,
707
+ "loss": 0.1986,
708
+ "step": 4300
709
+ },
710
+ {
711
+ "epoch": 8.7,
712
+ "grad_norm": 0.3240673243999481,
713
+ "learning_rate": 2.6040000000000004e-06,
714
+ "loss": 0.1442,
715
+ "step": 4350
716
+ },
717
+ {
718
+ "epoch": 8.8,
719
+ "grad_norm": 3.653069496154785,
720
+ "learning_rate": 2.404e-06,
721
+ "loss": 0.1531,
722
+ "step": 4400
723
+ },
724
+ {
725
+ "epoch": 8.9,
726
+ "grad_norm": 5.728586196899414,
727
+ "learning_rate": 2.2040000000000004e-06,
728
+ "loss": 0.2052,
729
+ "step": 4450
730
+ },
731
+ {
732
+ "epoch": 9.0,
733
+ "grad_norm": 7.657519340515137,
734
+ "learning_rate": 2.004e-06,
735
+ "loss": 0.1738,
736
+ "step": 4500
737
+ },
738
+ {
739
+ "epoch": 9.0,
740
+ "eval_accuracy": 0.852,
741
+ "eval_f1": 0.8512761651471602,
742
+ "eval_loss": 0.8877223134040833,
743
+ "eval_precision": 0.8546560428736699,
744
+ "eval_recall": 0.852,
745
+ "eval_runtime": 16.5529,
746
+ "eval_samples_per_second": 60.412,
747
+ "eval_steps_per_second": 7.552,
748
+ "step": 4500
749
+ },
750
+ {
751
+ "epoch": 9.1,
752
+ "grad_norm": 5.215471267700195,
753
+ "learning_rate": 1.8040000000000002e-06,
754
+ "loss": 0.1422,
755
+ "step": 4550
756
+ },
757
+ {
758
+ "epoch": 9.2,
759
+ "grad_norm": 0.04474487900733948,
760
+ "learning_rate": 1.604e-06,
761
+ "loss": 0.1342,
762
+ "step": 4600
763
+ },
764
+ {
765
+ "epoch": 9.3,
766
+ "grad_norm": 0.04985995218157768,
767
+ "learning_rate": 1.404e-06,
768
+ "loss": 0.1891,
769
+ "step": 4650
770
+ },
771
+ {
772
+ "epoch": 9.4,
773
+ "grad_norm": 0.4687720239162445,
774
+ "learning_rate": 1.204e-06,
775
+ "loss": 0.1469,
776
+ "step": 4700
777
+ },
778
+ {
779
+ "epoch": 9.5,
780
+ "grad_norm": 6.679148197174072,
781
+ "learning_rate": 1.004e-06,
782
+ "loss": 0.1893,
783
+ "step": 4750
784
+ },
785
+ {
786
+ "epoch": 9.6,
787
+ "grad_norm": 0.26801905035972595,
788
+ "learning_rate": 8.04e-07,
789
+ "loss": 0.1194,
790
+ "step": 4800
791
+ },
792
+ {
793
+ "epoch": 9.7,
794
+ "grad_norm": 0.0896608754992485,
795
+ "learning_rate": 6.040000000000001e-07,
796
+ "loss": 0.1187,
797
+ "step": 4850
798
+ },
799
+ {
800
+ "epoch": 9.8,
801
+ "grad_norm": 0.16843467950820923,
802
+ "learning_rate": 4.04e-07,
803
+ "loss": 0.1472,
804
+ "step": 4900
805
+ },
806
+ {
807
+ "epoch": 9.9,
808
+ "grad_norm": 0.023157618939876556,
809
+ "learning_rate": 2.0400000000000003e-07,
810
+ "loss": 0.1273,
811
+ "step": 4950
812
+ },
813
+ {
814
+ "epoch": 10.0,
815
+ "grad_norm": 9.227570533752441,
816
+ "learning_rate": 4e-09,
817
+ "loss": 0.1248,
818
+ "step": 5000
819
+ },
820
+ {
821
+ "epoch": 10.0,
822
+ "eval_accuracy": 0.846,
823
+ "eval_f1": 0.8453734677993351,
824
+ "eval_loss": 0.8945107460021973,
825
+ "eval_precision": 0.8485569397622855,
826
+ "eval_recall": 0.846,
827
+ "eval_runtime": 16.575,
828
+ "eval_samples_per_second": 60.332,
829
+ "eval_steps_per_second": 7.541,
830
+ "step": 5000
831
+ }
832
+ ],
833
+ "logging_steps": 50,
834
+ "max_steps": 5000,
835
+ "num_input_tokens_seen": 0,
836
+ "num_train_epochs": 10,
837
+ "save_steps": 500,
838
+ "stateful_callbacks": {
839
+ "EarlyStoppingCallback": {
840
+ "args": {
841
+ "early_stopping_patience": 2,
842
+ "early_stopping_threshold": 0.0
843
+ },
844
+ "attributes": {
845
+ "early_stopping_patience_counter": 1
846
+ }
847
+ },
848
+ "TrainerControl": {
849
+ "args": {
850
+ "should_epoch_stop": false,
851
+ "should_evaluate": false,
852
+ "should_log": false,
853
+ "should_save": true,
854
+ "should_training_stop": true
855
+ },
856
+ "attributes": {}
857
+ }
858
+ },
859
+ "total_flos": 1.0526235648e+16,
860
+ "train_batch_size": 8,
861
+ "trial_name": null,
862
+ "trial_params": null
863
+ }
checkpoint-5000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89b560840c2c90d1414d71d66ff5e4a2e615b3668acd2084f3457d01ca8c7238
3
+ size 5304
checkpoint-5000/vocab.txt ADDED
The diff for this file is too large to render. See raw diff