Hugging Face's logo

itsGoodman
/
electra_small_heading_classifier

Safetensors
electra
Model card Files
xet
Community
itsGoodman commited on
Commit
cbb8bc0
·
verified ·
1 Parent(s): 36b4c0b

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. checkpoint-1497/config.json +38 -0
  2. checkpoint-1497/model.safetensors +3 -0
  3. checkpoint-1497/optimizer.pt +3 -0
  4. checkpoint-1497/rng_state.pth +3 -0
  5. checkpoint-1497/scheduler.pt +3 -0
  6. checkpoint-1497/special_tokens_map.json +7 -0
  7. checkpoint-1497/tokenizer.json +0 -0
  8. checkpoint-1497/tokenizer_config.json +56 -0
  9. checkpoint-1497/trainer_state.json +247 -0
  10. checkpoint-1497/training_args.bin +3 -0
  11. checkpoint-1497/vocab.txt +0 -0
  12. checkpoint-2994/config.json +38 -0
  13. checkpoint-2994/model.safetensors +3 -0
  14. checkpoint-2994/optimizer.pt +3 -0
  15. checkpoint-2994/rng_state.pth +3 -0
  16. checkpoint-2994/scheduler.pt +3 -0
  17. checkpoint-2994/special_tokens_map.json +7 -0
  18. checkpoint-2994/tokenizer.json +0 -0
  19. checkpoint-2994/tokenizer_config.json +56 -0
  20. checkpoint-2994/trainer_state.json +467 -0
  21. checkpoint-2994/training_args.bin +3 -0
  22. checkpoint-2994/vocab.txt +0 -0
  23. checkpoint-4491/config.json +38 -0
  24. checkpoint-4491/model.safetensors +3 -0
  25. checkpoint-4491/optimizer.pt +3 -0
  26. checkpoint-4491/rng_state.pth +3 -0
  27. checkpoint-4491/scheduler.pt +3 -0
  28. checkpoint-4491/special_tokens_map.json +7 -0
  29. checkpoint-4491/tokenizer.json +0 -0
  30. checkpoint-4491/tokenizer_config.json +56 -0
  31. checkpoint-4491/trainer_state.json +687 -0
  32. checkpoint-4491/training_args.bin +3 -0
  33. checkpoint-4491/vocab.txt +0 -0
  34. checkpoint-5988/config.json +38 -0
  35. checkpoint-5988/model.safetensors +3 -0
  36. checkpoint-5988/optimizer.pt +3 -0
  37. checkpoint-5988/rng_state.pth +3 -0
  38. checkpoint-5988/scheduler.pt +3 -0
  39. checkpoint-5988/special_tokens_map.json +7 -0
  40. checkpoint-5988/tokenizer.json +0 -0
  41. checkpoint-5988/tokenizer_config.json +56 -0
  42. checkpoint-5988/trainer_state.json +907 -0
  43. checkpoint-5988/training_args.bin +3 -0
  44. checkpoint-5988/vocab.txt +0 -0
  45. config.json +38 -0
  46. model.safetensors +3 -0
  47. special_tokens_map.json +7 -0
  48. tokenizer.json +0 -0
  49. tokenizer_config.json +56 -0
  50. training_args.bin +3 -0
checkpoint-1497/config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ElectraForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "embedding_size": 128,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 256,
11
+ "id2label": {
12
+ "0": "NOT_HEADING",
13
+ "1": "HEADING"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 1024,
17
+ "label2id": {
18
+ "HEADING": 1,
19
+ "NOT_HEADING": 0
20
+ },
21
+ "layer_norm_eps": 1e-12,
22
+ "max_position_embeddings": 512,
23
+ "model_type": "electra",
24
+ "num_attention_heads": 4,
25
+ "num_hidden_layers": 12,
26
+ "pad_token_id": 0,
27
+ "position_embedding_type": "absolute",
28
+ "problem_type": "single_label_classification",
29
+ "summary_activation": "gelu",
30
+ "summary_last_dropout": 0.1,
31
+ "summary_type": "first",
32
+ "summary_use_proj": true,
33
+ "torch_dtype": "float32",
34
+ "transformers_version": "4.53.3",
35
+ "type_vocab_size": 2,
36
+ "use_cache": true,
37
+ "vocab_size": 30522
38
+ }
checkpoint-1497/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4767a96956fe33ceeaaa34e5787b460e43d9e9a0c11977738c5ab2821eacb8e0
3
+ size 54221200
checkpoint-1497/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e864071f48568d524d1abfcebea3ae2870db0c0c657ada32e8eb506b2af77931
3
+ size 108564747
checkpoint-1497/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9aae9c442391ce149aa949f1a5566e50dfa6f2cb167877dc19bbd772d752a704
3
+ size 14645
checkpoint-1497/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66f8e2f574f8264774ce03598e0126d9f33948e2724461dc6ea08b724841d869
3
+ size 1465
checkpoint-1497/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-1497/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1497/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": "ElectraTokenizer",
55
+ "unk_token": "[UNK]"
56
+ }
checkpoint-1497/trainer_state.json ADDED
@@ -0,0 +1,247 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1497,
3
+ "best_metric": 0.7813953488372093,
4
+ "best_model_checkpoint": "./electra-small-heading-classifier-expanded\\checkpoint-1497",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1497,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.033400133600534405,
14
+ "grad_norm": 1.82759428024292,
15
+ "learning_rate": 2.45e-05,
16
+ "loss": 0.5801,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.06680026720106881,
21
+ "grad_norm": 0.7336680293083191,
22
+ "learning_rate": 4.9500000000000004e-05,
23
+ "loss": 0.3191,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.10020040080160321,
28
+ "grad_norm": 0.759293258190155,
29
+ "learning_rate": 4.9583899456521745e-05,
30
+ "loss": 0.2216,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.13360053440213762,
35
+ "grad_norm": 0.5498169660568237,
36
+ "learning_rate": 4.915930706521739e-05,
37
+ "loss": 0.2167,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 0.16700066800267202,
42
+ "grad_norm": 3.572784662246704,
43
+ "learning_rate": 4.8734714673913044e-05,
44
+ "loss": 0.1797,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 0.20040080160320642,
49
+ "grad_norm": 1.3190865516662598,
50
+ "learning_rate": 4.83101222826087e-05,
51
+ "loss": 0.2202,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.23380093520374082,
56
+ "grad_norm": 1.5133466720581055,
57
+ "learning_rate": 4.788552989130435e-05,
58
+ "loss": 0.1925,
59
+ "step": 350
60
+ },
61
+ {
62
+ "epoch": 0.26720106880427524,
63
+ "grad_norm": 1.3208128213882446,
64
+ "learning_rate": 4.7460937500000004e-05,
65
+ "loss": 0.265,
66
+ "step": 400
67
+ },
68
+ {
69
+ "epoch": 0.30060120240480964,
70
+ "grad_norm": 0.9725052118301392,
71
+ "learning_rate": 4.703634510869566e-05,
72
+ "loss": 0.2866,
73
+ "step": 450
74
+ },
75
+ {
76
+ "epoch": 0.33400133600534404,
77
+ "grad_norm": 1.570516586303711,
78
+ "learning_rate": 4.6611752717391304e-05,
79
+ "loss": 0.1881,
80
+ "step": 500
81
+ },
82
+ {
83
+ "epoch": 0.36740146960587844,
84
+ "grad_norm": 4.067606449127197,
85
+ "learning_rate": 4.618716032608696e-05,
86
+ "loss": 0.2258,
87
+ "step": 550
88
+ },
89
+ {
90
+ "epoch": 0.40080160320641284,
91
+ "grad_norm": 0.2925869822502136,
92
+ "learning_rate": 4.576256793478261e-05,
93
+ "loss": 0.1488,
94
+ "step": 600
95
+ },
96
+ {
97
+ "epoch": 0.43420173680694724,
98
+ "grad_norm": 0.9354973435401917,
99
+ "learning_rate": 4.5337975543478264e-05,
100
+ "loss": 0.1611,
101
+ "step": 650
102
+ },
103
+ {
104
+ "epoch": 0.46760187040748163,
105
+ "grad_norm": 0.3285213112831116,
106
+ "learning_rate": 4.491338315217392e-05,
107
+ "loss": 0.1259,
108
+ "step": 700
109
+ },
110
+ {
111
+ "epoch": 0.501002004008016,
112
+ "grad_norm": 13.97227954864502,
113
+ "learning_rate": 4.448879076086957e-05,
114
+ "loss": 0.1593,
115
+ "step": 750
116
+ },
117
+ {
118
+ "epoch": 0.5344021376085505,
119
+ "grad_norm": 0.10469996929168701,
120
+ "learning_rate": 4.406419836956522e-05,
121
+ "loss": 0.0923,
122
+ "step": 800
123
+ },
124
+ {
125
+ "epoch": 0.5678022712090849,
126
+ "grad_norm": 0.19770751893520355,
127
+ "learning_rate": 4.363960597826087e-05,
128
+ "loss": 0.1067,
129
+ "step": 850
130
+ },
131
+ {
132
+ "epoch": 0.6012024048096193,
133
+ "grad_norm": 0.10104553401470184,
134
+ "learning_rate": 4.321501358695652e-05,
135
+ "loss": 0.1214,
136
+ "step": 900
137
+ },
138
+ {
139
+ "epoch": 0.6346025384101537,
140
+ "grad_norm": 0.10949808359146118,
141
+ "learning_rate": 4.279042119565218e-05,
142
+ "loss": 0.1175,
143
+ "step": 950
144
+ },
145
+ {
146
+ "epoch": 0.6680026720106881,
147
+ "grad_norm": 0.10336494445800781,
148
+ "learning_rate": 4.236582880434783e-05,
149
+ "loss": 0.0849,
150
+ "step": 1000
151
+ },
152
+ {
153
+ "epoch": 0.7014028056112225,
154
+ "grad_norm": 0.12163177132606506,
155
+ "learning_rate": 4.1941236413043476e-05,
156
+ "loss": 0.1486,
157
+ "step": 1050
158
+ },
159
+ {
160
+ "epoch": 0.7348029392117569,
161
+ "grad_norm": 0.0906616598367691,
162
+ "learning_rate": 4.151664402173913e-05,
163
+ "loss": 0.0671,
164
+ "step": 1100
165
+ },
166
+ {
167
+ "epoch": 0.7682030728122913,
168
+ "grad_norm": 0.12954840064048767,
169
+ "learning_rate": 4.109205163043478e-05,
170
+ "loss": 0.072,
171
+ "step": 1150
172
+ },
173
+ {
174
+ "epoch": 0.8016032064128257,
175
+ "grad_norm": 0.06860172003507614,
176
+ "learning_rate": 4.0667459239130436e-05,
177
+ "loss": 0.0735,
178
+ "step": 1200
179
+ },
180
+ {
181
+ "epoch": 0.8350033400133601,
182
+ "grad_norm": 0.06437909603118896,
183
+ "learning_rate": 4.024286684782609e-05,
184
+ "loss": 0.16,
185
+ "step": 1250
186
+ },
187
+ {
188
+ "epoch": 0.8684034736138945,
189
+ "grad_norm": 3.2551543712615967,
190
+ "learning_rate": 3.981827445652174e-05,
191
+ "loss": 0.0868,
192
+ "step": 1300
193
+ },
194
+ {
195
+ "epoch": 0.9018036072144289,
196
+ "grad_norm": 0.12183782458305359,
197
+ "learning_rate": 3.939368206521739e-05,
198
+ "loss": 0.1346,
199
+ "step": 1350
200
+ },
201
+ {
202
+ "epoch": 0.9352037408149633,
203
+ "grad_norm": 5.742549419403076,
204
+ "learning_rate": 3.896908967391304e-05,
205
+ "loss": 0.0977,
206
+ "step": 1400
207
+ },
208
+ {
209
+ "epoch": 0.9686038744154977,
210
+ "grad_norm": 0.049525078386068344,
211
+ "learning_rate": 3.85444972826087e-05,
212
+ "loss": 0.0771,
213
+ "step": 1450
214
+ },
215
+ {
216
+ "epoch": 1.0,
217
+ "eval_accuracy": 0.9777567439659253,
218
+ "eval_f1": 0.7813953488372093,
219
+ "eval_loss": 0.07437803596258163,
220
+ "eval_runtime": 8.1317,
221
+ "eval_samples_per_second": 259.846,
222
+ "eval_steps_per_second": 32.588,
223
+ "step": 1497
224
+ }
225
+ ],
226
+ "logging_steps": 50,
227
+ "max_steps": 5988,
228
+ "num_input_tokens_seen": 0,
229
+ "num_train_epochs": 4,
230
+ "save_steps": 500,
231
+ "stateful_callbacks": {
232
+ "TrainerControl": {
233
+ "args": {
234
+ "should_epoch_stop": false,
235
+ "should_evaluate": false,
236
+ "should_log": false,
237
+ "should_save": true,
238
+ "should_training_stop": false
239
+ },
240
+ "attributes": {}
241
+ }
242
+ },
243
+ "total_flos": 352211940040704.0,
244
+ "train_batch_size": 8,
245
+ "trial_name": null,
246
+ "trial_params": null
247
+ }
checkpoint-1497/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e03a8046e1b1da0fe3b9b933f56a9589c401ad31953c08b2477ec29e5fff7bf3
3
+ size 5713
checkpoint-1497/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2994/config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ElectraForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "embedding_size": 128,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 256,
11
+ "id2label": {
12
+ "0": "NOT_HEADING",
13
+ "1": "HEADING"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 1024,
17
+ "label2id": {
18
+ "HEADING": 1,
19
+ "NOT_HEADING": 0
20
+ },
21
+ "layer_norm_eps": 1e-12,
22
+ "max_position_embeddings": 512,
23
+ "model_type": "electra",
24
+ "num_attention_heads": 4,
25
+ "num_hidden_layers": 12,
26
+ "pad_token_id": 0,
27
+ "position_embedding_type": "absolute",
28
+ "problem_type": "single_label_classification",
29
+ "summary_activation": "gelu",
30
+ "summary_last_dropout": 0.1,
31
+ "summary_type": "first",
32
+ "summary_use_proj": true,
33
+ "torch_dtype": "float32",
34
+ "transformers_version": "4.53.3",
35
+ "type_vocab_size": 2,
36
+ "use_cache": true,
37
+ "vocab_size": 30522
38
+ }
checkpoint-2994/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e77553afca2eb6ed95d401770181575e774c5941429a5412980245ff9838892c
3
+ size 54221200
checkpoint-2994/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc4bf55c883a466be2d90fbf2168909be6d90d354df00e00ff6124da38621006
3
+ size 108564747
checkpoint-2994/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01c46fdd2648600b857e569143f0ea96e1de1e077f9820d414bbfad28756d67c
3
+ size 14645
checkpoint-2994/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fa9a19dc96de60544a5645667b48e08ad09100f86dd469759f58aa33bfbd4e6
3
+ size 1465
checkpoint-2994/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-2994/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2994/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": "ElectraTokenizer",
55
+ "unk_token": "[UNK]"
56
+ }
checkpoint-2994/trainer_state.json ADDED
@@ -0,0 +1,467 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2994,
3
+ "best_metric": 0.8489208633093526,
4
+ "best_model_checkpoint": "./electra-small-heading-classifier-expanded\\checkpoint-2994",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2994,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.033400133600534405,
14
+ "grad_norm": 1.82759428024292,
15
+ "learning_rate": 2.45e-05,
16
+ "loss": 0.5801,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.06680026720106881,
21
+ "grad_norm": 0.7336680293083191,
22
+ "learning_rate": 4.9500000000000004e-05,
23
+ "loss": 0.3191,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.10020040080160321,
28
+ "grad_norm": 0.759293258190155,
29
+ "learning_rate": 4.9583899456521745e-05,
30
+ "loss": 0.2216,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.13360053440213762,
35
+ "grad_norm": 0.5498169660568237,
36
+ "learning_rate": 4.915930706521739e-05,
37
+ "loss": 0.2167,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 0.16700066800267202,
42
+ "grad_norm": 3.572784662246704,
43
+ "learning_rate": 4.8734714673913044e-05,
44
+ "loss": 0.1797,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 0.20040080160320642,
49
+ "grad_norm": 1.3190865516662598,
50
+ "learning_rate": 4.83101222826087e-05,
51
+ "loss": 0.2202,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.23380093520374082,
56
+ "grad_norm": 1.5133466720581055,
57
+ "learning_rate": 4.788552989130435e-05,
58
+ "loss": 0.1925,
59
+ "step": 350
60
+ },
61
+ {
62
+ "epoch": 0.26720106880427524,
63
+ "grad_norm": 1.3208128213882446,
64
+ "learning_rate": 4.7460937500000004e-05,
65
+ "loss": 0.265,
66
+ "step": 400
67
+ },
68
+ {
69
+ "epoch": 0.30060120240480964,
70
+ "grad_norm": 0.9725052118301392,
71
+ "learning_rate": 4.703634510869566e-05,
72
+ "loss": 0.2866,
73
+ "step": 450
74
+ },
75
+ {
76
+ "epoch": 0.33400133600534404,
77
+ "grad_norm": 1.570516586303711,
78
+ "learning_rate": 4.6611752717391304e-05,
79
+ "loss": 0.1881,
80
+ "step": 500
81
+ },
82
+ {
83
+ "epoch": 0.36740146960587844,
84
+ "grad_norm": 4.067606449127197,
85
+ "learning_rate": 4.618716032608696e-05,
86
+ "loss": 0.2258,
87
+ "step": 550
88
+ },
89
+ {
90
+ "epoch": 0.40080160320641284,
91
+ "grad_norm": 0.2925869822502136,
92
+ "learning_rate": 4.576256793478261e-05,
93
+ "loss": 0.1488,
94
+ "step": 600
95
+ },
96
+ {
97
+ "epoch": 0.43420173680694724,
98
+ "grad_norm": 0.9354973435401917,
99
+ "learning_rate": 4.5337975543478264e-05,
100
+ "loss": 0.1611,
101
+ "step": 650
102
+ },
103
+ {
104
+ "epoch": 0.46760187040748163,
105
+ "grad_norm": 0.3285213112831116,
106
+ "learning_rate": 4.491338315217392e-05,
107
+ "loss": 0.1259,
108
+ "step": 700
109
+ },
110
+ {
111
+ "epoch": 0.501002004008016,
112
+ "grad_norm": 13.97227954864502,
113
+ "learning_rate": 4.448879076086957e-05,
114
+ "loss": 0.1593,
115
+ "step": 750
116
+ },
117
+ {
118
+ "epoch": 0.5344021376085505,
119
+ "grad_norm": 0.10469996929168701,
120
+ "learning_rate": 4.406419836956522e-05,
121
+ "loss": 0.0923,
122
+ "step": 800
123
+ },
124
+ {
125
+ "epoch": 0.5678022712090849,
126
+ "grad_norm": 0.19770751893520355,
127
+ "learning_rate": 4.363960597826087e-05,
128
+ "loss": 0.1067,
129
+ "step": 850
130
+ },
131
+ {
132
+ "epoch": 0.6012024048096193,
133
+ "grad_norm": 0.10104553401470184,
134
+ "learning_rate": 4.321501358695652e-05,
135
+ "loss": 0.1214,
136
+ "step": 900
137
+ },
138
+ {
139
+ "epoch": 0.6346025384101537,
140
+ "grad_norm": 0.10949808359146118,
141
+ "learning_rate": 4.279042119565218e-05,
142
+ "loss": 0.1175,
143
+ "step": 950
144
+ },
145
+ {
146
+ "epoch": 0.6680026720106881,
147
+ "grad_norm": 0.10336494445800781,
148
+ "learning_rate": 4.236582880434783e-05,
149
+ "loss": 0.0849,
150
+ "step": 1000
151
+ },
152
+ {
153
+ "epoch": 0.7014028056112225,
154
+ "grad_norm": 0.12163177132606506,
155
+ "learning_rate": 4.1941236413043476e-05,
156
+ "loss": 0.1486,
157
+ "step": 1050
158
+ },
159
+ {
160
+ "epoch": 0.7348029392117569,
161
+ "grad_norm": 0.0906616598367691,
162
+ "learning_rate": 4.151664402173913e-05,
163
+ "loss": 0.0671,
164
+ "step": 1100
165
+ },
166
+ {
167
+ "epoch": 0.7682030728122913,
168
+ "grad_norm": 0.12954840064048767,
169
+ "learning_rate": 4.109205163043478e-05,
170
+ "loss": 0.072,
171
+ "step": 1150
172
+ },
173
+ {
174
+ "epoch": 0.8016032064128257,
175
+ "grad_norm": 0.06860172003507614,
176
+ "learning_rate": 4.0667459239130436e-05,
177
+ "loss": 0.0735,
178
+ "step": 1200
179
+ },
180
+ {
181
+ "epoch": 0.8350033400133601,
182
+ "grad_norm": 0.06437909603118896,
183
+ "learning_rate": 4.024286684782609e-05,
184
+ "loss": 0.16,
185
+ "step": 1250
186
+ },
187
+ {
188
+ "epoch": 0.8684034736138945,
189
+ "grad_norm": 3.2551543712615967,
190
+ "learning_rate": 3.981827445652174e-05,
191
+ "loss": 0.0868,
192
+ "step": 1300
193
+ },
194
+ {
195
+ "epoch": 0.9018036072144289,
196
+ "grad_norm": 0.12183782458305359,
197
+ "learning_rate": 3.939368206521739e-05,
198
+ "loss": 0.1346,
199
+ "step": 1350
200
+ },
201
+ {
202
+ "epoch": 0.9352037408149633,
203
+ "grad_norm": 5.742549419403076,
204
+ "learning_rate": 3.896908967391304e-05,
205
+ "loss": 0.0977,
206
+ "step": 1400
207
+ },
208
+ {
209
+ "epoch": 0.9686038744154977,
210
+ "grad_norm": 0.049525078386068344,
211
+ "learning_rate": 3.85444972826087e-05,
212
+ "loss": 0.0771,
213
+ "step": 1450
214
+ },
215
+ {
216
+ "epoch": 1.0,
217
+ "eval_accuracy": 0.9777567439659253,
218
+ "eval_f1": 0.7813953488372093,
219
+ "eval_loss": 0.07437803596258163,
220
+ "eval_runtime": 8.1317,
221
+ "eval_samples_per_second": 259.846,
222
+ "eval_steps_per_second": 32.588,
223
+ "step": 1497
224
+ },
225
+ {
226
+ "epoch": 1.002004008016032,
227
+ "grad_norm": 0.08354274183511734,
228
+ "learning_rate": 3.811990489130435e-05,
229
+ "loss": 0.1122,
230
+ "step": 1500
231
+ },
232
+ {
233
+ "epoch": 1.0354041416165665,
234
+ "grad_norm": 0.07657646387815475,
235
+ "learning_rate": 3.76953125e-05,
236
+ "loss": 0.0776,
237
+ "step": 1550
238
+ },
239
+ {
240
+ "epoch": 1.0688042752171008,
241
+ "grad_norm": 0.029183173552155495,
242
+ "learning_rate": 3.7270720108695656e-05,
243
+ "loss": 0.048,
244
+ "step": 1600
245
+ },
246
+ {
247
+ "epoch": 1.1022044088176353,
248
+ "grad_norm": 0.6650001406669617,
249
+ "learning_rate": 3.68461277173913e-05,
250
+ "loss": 0.0651,
251
+ "step": 1650
252
+ },
253
+ {
254
+ "epoch": 1.1356045424181698,
255
+ "grad_norm": 0.13291126489639282,
256
+ "learning_rate": 3.6421535326086955e-05,
257
+ "loss": 0.1035,
258
+ "step": 1700
259
+ },
260
+ {
261
+ "epoch": 1.169004676018704,
262
+ "grad_norm": 0.14512984454631805,
263
+ "learning_rate": 3.5996942934782615e-05,
264
+ "loss": 0.0896,
265
+ "step": 1750
266
+ },
267
+ {
268
+ "epoch": 1.2024048096192386,
269
+ "grad_norm": 0.09467575699090958,
270
+ "learning_rate": 3.557235054347826e-05,
271
+ "loss": 0.0763,
272
+ "step": 1800
273
+ },
274
+ {
275
+ "epoch": 1.2358049432197729,
276
+ "grad_norm": 0.5623799562454224,
277
+ "learning_rate": 3.5147758152173915e-05,
278
+ "loss": 0.059,
279
+ "step": 1850
280
+ },
281
+ {
282
+ "epoch": 1.2692050768203074,
283
+ "grad_norm": 0.16416147351264954,
284
+ "learning_rate": 3.472316576086957e-05,
285
+ "loss": 0.0483,
286
+ "step": 1900
287
+ },
288
+ {
289
+ "epoch": 1.3026052104208417,
290
+ "grad_norm": 3.0964038372039795,
291
+ "learning_rate": 3.4298573369565215e-05,
292
+ "loss": 0.1058,
293
+ "step": 1950
294
+ },
295
+ {
296
+ "epoch": 1.3360053440213762,
297
+ "grad_norm": 0.030443737283349037,
298
+ "learning_rate": 3.3873980978260875e-05,
299
+ "loss": 0.0517,
300
+ "step": 2000
301
+ },
302
+ {
303
+ "epoch": 1.3694054776219104,
304
+ "grad_norm": 0.04263285547494888,
305
+ "learning_rate": 3.344938858695653e-05,
306
+ "loss": 0.0486,
307
+ "step": 2050
308
+ },
309
+ {
310
+ "epoch": 1.402805611222445,
311
+ "grad_norm": 0.06713565438985825,
312
+ "learning_rate": 3.3024796195652175e-05,
313
+ "loss": 0.0494,
314
+ "step": 2100
315
+ },
316
+ {
317
+ "epoch": 1.4362057448229792,
318
+ "grad_norm": 0.04323391616344452,
319
+ "learning_rate": 3.260020380434783e-05,
320
+ "loss": 0.0518,
321
+ "step": 2150
322
+ },
323
+ {
324
+ "epoch": 1.4696058784235138,
325
+ "grad_norm": 0.9192191362380981,
326
+ "learning_rate": 3.2175611413043474e-05,
327
+ "loss": 0.0498,
328
+ "step": 2200
329
+ },
330
+ {
331
+ "epoch": 1.503006012024048,
332
+ "grad_norm": 0.08719488233327866,
333
+ "learning_rate": 3.1751019021739135e-05,
334
+ "loss": 0.1155,
335
+ "step": 2250
336
+ },
337
+ {
338
+ "epoch": 1.5364061456245826,
339
+ "grad_norm": 6.060549259185791,
340
+ "learning_rate": 3.132642663043479e-05,
341
+ "loss": 0.1349,
342
+ "step": 2300
343
+ },
344
+ {
345
+ "epoch": 1.569806279225117,
346
+ "grad_norm": 0.09453130513429642,
347
+ "learning_rate": 3.0901834239130434e-05,
348
+ "loss": 0.0639,
349
+ "step": 2350
350
+ },
351
+ {
352
+ "epoch": 1.6032064128256514,
353
+ "grad_norm": 0.02592223510146141,
354
+ "learning_rate": 3.0477241847826088e-05,
355
+ "loss": 0.0929,
356
+ "step": 2400
357
+ },
358
+ {
359
+ "epoch": 1.6366065464261856,
360
+ "grad_norm": 0.027360519394278526,
361
+ "learning_rate": 3.005264945652174e-05,
362
+ "loss": 0.071,
363
+ "step": 2450
364
+ },
365
+ {
366
+ "epoch": 1.6700066800267201,
367
+ "grad_norm": 0.14285916090011597,
368
+ "learning_rate": 2.962805706521739e-05,
369
+ "loss": 0.0768,
370
+ "step": 2500
371
+ },
372
+ {
373
+ "epoch": 1.7034068136272547,
374
+ "grad_norm": 0.04874153807759285,
375
+ "learning_rate": 2.9203464673913044e-05,
376
+ "loss": 0.076,
377
+ "step": 2550
378
+ },
379
+ {
380
+ "epoch": 1.736806947227789,
381
+ "grad_norm": 0.09328486025333405,
382
+ "learning_rate": 2.8778872282608697e-05,
383
+ "loss": 0.0505,
384
+ "step": 2600
385
+ },
386
+ {
387
+ "epoch": 1.7702070808283232,
388
+ "grad_norm": 0.34670591354370117,
389
+ "learning_rate": 2.8354279891304347e-05,
390
+ "loss": 0.0763,
391
+ "step": 2650
392
+ },
393
+ {
394
+ "epoch": 1.8036072144288577,
395
+ "grad_norm": 0.057789236307144165,
396
+ "learning_rate": 2.79296875e-05,
397
+ "loss": 0.0832,
398
+ "step": 2700
399
+ },
400
+ {
401
+ "epoch": 1.8370073480293923,
402
+ "grad_norm": 0.059989944100379944,
403
+ "learning_rate": 2.7505095108695657e-05,
404
+ "loss": 0.062,
405
+ "step": 2750
406
+ },
407
+ {
408
+ "epoch": 1.8704074816299265,
409
+ "grad_norm": 0.02221057377755642,
410
+ "learning_rate": 2.7080502717391304e-05,
411
+ "loss": 0.0649,
412
+ "step": 2800
413
+ },
414
+ {
415
+ "epoch": 1.9038076152304608,
416
+ "grad_norm": 0.154799684882164,
417
+ "learning_rate": 2.6655910326086957e-05,
418
+ "loss": 0.144,
419
+ "step": 2850
420
+ },
421
+ {
422
+ "epoch": 1.9372077488309953,
423
+ "grad_norm": 0.0422816276550293,
424
+ "learning_rate": 2.6231317934782613e-05,
425
+ "loss": 0.0531,
426
+ "step": 2900
427
+ },
428
+ {
429
+ "epoch": 1.9706078824315298,
430
+ "grad_norm": 0.019839206710457802,
431
+ "learning_rate": 2.580672554347826e-05,
432
+ "loss": 0.0545,
433
+ "step": 2950
434
+ },
435
+ {
436
+ "epoch": 2.0,
437
+ "eval_accuracy": 0.9801230477993375,
438
+ "eval_f1": 0.8489208633093526,
439
+ "eval_loss": 0.07019170373678207,
440
+ "eval_runtime": 8.2123,
441
+ "eval_samples_per_second": 257.298,
442
+ "eval_steps_per_second": 32.269,
443
+ "step": 2994
444
+ }
445
+ ],
446
+ "logging_steps": 50,
447
+ "max_steps": 5988,
448
+ "num_input_tokens_seen": 0,
449
+ "num_train_epochs": 4,
450
+ "save_steps": 500,
451
+ "stateful_callbacks": {
452
+ "TrainerControl": {
453
+ "args": {
454
+ "should_epoch_stop": false,
455
+ "should_evaluate": false,
456
+ "should_log": false,
457
+ "should_save": true,
458
+ "should_training_stop": false
459
+ },
460
+ "attributes": {}
461
+ }
462
+ },
463
+ "total_flos": 704423880081408.0,
464
+ "train_batch_size": 8,
465
+ "trial_name": null,
466
+ "trial_params": null
467
+ }
checkpoint-2994/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e03a8046e1b1da0fe3b9b933f56a9589c401ad31953c08b2477ec29e5fff7bf3
3
+ size 5713
checkpoint-2994/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-4491/config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ElectraForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "embedding_size": 128,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 256,
11
+ "id2label": {
12
+ "0": "NOT_HEADING",
13
+ "1": "HEADING"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 1024,
17
+ "label2id": {
18
+ "HEADING": 1,
19
+ "NOT_HEADING": 0
20
+ },
21
+ "layer_norm_eps": 1e-12,
22
+ "max_position_embeddings": 512,
23
+ "model_type": "electra",
24
+ "num_attention_heads": 4,
25
+ "num_hidden_layers": 12,
26
+ "pad_token_id": 0,
27
+ "position_embedding_type": "absolute",
28
+ "problem_type": "single_label_classification",
29
+ "summary_activation": "gelu",
30
+ "summary_last_dropout": 0.1,
31
+ "summary_type": "first",
32
+ "summary_use_proj": true,
33
+ "torch_dtype": "float32",
34
+ "transformers_version": "4.53.3",
35
+ "type_vocab_size": 2,
36
+ "use_cache": true,
37
+ "vocab_size": 30522
38
+ }
checkpoint-4491/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f98cd1586be43f601de6319be13b04c8112e35abda46dd6f4e9a65839a17f5d9
3
+ size 54221200
checkpoint-4491/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c00661ed4714152fdf0c7fd5f8bb51c74247625d587a82d1ee267b7ffc41bf8
3
+ size 108564747
checkpoint-4491/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6bb93161bdb50e478e1b83319179fad4631e71552288f10d1b232e24fedf8a1d
3
+ size 14645
checkpoint-4491/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4eca8fe981e4a7c3d38bc270169824f344b57d264984fb621e6591014bef2f3
3
+ size 1465
checkpoint-4491/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-4491/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-4491/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": "ElectraTokenizer",
55
+ "unk_token": "[UNK]"
56
+ }
checkpoint-4491/trainer_state.json ADDED
@@ -0,0 +1,687 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 4491,
3
+ "best_metric": 0.888030888030888,
4
+ "best_model_checkpoint": "./electra-small-heading-classifier-expanded\\checkpoint-4491",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 4491,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.033400133600534405,
14
+ "grad_norm": 1.82759428024292,
15
+ "learning_rate": 2.45e-05,
16
+ "loss": 0.5801,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.06680026720106881,
21
+ "grad_norm": 0.7336680293083191,
22
+ "learning_rate": 4.9500000000000004e-05,
23
+ "loss": 0.3191,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.10020040080160321,
28
+ "grad_norm": 0.759293258190155,
29
+ "learning_rate": 4.9583899456521745e-05,
30
+ "loss": 0.2216,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.13360053440213762,
35
+ "grad_norm": 0.5498169660568237,
36
+ "learning_rate": 4.915930706521739e-05,
37
+ "loss": 0.2167,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 0.16700066800267202,
42
+ "grad_norm": 3.572784662246704,
43
+ "learning_rate": 4.8734714673913044e-05,
44
+ "loss": 0.1797,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 0.20040080160320642,
49
+ "grad_norm": 1.3190865516662598,
50
+ "learning_rate": 4.83101222826087e-05,
51
+ "loss": 0.2202,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.23380093520374082,
56
+ "grad_norm": 1.5133466720581055,
57
+ "learning_rate": 4.788552989130435e-05,
58
+ "loss": 0.1925,
59
+ "step": 350
60
+ },
61
+ {
62
+ "epoch": 0.26720106880427524,
63
+ "grad_norm": 1.3208128213882446,
64
+ "learning_rate": 4.7460937500000004e-05,
65
+ "loss": 0.265,
66
+ "step": 400
67
+ },
68
+ {
69
+ "epoch": 0.30060120240480964,
70
+ "grad_norm": 0.9725052118301392,
71
+ "learning_rate": 4.703634510869566e-05,
72
+ "loss": 0.2866,
73
+ "step": 450
74
+ },
75
+ {
76
+ "epoch": 0.33400133600534404,
77
+ "grad_norm": 1.570516586303711,
78
+ "learning_rate": 4.6611752717391304e-05,
79
+ "loss": 0.1881,
80
+ "step": 500
81
+ },
82
+ {
83
+ "epoch": 0.36740146960587844,
84
+ "grad_norm": 4.067606449127197,
85
+ "learning_rate": 4.618716032608696e-05,
86
+ "loss": 0.2258,
87
+ "step": 550
88
+ },
89
+ {
90
+ "epoch": 0.40080160320641284,
91
+ "grad_norm": 0.2925869822502136,
92
+ "learning_rate": 4.576256793478261e-05,
93
+ "loss": 0.1488,
94
+ "step": 600
95
+ },
96
+ {
97
+ "epoch": 0.43420173680694724,
98
+ "grad_norm": 0.9354973435401917,
99
+ "learning_rate": 4.5337975543478264e-05,
100
+ "loss": 0.1611,
101
+ "step": 650
102
+ },
103
+ {
104
+ "epoch": 0.46760187040748163,
105
+ "grad_norm": 0.3285213112831116,
106
+ "learning_rate": 4.491338315217392e-05,
107
+ "loss": 0.1259,
108
+ "step": 700
109
+ },
110
+ {
111
+ "epoch": 0.501002004008016,
112
+ "grad_norm": 13.97227954864502,
113
+ "learning_rate": 4.448879076086957e-05,
114
+ "loss": 0.1593,
115
+ "step": 750
116
+ },
117
+ {
118
+ "epoch": 0.5344021376085505,
119
+ "grad_norm": 0.10469996929168701,
120
+ "learning_rate": 4.406419836956522e-05,
121
+ "loss": 0.0923,
122
+ "step": 800
123
+ },
124
+ {
125
+ "epoch": 0.5678022712090849,
126
+ "grad_norm": 0.19770751893520355,
127
+ "learning_rate": 4.363960597826087e-05,
128
+ "loss": 0.1067,
129
+ "step": 850
130
+ },
131
+ {
132
+ "epoch": 0.6012024048096193,
133
+ "grad_norm": 0.10104553401470184,
134
+ "learning_rate": 4.321501358695652e-05,
135
+ "loss": 0.1214,
136
+ "step": 900
137
+ },
138
+ {
139
+ "epoch": 0.6346025384101537,
140
+ "grad_norm": 0.10949808359146118,
141
+ "learning_rate": 4.279042119565218e-05,
142
+ "loss": 0.1175,
143
+ "step": 950
144
+ },
145
+ {
146
+ "epoch": 0.6680026720106881,
147
+ "grad_norm": 0.10336494445800781,
148
+ "learning_rate": 4.236582880434783e-05,
149
+ "loss": 0.0849,
150
+ "step": 1000
151
+ },
152
+ {
153
+ "epoch": 0.7014028056112225,
154
+ "grad_norm": 0.12163177132606506,
155
+ "learning_rate": 4.1941236413043476e-05,
156
+ "loss": 0.1486,
157
+ "step": 1050
158
+ },
159
+ {
160
+ "epoch": 0.7348029392117569,
161
+ "grad_norm": 0.0906616598367691,
162
+ "learning_rate": 4.151664402173913e-05,
163
+ "loss": 0.0671,
164
+ "step": 1100
165
+ },
166
+ {
167
+ "epoch": 0.7682030728122913,
168
+ "grad_norm": 0.12954840064048767,
169
+ "learning_rate": 4.109205163043478e-05,
170
+ "loss": 0.072,
171
+ "step": 1150
172
+ },
173
+ {
174
+ "epoch": 0.8016032064128257,
175
+ "grad_norm": 0.06860172003507614,
176
+ "learning_rate": 4.0667459239130436e-05,
177
+ "loss": 0.0735,
178
+ "step": 1200
179
+ },
180
+ {
181
+ "epoch": 0.8350033400133601,
182
+ "grad_norm": 0.06437909603118896,
183
+ "learning_rate": 4.024286684782609e-05,
184
+ "loss": 0.16,
185
+ "step": 1250
186
+ },
187
+ {
188
+ "epoch": 0.8684034736138945,
189
+ "grad_norm": 3.2551543712615967,
190
+ "learning_rate": 3.981827445652174e-05,
191
+ "loss": 0.0868,
192
+ "step": 1300
193
+ },
194
+ {
195
+ "epoch": 0.9018036072144289,
196
+ "grad_norm": 0.12183782458305359,
197
+ "learning_rate": 3.939368206521739e-05,
198
+ "loss": 0.1346,
199
+ "step": 1350
200
+ },
201
+ {
202
+ "epoch": 0.9352037408149633,
203
+ "grad_norm": 5.742549419403076,
204
+ "learning_rate": 3.896908967391304e-05,
205
+ "loss": 0.0977,
206
+ "step": 1400
207
+ },
208
+ {
209
+ "epoch": 0.9686038744154977,
210
+ "grad_norm": 0.049525078386068344,
211
+ "learning_rate": 3.85444972826087e-05,
212
+ "loss": 0.0771,
213
+ "step": 1450
214
+ },
215
+ {
216
+ "epoch": 1.0,
217
+ "eval_accuracy": 0.9777567439659253,
218
+ "eval_f1": 0.7813953488372093,
219
+ "eval_loss": 0.07437803596258163,
220
+ "eval_runtime": 8.1317,
221
+ "eval_samples_per_second": 259.846,
222
+ "eval_steps_per_second": 32.588,
223
+ "step": 1497
224
+ },
225
+ {
226
+ "epoch": 1.002004008016032,
227
+ "grad_norm": 0.08354274183511734,
228
+ "learning_rate": 3.811990489130435e-05,
229
+ "loss": 0.1122,
230
+ "step": 1500
231
+ },
232
+ {
233
+ "epoch": 1.0354041416165665,
234
+ "grad_norm": 0.07657646387815475,
235
+ "learning_rate": 3.76953125e-05,
236
+ "loss": 0.0776,
237
+ "step": 1550
238
+ },
239
+ {
240
+ "epoch": 1.0688042752171008,
241
+ "grad_norm": 0.029183173552155495,
242
+ "learning_rate": 3.7270720108695656e-05,
243
+ "loss": 0.048,
244
+ "step": 1600
245
+ },
246
+ {
247
+ "epoch": 1.1022044088176353,
248
+ "grad_norm": 0.6650001406669617,
249
+ "learning_rate": 3.68461277173913e-05,
250
+ "loss": 0.0651,
251
+ "step": 1650
252
+ },
253
+ {
254
+ "epoch": 1.1356045424181698,
255
+ "grad_norm": 0.13291126489639282,
256
+ "learning_rate": 3.6421535326086955e-05,
257
+ "loss": 0.1035,
258
+ "step": 1700
259
+ },
260
+ {
261
+ "epoch": 1.169004676018704,
262
+ "grad_norm": 0.14512984454631805,
263
+ "learning_rate": 3.5996942934782615e-05,
264
+ "loss": 0.0896,
265
+ "step": 1750
266
+ },
267
+ {
268
+ "epoch": 1.2024048096192386,
269
+ "grad_norm": 0.09467575699090958,
270
+ "learning_rate": 3.557235054347826e-05,
271
+ "loss": 0.0763,
272
+ "step": 1800
273
+ },
274
+ {
275
+ "epoch": 1.2358049432197729,
276
+ "grad_norm": 0.5623799562454224,
277
+ "learning_rate": 3.5147758152173915e-05,
278
+ "loss": 0.059,
279
+ "step": 1850
280
+ },
281
+ {
282
+ "epoch": 1.2692050768203074,
283
+ "grad_norm": 0.16416147351264954,
284
+ "learning_rate": 3.472316576086957e-05,
285
+ "loss": 0.0483,
286
+ "step": 1900
287
+ },
288
+ {
289
+ "epoch": 1.3026052104208417,
290
+ "grad_norm": 3.0964038372039795,
291
+ "learning_rate": 3.4298573369565215e-05,
292
+ "loss": 0.1058,
293
+ "step": 1950
294
+ },
295
+ {
296
+ "epoch": 1.3360053440213762,
297
+ "grad_norm": 0.030443737283349037,
298
+ "learning_rate": 3.3873980978260875e-05,
299
+ "loss": 0.0517,
300
+ "step": 2000
301
+ },
302
+ {
303
+ "epoch": 1.3694054776219104,
304
+ "grad_norm": 0.04263285547494888,
305
+ "learning_rate": 3.344938858695653e-05,
306
+ "loss": 0.0486,
307
+ "step": 2050
308
+ },
309
+ {
310
+ "epoch": 1.402805611222445,
311
+ "grad_norm": 0.06713565438985825,
312
+ "learning_rate": 3.3024796195652175e-05,
313
+ "loss": 0.0494,
314
+ "step": 2100
315
+ },
316
+ {
317
+ "epoch": 1.4362057448229792,
318
+ "grad_norm": 0.04323391616344452,
319
+ "learning_rate": 3.260020380434783e-05,
320
+ "loss": 0.0518,
321
+ "step": 2150
322
+ },
323
+ {
324
+ "epoch": 1.4696058784235138,
325
+ "grad_norm": 0.9192191362380981,
326
+ "learning_rate": 3.2175611413043474e-05,
327
+ "loss": 0.0498,
328
+ "step": 2200
329
+ },
330
+ {
331
+ "epoch": 1.503006012024048,
332
+ "grad_norm": 0.08719488233327866,
333
+ "learning_rate": 3.1751019021739135e-05,
334
+ "loss": 0.1155,
335
+ "step": 2250
336
+ },
337
+ {
338
+ "epoch": 1.5364061456245826,
339
+ "grad_norm": 6.060549259185791,
340
+ "learning_rate": 3.132642663043479e-05,
341
+ "loss": 0.1349,
342
+ "step": 2300
343
+ },
344
+ {
345
+ "epoch": 1.569806279225117,
346
+ "grad_norm": 0.09453130513429642,
347
+ "learning_rate": 3.0901834239130434e-05,
348
+ "loss": 0.0639,
349
+ "step": 2350
350
+ },
351
+ {
352
+ "epoch": 1.6032064128256514,
353
+ "grad_norm": 0.02592223510146141,
354
+ "learning_rate": 3.0477241847826088e-05,
355
+ "loss": 0.0929,
356
+ "step": 2400
357
+ },
358
+ {
359
+ "epoch": 1.6366065464261856,
360
+ "grad_norm": 0.027360519394278526,
361
+ "learning_rate": 3.005264945652174e-05,
362
+ "loss": 0.071,
363
+ "step": 2450
364
+ },
365
+ {
366
+ "epoch": 1.6700066800267201,
367
+ "grad_norm": 0.14285916090011597,
368
+ "learning_rate": 2.962805706521739e-05,
369
+ "loss": 0.0768,
370
+ "step": 2500
371
+ },
372
+ {
373
+ "epoch": 1.7034068136272547,
374
+ "grad_norm": 0.04874153807759285,
375
+ "learning_rate": 2.9203464673913044e-05,
376
+ "loss": 0.076,
377
+ "step": 2550
378
+ },
379
+ {
380
+ "epoch": 1.736806947227789,
381
+ "grad_norm": 0.09328486025333405,
382
+ "learning_rate": 2.8778872282608697e-05,
383
+ "loss": 0.0505,
384
+ "step": 2600
385
+ },
386
+ {
387
+ "epoch": 1.7702070808283232,
388
+ "grad_norm": 0.34670591354370117,
389
+ "learning_rate": 2.8354279891304347e-05,
390
+ "loss": 0.0763,
391
+ "step": 2650
392
+ },
393
+ {
394
+ "epoch": 1.8036072144288577,
395
+ "grad_norm": 0.057789236307144165,
396
+ "learning_rate": 2.79296875e-05,
397
+ "loss": 0.0832,
398
+ "step": 2700
399
+ },
400
+ {
401
+ "epoch": 1.8370073480293923,
402
+ "grad_norm": 0.059989944100379944,
403
+ "learning_rate": 2.7505095108695657e-05,
404
+ "loss": 0.062,
405
+ "step": 2750
406
+ },
407
+ {
408
+ "epoch": 1.8704074816299265,
409
+ "grad_norm": 0.02221057377755642,
410
+ "learning_rate": 2.7080502717391304e-05,
411
+ "loss": 0.0649,
412
+ "step": 2800
413
+ },
414
+ {
415
+ "epoch": 1.9038076152304608,
416
+ "grad_norm": 0.154799684882164,
417
+ "learning_rate": 2.6655910326086957e-05,
418
+ "loss": 0.144,
419
+ "step": 2850
420
+ },
421
+ {
422
+ "epoch": 1.9372077488309953,
423
+ "grad_norm": 0.0422816276550293,
424
+ "learning_rate": 2.6231317934782613e-05,
425
+ "loss": 0.0531,
426
+ "step": 2900
427
+ },
428
+ {
429
+ "epoch": 1.9706078824315298,
430
+ "grad_norm": 0.019839206710457802,
431
+ "learning_rate": 2.580672554347826e-05,
432
+ "loss": 0.0545,
433
+ "step": 2950
434
+ },
435
+ {
436
+ "epoch": 2.0,
437
+ "eval_accuracy": 0.9801230477993375,
438
+ "eval_f1": 0.8489208633093526,
439
+ "eval_loss": 0.07019170373678207,
440
+ "eval_runtime": 8.2123,
441
+ "eval_samples_per_second": 257.298,
442
+ "eval_steps_per_second": 32.269,
443
+ "step": 2994
444
+ },
445
+ {
446
+ "epoch": 2.004008016032064,
447
+ "grad_norm": 3.3859903812408447,
448
+ "learning_rate": 2.5382133152173913e-05,
449
+ "loss": 0.0891,
450
+ "step": 3000
451
+ },
452
+ {
453
+ "epoch": 2.0374081496325984,
454
+ "grad_norm": 0.055893734097480774,
455
+ "learning_rate": 2.4957540760869567e-05,
456
+ "loss": 0.0498,
457
+ "step": 3050
458
+ },
459
+ {
460
+ "epoch": 2.070808283233133,
461
+ "grad_norm": 0.030930357053875923,
462
+ "learning_rate": 2.4532948369565216e-05,
463
+ "loss": 0.0272,
464
+ "step": 3100
465
+ },
466
+ {
467
+ "epoch": 2.1042084168336674,
468
+ "grad_norm": 13.285860061645508,
469
+ "learning_rate": 2.4108355978260873e-05,
470
+ "loss": 0.0327,
471
+ "step": 3150
472
+ },
473
+ {
474
+ "epoch": 2.1376085504342015,
475
+ "grad_norm": 0.06732609122991562,
476
+ "learning_rate": 2.3683763586956523e-05,
477
+ "loss": 0.0366,
478
+ "step": 3200
479
+ },
480
+ {
481
+ "epoch": 2.171008684034736,
482
+ "grad_norm": 0.042977988719940186,
483
+ "learning_rate": 2.3259171195652173e-05,
484
+ "loss": 0.0979,
485
+ "step": 3250
486
+ },
487
+ {
488
+ "epoch": 2.2044088176352705,
489
+ "grad_norm": 0.2444009780883789,
490
+ "learning_rate": 2.283457880434783e-05,
491
+ "loss": 0.0185,
492
+ "step": 3300
493
+ },
494
+ {
495
+ "epoch": 2.237808951235805,
496
+ "grad_norm": 0.013526637107133865,
497
+ "learning_rate": 2.240998641304348e-05,
498
+ "loss": 0.049,
499
+ "step": 3350
500
+ },
501
+ {
502
+ "epoch": 2.2712090848363395,
503
+ "grad_norm": 0.034718696027994156,
504
+ "learning_rate": 2.198539402173913e-05,
505
+ "loss": 0.0512,
506
+ "step": 3400
507
+ },
508
+ {
509
+ "epoch": 2.3046092184368736,
510
+ "grad_norm": 4.7193403244018555,
511
+ "learning_rate": 2.1560801630434786e-05,
512
+ "loss": 0.0479,
513
+ "step": 3450
514
+ },
515
+ {
516
+ "epoch": 2.338009352037408,
517
+ "grad_norm": 0.019014885649085045,
518
+ "learning_rate": 2.1136209239130436e-05,
519
+ "loss": 0.0924,
520
+ "step": 3500
521
+ },
522
+ {
523
+ "epoch": 2.3714094856379426,
524
+ "grad_norm": 0.03284880891442299,
525
+ "learning_rate": 2.071161684782609e-05,
526
+ "loss": 0.045,
527
+ "step": 3550
528
+ },
529
+ {
530
+ "epoch": 2.404809619238477,
531
+ "grad_norm": 0.014006136916577816,
532
+ "learning_rate": 2.028702445652174e-05,
533
+ "loss": 0.0518,
534
+ "step": 3600
535
+ },
536
+ {
537
+ "epoch": 2.438209752839011,
538
+ "grad_norm": 0.05049414187669754,
539
+ "learning_rate": 1.9862432065217392e-05,
540
+ "loss": 0.0482,
541
+ "step": 3650
542
+ },
543
+ {
544
+ "epoch": 2.4716098864395457,
545
+ "grad_norm": 0.045781753957271576,
546
+ "learning_rate": 1.9437839673913045e-05,
547
+ "loss": 0.0143,
548
+ "step": 3700
549
+ },
550
+ {
551
+ "epoch": 2.50501002004008,
552
+ "grad_norm": 0.009661749005317688,
553
+ "learning_rate": 1.9013247282608695e-05,
554
+ "loss": 0.0551,
555
+ "step": 3750
556
+ },
557
+ {
558
+ "epoch": 2.5384101536406147,
559
+ "grad_norm": 0.009859848767518997,
560
+ "learning_rate": 1.858865489130435e-05,
561
+ "loss": 0.0495,
562
+ "step": 3800
563
+ },
564
+ {
565
+ "epoch": 2.5718102872411492,
566
+ "grad_norm": 10.117756843566895,
567
+ "learning_rate": 1.8164062500000002e-05,
568
+ "loss": 0.0412,
569
+ "step": 3850
570
+ },
571
+ {
572
+ "epoch": 2.6052104208416833,
573
+ "grad_norm": 0.015856023877859116,
574
+ "learning_rate": 1.7739470108695652e-05,
575
+ "loss": 0.0505,
576
+ "step": 3900
577
+ },
578
+ {
579
+ "epoch": 2.638610554442218,
580
+ "grad_norm": 0.08077731728553772,
581
+ "learning_rate": 1.7314877717391305e-05,
582
+ "loss": 0.0458,
583
+ "step": 3950
584
+ },
585
+ {
586
+ "epoch": 2.6720106880427523,
587
+ "grad_norm": 0.1191636249423027,
588
+ "learning_rate": 1.689028532608696e-05,
589
+ "loss": 0.0473,
590
+ "step": 4000
591
+ },
592
+ {
593
+ "epoch": 2.7054108216432864,
594
+ "grad_norm": 0.012686469592154026,
595
+ "learning_rate": 1.6465692934782608e-05,
596
+ "loss": 0.0546,
597
+ "step": 4050
598
+ },
599
+ {
600
+ "epoch": 2.738810955243821,
601
+ "grad_norm": 0.011968459002673626,
602
+ "learning_rate": 1.604110054347826e-05,
603
+ "loss": 0.0259,
604
+ "step": 4100
605
+ },
606
+ {
607
+ "epoch": 2.7722110888443554,
608
+ "grad_norm": 0.013011530973017216,
609
+ "learning_rate": 1.5616508152173915e-05,
610
+ "loss": 0.0165,
611
+ "step": 4150
612
+ },
613
+ {
614
+ "epoch": 2.80561122244489,
615
+ "grad_norm": 0.03517712280154228,
616
+ "learning_rate": 1.5191915760869566e-05,
617
+ "loss": 0.0291,
618
+ "step": 4200
619
+ },
620
+ {
621
+ "epoch": 2.8390113560454244,
622
+ "grad_norm": 0.03560580685734749,
623
+ "learning_rate": 1.4767323369565216e-05,
624
+ "loss": 0.0686,
625
+ "step": 4250
626
+ },
627
+ {
628
+ "epoch": 2.8724114896459585,
629
+ "grad_norm": 0.021690141409635544,
630
+ "learning_rate": 1.4342730978260871e-05,
631
+ "loss": 0.0482,
632
+ "step": 4300
633
+ },
634
+ {
635
+ "epoch": 2.905811623246493,
636
+ "grad_norm": 0.015762945637106895,
637
+ "learning_rate": 1.3918138586956523e-05,
638
+ "loss": 0.0242,
639
+ "step": 4350
640
+ },
641
+ {
642
+ "epoch": 2.9392117568470275,
643
+ "grad_norm": 0.12395244836807251,
644
+ "learning_rate": 1.3493546195652174e-05,
645
+ "loss": 0.0451,
646
+ "step": 4400
647
+ },
648
+ {
649
+ "epoch": 2.9726118904475616,
650
+ "grad_norm": 0.012854584492743015,
651
+ "learning_rate": 1.3068953804347828e-05,
652
+ "loss": 0.0521,
653
+ "step": 4450
654
+ },
655
+ {
656
+ "epoch": 3.0,
657
+ "eval_accuracy": 0.9862754377662092,
658
+ "eval_f1": 0.888030888030888,
659
+ "eval_loss": 0.056181490421295166,
660
+ "eval_runtime": 8.2762,
661
+ "eval_samples_per_second": 255.31,
662
+ "eval_steps_per_second": 32.02,
663
+ "step": 4491
664
+ }
665
+ ],
666
+ "logging_steps": 50,
667
+ "max_steps": 5988,
668
+ "num_input_tokens_seen": 0,
669
+ "num_train_epochs": 4,
670
+ "save_steps": 500,
671
+ "stateful_callbacks": {
672
+ "TrainerControl": {
673
+ "args": {
674
+ "should_epoch_stop": false,
675
+ "should_evaluate": false,
676
+ "should_log": false,
677
+ "should_save": true,
678
+ "should_training_stop": false
679
+ },
680
+ "attributes": {}
681
+ }
682
+ },
683
+ "total_flos": 1056635820122112.0,
684
+ "train_batch_size": 8,
685
+ "trial_name": null,
686
+ "trial_params": null
687
+ }
checkpoint-4491/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e03a8046e1b1da0fe3b9b933f56a9589c401ad31953c08b2477ec29e5fff7bf3
3
+ size 5713
checkpoint-4491/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-5988/config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ElectraForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "embedding_size": 128,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 256,
11
+ "id2label": {
12
+ "0": "NOT_HEADING",
13
+ "1": "HEADING"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 1024,
17
+ "label2id": {
18
+ "HEADING": 1,
19
+ "NOT_HEADING": 0
20
+ },
21
+ "layer_norm_eps": 1e-12,
22
+ "max_position_embeddings": 512,
23
+ "model_type": "electra",
24
+ "num_attention_heads": 4,
25
+ "num_hidden_layers": 12,
26
+ "pad_token_id": 0,
27
+ "position_embedding_type": "absolute",
28
+ "problem_type": "single_label_classification",
29
+ "summary_activation": "gelu",
30
+ "summary_last_dropout": 0.1,
31
+ "summary_type": "first",
32
+ "summary_use_proj": true,
33
+ "torch_dtype": "float32",
34
+ "transformers_version": "4.53.3",
35
+ "type_vocab_size": 2,
36
+ "use_cache": true,
37
+ "vocab_size": 30522
38
+ }
checkpoint-5988/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd44b6dd249aee7ff0f6ea19dc5b6a1da7bf4fd922e025166e1fbfa57ee6da3f
3
+ size 54221200
checkpoint-5988/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dce673cb0f10bf0ec58888ca4a52615874bfa3f88319adf6b83c4279acacef15
3
+ size 108564747
checkpoint-5988/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55c340978e106df61e3665af36a34df3227ab90ceb40186fb8827e9f6c020e1d
3
+ size 14645
checkpoint-5988/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0439fd7c1120bbbda24e25e1d56b66ff1c606dfe44f054ad7427f541d074f64
3
+ size 1465
checkpoint-5988/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-5988/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-5988/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": "ElectraTokenizer",
55
+ "unk_token": "[UNK]"
56
+ }
checkpoint-5988/trainer_state.json ADDED
@@ -0,0 +1,907 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 5988,
3
+ "best_metric": 0.9053497942386831,
4
+ "best_model_checkpoint": "./electra-small-heading-classifier-expanded\\checkpoint-5988",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 5988,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.033400133600534405,
14
+ "grad_norm": 1.82759428024292,
15
+ "learning_rate": 2.45e-05,
16
+ "loss": 0.5801,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.06680026720106881,
21
+ "grad_norm": 0.7336680293083191,
22
+ "learning_rate": 4.9500000000000004e-05,
23
+ "loss": 0.3191,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.10020040080160321,
28
+ "grad_norm": 0.759293258190155,
29
+ "learning_rate": 4.9583899456521745e-05,
30
+ "loss": 0.2216,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.13360053440213762,
35
+ "grad_norm": 0.5498169660568237,
36
+ "learning_rate": 4.915930706521739e-05,
37
+ "loss": 0.2167,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 0.16700066800267202,
42
+ "grad_norm": 3.572784662246704,
43
+ "learning_rate": 4.8734714673913044e-05,
44
+ "loss": 0.1797,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 0.20040080160320642,
49
+ "grad_norm": 1.3190865516662598,
50
+ "learning_rate": 4.83101222826087e-05,
51
+ "loss": 0.2202,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.23380093520374082,
56
+ "grad_norm": 1.5133466720581055,
57
+ "learning_rate": 4.788552989130435e-05,
58
+ "loss": 0.1925,
59
+ "step": 350
60
+ },
61
+ {
62
+ "epoch": 0.26720106880427524,
63
+ "grad_norm": 1.3208128213882446,
64
+ "learning_rate": 4.7460937500000004e-05,
65
+ "loss": 0.265,
66
+ "step": 400
67
+ },
68
+ {
69
+ "epoch": 0.30060120240480964,
70
+ "grad_norm": 0.9725052118301392,
71
+ "learning_rate": 4.703634510869566e-05,
72
+ "loss": 0.2866,
73
+ "step": 450
74
+ },
75
+ {
76
+ "epoch": 0.33400133600534404,
77
+ "grad_norm": 1.570516586303711,
78
+ "learning_rate": 4.6611752717391304e-05,
79
+ "loss": 0.1881,
80
+ "step": 500
81
+ },
82
+ {
83
+ "epoch": 0.36740146960587844,
84
+ "grad_norm": 4.067606449127197,
85
+ "learning_rate": 4.618716032608696e-05,
86
+ "loss": 0.2258,
87
+ "step": 550
88
+ },
89
+ {
90
+ "epoch": 0.40080160320641284,
91
+ "grad_norm": 0.2925869822502136,
92
+ "learning_rate": 4.576256793478261e-05,
93
+ "loss": 0.1488,
94
+ "step": 600
95
+ },
96
+ {
97
+ "epoch": 0.43420173680694724,
98
+ "grad_norm": 0.9354973435401917,
99
+ "learning_rate": 4.5337975543478264e-05,
100
+ "loss": 0.1611,
101
+ "step": 650
102
+ },
103
+ {
104
+ "epoch": 0.46760187040748163,
105
+ "grad_norm": 0.3285213112831116,
106
+ "learning_rate": 4.491338315217392e-05,
107
+ "loss": 0.1259,
108
+ "step": 700
109
+ },
110
+ {
111
+ "epoch": 0.501002004008016,
112
+ "grad_norm": 13.97227954864502,
113
+ "learning_rate": 4.448879076086957e-05,
114
+ "loss": 0.1593,
115
+ "step": 750
116
+ },
117
+ {
118
+ "epoch": 0.5344021376085505,
119
+ "grad_norm": 0.10469996929168701,
120
+ "learning_rate": 4.406419836956522e-05,
121
+ "loss": 0.0923,
122
+ "step": 800
123
+ },
124
+ {
125
+ "epoch": 0.5678022712090849,
126
+ "grad_norm": 0.19770751893520355,
127
+ "learning_rate": 4.363960597826087e-05,
128
+ "loss": 0.1067,
129
+ "step": 850
130
+ },
131
+ {
132
+ "epoch": 0.6012024048096193,
133
+ "grad_norm": 0.10104553401470184,
134
+ "learning_rate": 4.321501358695652e-05,
135
+ "loss": 0.1214,
136
+ "step": 900
137
+ },
138
+ {
139
+ "epoch": 0.6346025384101537,
140
+ "grad_norm": 0.10949808359146118,
141
+ "learning_rate": 4.279042119565218e-05,
142
+ "loss": 0.1175,
143
+ "step": 950
144
+ },
145
+ {
146
+ "epoch": 0.6680026720106881,
147
+ "grad_norm": 0.10336494445800781,
148
+ "learning_rate": 4.236582880434783e-05,
149
+ "loss": 0.0849,
150
+ "step": 1000
151
+ },
152
+ {
153
+ "epoch": 0.7014028056112225,
154
+ "grad_norm": 0.12163177132606506,
155
+ "learning_rate": 4.1941236413043476e-05,
156
+ "loss": 0.1486,
157
+ "step": 1050
158
+ },
159
+ {
160
+ "epoch": 0.7348029392117569,
161
+ "grad_norm": 0.0906616598367691,
162
+ "learning_rate": 4.151664402173913e-05,
163
+ "loss": 0.0671,
164
+ "step": 1100
165
+ },
166
+ {
167
+ "epoch": 0.7682030728122913,
168
+ "grad_norm": 0.12954840064048767,
169
+ "learning_rate": 4.109205163043478e-05,
170
+ "loss": 0.072,
171
+ "step": 1150
172
+ },
173
+ {
174
+ "epoch": 0.8016032064128257,
175
+ "grad_norm": 0.06860172003507614,
176
+ "learning_rate": 4.0667459239130436e-05,
177
+ "loss": 0.0735,
178
+ "step": 1200
179
+ },
180
+ {
181
+ "epoch": 0.8350033400133601,
182
+ "grad_norm": 0.06437909603118896,
183
+ "learning_rate": 4.024286684782609e-05,
184
+ "loss": 0.16,
185
+ "step": 1250
186
+ },
187
+ {
188
+ "epoch": 0.8684034736138945,
189
+ "grad_norm": 3.2551543712615967,
190
+ "learning_rate": 3.981827445652174e-05,
191
+ "loss": 0.0868,
192
+ "step": 1300
193
+ },
194
+ {
195
+ "epoch": 0.9018036072144289,
196
+ "grad_norm": 0.12183782458305359,
197
+ "learning_rate": 3.939368206521739e-05,
198
+ "loss": 0.1346,
199
+ "step": 1350
200
+ },
201
+ {
202
+ "epoch": 0.9352037408149633,
203
+ "grad_norm": 5.742549419403076,
204
+ "learning_rate": 3.896908967391304e-05,
205
+ "loss": 0.0977,
206
+ "step": 1400
207
+ },
208
+ {
209
+ "epoch": 0.9686038744154977,
210
+ "grad_norm": 0.049525078386068344,
211
+ "learning_rate": 3.85444972826087e-05,
212
+ "loss": 0.0771,
213
+ "step": 1450
214
+ },
215
+ {
216
+ "epoch": 1.0,
217
+ "eval_accuracy": 0.9777567439659253,
218
+ "eval_f1": 0.7813953488372093,
219
+ "eval_loss": 0.07437803596258163,
220
+ "eval_runtime": 8.1317,
221
+ "eval_samples_per_second": 259.846,
222
+ "eval_steps_per_second": 32.588,
223
+ "step": 1497
224
+ },
225
+ {
226
+ "epoch": 1.002004008016032,
227
+ "grad_norm": 0.08354274183511734,
228
+ "learning_rate": 3.811990489130435e-05,
229
+ "loss": 0.1122,
230
+ "step": 1500
231
+ },
232
+ {
233
+ "epoch": 1.0354041416165665,
234
+ "grad_norm": 0.07657646387815475,
235
+ "learning_rate": 3.76953125e-05,
236
+ "loss": 0.0776,
237
+ "step": 1550
238
+ },
239
+ {
240
+ "epoch": 1.0688042752171008,
241
+ "grad_norm": 0.029183173552155495,
242
+ "learning_rate": 3.7270720108695656e-05,
243
+ "loss": 0.048,
244
+ "step": 1600
245
+ },
246
+ {
247
+ "epoch": 1.1022044088176353,
248
+ "grad_norm": 0.6650001406669617,
249
+ "learning_rate": 3.68461277173913e-05,
250
+ "loss": 0.0651,
251
+ "step": 1650
252
+ },
253
+ {
254
+ "epoch": 1.1356045424181698,
255
+ "grad_norm": 0.13291126489639282,
256
+ "learning_rate": 3.6421535326086955e-05,
257
+ "loss": 0.1035,
258
+ "step": 1700
259
+ },
260
+ {
261
+ "epoch": 1.169004676018704,
262
+ "grad_norm": 0.14512984454631805,
263
+ "learning_rate": 3.5996942934782615e-05,
264
+ "loss": 0.0896,
265
+ "step": 1750
266
+ },
267
+ {
268
+ "epoch": 1.2024048096192386,
269
+ "grad_norm": 0.09467575699090958,
270
+ "learning_rate": 3.557235054347826e-05,
271
+ "loss": 0.0763,
272
+ "step": 1800
273
+ },
274
+ {
275
+ "epoch": 1.2358049432197729,
276
+ "grad_norm": 0.5623799562454224,
277
+ "learning_rate": 3.5147758152173915e-05,
278
+ "loss": 0.059,
279
+ "step": 1850
280
+ },
281
+ {
282
+ "epoch": 1.2692050768203074,
283
+ "grad_norm": 0.16416147351264954,
284
+ "learning_rate": 3.472316576086957e-05,
285
+ "loss": 0.0483,
286
+ "step": 1900
287
+ },
288
+ {
289
+ "epoch": 1.3026052104208417,
290
+ "grad_norm": 3.0964038372039795,
291
+ "learning_rate": 3.4298573369565215e-05,
292
+ "loss": 0.1058,
293
+ "step": 1950
294
+ },
295
+ {
296
+ "epoch": 1.3360053440213762,
297
+ "grad_norm": 0.030443737283349037,
298
+ "learning_rate": 3.3873980978260875e-05,
299
+ "loss": 0.0517,
300
+ "step": 2000
301
+ },
302
+ {
303
+ "epoch": 1.3694054776219104,
304
+ "grad_norm": 0.04263285547494888,
305
+ "learning_rate": 3.344938858695653e-05,
306
+ "loss": 0.0486,
307
+ "step": 2050
308
+ },
309
+ {
310
+ "epoch": 1.402805611222445,
311
+ "grad_norm": 0.06713565438985825,
312
+ "learning_rate": 3.3024796195652175e-05,
313
+ "loss": 0.0494,
314
+ "step": 2100
315
+ },
316
+ {
317
+ "epoch": 1.4362057448229792,
318
+ "grad_norm": 0.04323391616344452,
319
+ "learning_rate": 3.260020380434783e-05,
320
+ "loss": 0.0518,
321
+ "step": 2150
322
+ },
323
+ {
324
+ "epoch": 1.4696058784235138,
325
+ "grad_norm": 0.9192191362380981,
326
+ "learning_rate": 3.2175611413043474e-05,
327
+ "loss": 0.0498,
328
+ "step": 2200
329
+ },
330
+ {
331
+ "epoch": 1.503006012024048,
332
+ "grad_norm": 0.08719488233327866,
333
+ "learning_rate": 3.1751019021739135e-05,
334
+ "loss": 0.1155,
335
+ "step": 2250
336
+ },
337
+ {
338
+ "epoch": 1.5364061456245826,
339
+ "grad_norm": 6.060549259185791,
340
+ "learning_rate": 3.132642663043479e-05,
341
+ "loss": 0.1349,
342
+ "step": 2300
343
+ },
344
+ {
345
+ "epoch": 1.569806279225117,
346
+ "grad_norm": 0.09453130513429642,
347
+ "learning_rate": 3.0901834239130434e-05,
348
+ "loss": 0.0639,
349
+ "step": 2350
350
+ },
351
+ {
352
+ "epoch": 1.6032064128256514,
353
+ "grad_norm": 0.02592223510146141,
354
+ "learning_rate": 3.0477241847826088e-05,
355
+ "loss": 0.0929,
356
+ "step": 2400
357
+ },
358
+ {
359
+ "epoch": 1.6366065464261856,
360
+ "grad_norm": 0.027360519394278526,
361
+ "learning_rate": 3.005264945652174e-05,
362
+ "loss": 0.071,
363
+ "step": 2450
364
+ },
365
+ {
366
+ "epoch": 1.6700066800267201,
367
+ "grad_norm": 0.14285916090011597,
368
+ "learning_rate": 2.962805706521739e-05,
369
+ "loss": 0.0768,
370
+ "step": 2500
371
+ },
372
+ {
373
+ "epoch": 1.7034068136272547,
374
+ "grad_norm": 0.04874153807759285,
375
+ "learning_rate": 2.9203464673913044e-05,
376
+ "loss": 0.076,
377
+ "step": 2550
378
+ },
379
+ {
380
+ "epoch": 1.736806947227789,
381
+ "grad_norm": 0.09328486025333405,
382
+ "learning_rate": 2.8778872282608697e-05,
383
+ "loss": 0.0505,
384
+ "step": 2600
385
+ },
386
+ {
387
+ "epoch": 1.7702070808283232,
388
+ "grad_norm": 0.34670591354370117,
389
+ "learning_rate": 2.8354279891304347e-05,
390
+ "loss": 0.0763,
391
+ "step": 2650
392
+ },
393
+ {
394
+ "epoch": 1.8036072144288577,
395
+ "grad_norm": 0.057789236307144165,
396
+ "learning_rate": 2.79296875e-05,
397
+ "loss": 0.0832,
398
+ "step": 2700
399
+ },
400
+ {
401
+ "epoch": 1.8370073480293923,
402
+ "grad_norm": 0.059989944100379944,
403
+ "learning_rate": 2.7505095108695657e-05,
404
+ "loss": 0.062,
405
+ "step": 2750
406
+ },
407
+ {
408
+ "epoch": 1.8704074816299265,
409
+ "grad_norm": 0.02221057377755642,
410
+ "learning_rate": 2.7080502717391304e-05,
411
+ "loss": 0.0649,
412
+ "step": 2800
413
+ },
414
+ {
415
+ "epoch": 1.9038076152304608,
416
+ "grad_norm": 0.154799684882164,
417
+ "learning_rate": 2.6655910326086957e-05,
418
+ "loss": 0.144,
419
+ "step": 2850
420
+ },
421
+ {
422
+ "epoch": 1.9372077488309953,
423
+ "grad_norm": 0.0422816276550293,
424
+ "learning_rate": 2.6231317934782613e-05,
425
+ "loss": 0.0531,
426
+ "step": 2900
427
+ },
428
+ {
429
+ "epoch": 1.9706078824315298,
430
+ "grad_norm": 0.019839206710457802,
431
+ "learning_rate": 2.580672554347826e-05,
432
+ "loss": 0.0545,
433
+ "step": 2950
434
+ },
435
+ {
436
+ "epoch": 2.0,
437
+ "eval_accuracy": 0.9801230477993375,
438
+ "eval_f1": 0.8489208633093526,
439
+ "eval_loss": 0.07019170373678207,
440
+ "eval_runtime": 8.2123,
441
+ "eval_samples_per_second": 257.298,
442
+ "eval_steps_per_second": 32.269,
443
+ "step": 2994
444
+ },
445
+ {
446
+ "epoch": 2.004008016032064,
447
+ "grad_norm": 3.3859903812408447,
448
+ "learning_rate": 2.5382133152173913e-05,
449
+ "loss": 0.0891,
450
+ "step": 3000
451
+ },
452
+ {
453
+ "epoch": 2.0374081496325984,
454
+ "grad_norm": 0.055893734097480774,
455
+ "learning_rate": 2.4957540760869567e-05,
456
+ "loss": 0.0498,
457
+ "step": 3050
458
+ },
459
+ {
460
+ "epoch": 2.070808283233133,
461
+ "grad_norm": 0.030930357053875923,
462
+ "learning_rate": 2.4532948369565216e-05,
463
+ "loss": 0.0272,
464
+ "step": 3100
465
+ },
466
+ {
467
+ "epoch": 2.1042084168336674,
468
+ "grad_norm": 13.285860061645508,
469
+ "learning_rate": 2.4108355978260873e-05,
470
+ "loss": 0.0327,
471
+ "step": 3150
472
+ },
473
+ {
474
+ "epoch": 2.1376085504342015,
475
+ "grad_norm": 0.06732609122991562,
476
+ "learning_rate": 2.3683763586956523e-05,
477
+ "loss": 0.0366,
478
+ "step": 3200
479
+ },
480
+ {
481
+ "epoch": 2.171008684034736,
482
+ "grad_norm": 0.042977988719940186,
483
+ "learning_rate": 2.3259171195652173e-05,
484
+ "loss": 0.0979,
485
+ "step": 3250
486
+ },
487
+ {
488
+ "epoch": 2.2044088176352705,
489
+ "grad_norm": 0.2444009780883789,
490
+ "learning_rate": 2.283457880434783e-05,
491
+ "loss": 0.0185,
492
+ "step": 3300
493
+ },
494
+ {
495
+ "epoch": 2.237808951235805,
496
+ "grad_norm": 0.013526637107133865,
497
+ "learning_rate": 2.240998641304348e-05,
498
+ "loss": 0.049,
499
+ "step": 3350
500
+ },
501
+ {
502
+ "epoch": 2.2712090848363395,
503
+ "grad_norm": 0.034718696027994156,
504
+ "learning_rate": 2.198539402173913e-05,
505
+ "loss": 0.0512,
506
+ "step": 3400
507
+ },
508
+ {
509
+ "epoch": 2.3046092184368736,
510
+ "grad_norm": 4.7193403244018555,
511
+ "learning_rate": 2.1560801630434786e-05,
512
+ "loss": 0.0479,
513
+ "step": 3450
514
+ },
515
+ {
516
+ "epoch": 2.338009352037408,
517
+ "grad_norm": 0.019014885649085045,
518
+ "learning_rate": 2.1136209239130436e-05,
519
+ "loss": 0.0924,
520
+ "step": 3500
521
+ },
522
+ {
523
+ "epoch": 2.3714094856379426,
524
+ "grad_norm": 0.03284880891442299,
525
+ "learning_rate": 2.071161684782609e-05,
526
+ "loss": 0.045,
527
+ "step": 3550
528
+ },
529
+ {
530
+ "epoch": 2.404809619238477,
531
+ "grad_norm": 0.014006136916577816,
532
+ "learning_rate": 2.028702445652174e-05,
533
+ "loss": 0.0518,
534
+ "step": 3600
535
+ },
536
+ {
537
+ "epoch": 2.438209752839011,
538
+ "grad_norm": 0.05049414187669754,
539
+ "learning_rate": 1.9862432065217392e-05,
540
+ "loss": 0.0482,
541
+ "step": 3650
542
+ },
543
+ {
544
+ "epoch": 2.4716098864395457,
545
+ "grad_norm": 0.045781753957271576,
546
+ "learning_rate": 1.9437839673913045e-05,
547
+ "loss": 0.0143,
548
+ "step": 3700
549
+ },
550
+ {
551
+ "epoch": 2.50501002004008,
552
+ "grad_norm": 0.009661749005317688,
553
+ "learning_rate": 1.9013247282608695e-05,
554
+ "loss": 0.0551,
555
+ "step": 3750
556
+ },
557
+ {
558
+ "epoch": 2.5384101536406147,
559
+ "grad_norm": 0.009859848767518997,
560
+ "learning_rate": 1.858865489130435e-05,
561
+ "loss": 0.0495,
562
+ "step": 3800
563
+ },
564
+ {
565
+ "epoch": 2.5718102872411492,
566
+ "grad_norm": 10.117756843566895,
567
+ "learning_rate": 1.8164062500000002e-05,
568
+ "loss": 0.0412,
569
+ "step": 3850
570
+ },
571
+ {
572
+ "epoch": 2.6052104208416833,
573
+ "grad_norm": 0.015856023877859116,
574
+ "learning_rate": 1.7739470108695652e-05,
575
+ "loss": 0.0505,
576
+ "step": 3900
577
+ },
578
+ {
579
+ "epoch": 2.638610554442218,
580
+ "grad_norm": 0.08077731728553772,
581
+ "learning_rate": 1.7314877717391305e-05,
582
+ "loss": 0.0458,
583
+ "step": 3950
584
+ },
585
+ {
586
+ "epoch": 2.6720106880427523,
587
+ "grad_norm": 0.1191636249423027,
588
+ "learning_rate": 1.689028532608696e-05,
589
+ "loss": 0.0473,
590
+ "step": 4000
591
+ },
592
+ {
593
+ "epoch": 2.7054108216432864,
594
+ "grad_norm": 0.012686469592154026,
595
+ "learning_rate": 1.6465692934782608e-05,
596
+ "loss": 0.0546,
597
+ "step": 4050
598
+ },
599
+ {
600
+ "epoch": 2.738810955243821,
601
+ "grad_norm": 0.011968459002673626,
602
+ "learning_rate": 1.604110054347826e-05,
603
+ "loss": 0.0259,
604
+ "step": 4100
605
+ },
606
+ {
607
+ "epoch": 2.7722110888443554,
608
+ "grad_norm": 0.013011530973017216,
609
+ "learning_rate": 1.5616508152173915e-05,
610
+ "loss": 0.0165,
611
+ "step": 4150
612
+ },
613
+ {
614
+ "epoch": 2.80561122244489,
615
+ "grad_norm": 0.03517712280154228,
616
+ "learning_rate": 1.5191915760869566e-05,
617
+ "loss": 0.0291,
618
+ "step": 4200
619
+ },
620
+ {
621
+ "epoch": 2.8390113560454244,
622
+ "grad_norm": 0.03560580685734749,
623
+ "learning_rate": 1.4767323369565216e-05,
624
+ "loss": 0.0686,
625
+ "step": 4250
626
+ },
627
+ {
628
+ "epoch": 2.8724114896459585,
629
+ "grad_norm": 0.021690141409635544,
630
+ "learning_rate": 1.4342730978260871e-05,
631
+ "loss": 0.0482,
632
+ "step": 4300
633
+ },
634
+ {
635
+ "epoch": 2.905811623246493,
636
+ "grad_norm": 0.015762945637106895,
637
+ "learning_rate": 1.3918138586956523e-05,
638
+ "loss": 0.0242,
639
+ "step": 4350
640
+ },
641
+ {
642
+ "epoch": 2.9392117568470275,
643
+ "grad_norm": 0.12395244836807251,
644
+ "learning_rate": 1.3493546195652174e-05,
645
+ "loss": 0.0451,
646
+ "step": 4400
647
+ },
648
+ {
649
+ "epoch": 2.9726118904475616,
650
+ "grad_norm": 0.012854584492743015,
651
+ "learning_rate": 1.3068953804347828e-05,
652
+ "loss": 0.0521,
653
+ "step": 4450
654
+ },
655
+ {
656
+ "epoch": 3.0,
657
+ "eval_accuracy": 0.9862754377662092,
658
+ "eval_f1": 0.888030888030888,
659
+ "eval_loss": 0.056181490421295166,
660
+ "eval_runtime": 8.2762,
661
+ "eval_samples_per_second": 255.31,
662
+ "eval_steps_per_second": 32.02,
663
+ "step": 4491
664
+ },
665
+ {
666
+ "epoch": 3.006012024048096,
667
+ "grad_norm": 0.020279008895158768,
668
+ "learning_rate": 1.264436141304348e-05,
669
+ "loss": 0.09,
670
+ "step": 4500
671
+ },
672
+ {
673
+ "epoch": 3.0394121576486306,
674
+ "grad_norm": 0.00978385005146265,
675
+ "learning_rate": 1.2219769021739132e-05,
676
+ "loss": 0.0353,
677
+ "step": 4550
678
+ },
679
+ {
680
+ "epoch": 3.072812291249165,
681
+ "grad_norm": 0.16378536820411682,
682
+ "learning_rate": 1.1795176630434782e-05,
683
+ "loss": 0.0214,
684
+ "step": 4600
685
+ },
686
+ {
687
+ "epoch": 3.1062124248496996,
688
+ "grad_norm": 0.00774746248498559,
689
+ "learning_rate": 1.1370584239130436e-05,
690
+ "loss": 0.0061,
691
+ "step": 4650
692
+ },
693
+ {
694
+ "epoch": 3.1396125584502337,
695
+ "grad_norm": 0.03240759298205376,
696
+ "learning_rate": 1.0945991847826087e-05,
697
+ "loss": 0.0437,
698
+ "step": 4700
699
+ },
700
+ {
701
+ "epoch": 3.173012692050768,
702
+ "grad_norm": 0.06572789698839188,
703
+ "learning_rate": 1.052139945652174e-05,
704
+ "loss": 0.0175,
705
+ "step": 4750
706
+ },
707
+ {
708
+ "epoch": 3.2064128256513027,
709
+ "grad_norm": 0.00863126665353775,
710
+ "learning_rate": 1.0096807065217392e-05,
711
+ "loss": 0.0357,
712
+ "step": 4800
713
+ },
714
+ {
715
+ "epoch": 3.239812959251837,
716
+ "grad_norm": 0.008642938919365406,
717
+ "learning_rate": 9.672214673913044e-06,
718
+ "loss": 0.0156,
719
+ "step": 4850
720
+ },
721
+ {
722
+ "epoch": 3.2732130928523713,
723
+ "grad_norm": 0.01715974509716034,
724
+ "learning_rate": 9.247622282608697e-06,
725
+ "loss": 0.03,
726
+ "step": 4900
727
+ },
728
+ {
729
+ "epoch": 3.306613226452906,
730
+ "grad_norm": 0.007449580822139978,
731
+ "learning_rate": 8.823029891304348e-06,
732
+ "loss": 0.0136,
733
+ "step": 4950
734
+ },
735
+ {
736
+ "epoch": 3.3400133600534403,
737
+ "grad_norm": 0.011774774640798569,
738
+ "learning_rate": 8.3984375e-06,
739
+ "loss": 0.0213,
740
+ "step": 5000
741
+ },
742
+ {
743
+ "epoch": 3.373413493653975,
744
+ "grad_norm": 0.006117586512118578,
745
+ "learning_rate": 7.973845108695653e-06,
746
+ "loss": 0.0438,
747
+ "step": 5050
748
+ },
749
+ {
750
+ "epoch": 3.406813627254509,
751
+ "grad_norm": 0.08415533602237701,
752
+ "learning_rate": 7.549252717391304e-06,
753
+ "loss": 0.0375,
754
+ "step": 5100
755
+ },
756
+ {
757
+ "epoch": 3.4402137608550434,
758
+ "grad_norm": 0.013756927102804184,
759
+ "learning_rate": 7.124660326086957e-06,
760
+ "loss": 0.0477,
761
+ "step": 5150
762
+ },
763
+ {
764
+ "epoch": 3.473613894455578,
765
+ "grad_norm": 0.05258077010512352,
766
+ "learning_rate": 6.700067934782608e-06,
767
+ "loss": 0.0408,
768
+ "step": 5200
769
+ },
770
+ {
771
+ "epoch": 3.507014028056112,
772
+ "grad_norm": 0.03193260356783867,
773
+ "learning_rate": 6.275475543478261e-06,
774
+ "loss": 0.0739,
775
+ "step": 5250
776
+ },
777
+ {
778
+ "epoch": 3.5404141616566465,
779
+ "grad_norm": 0.0606362447142601,
780
+ "learning_rate": 5.850883152173913e-06,
781
+ "loss": 0.0445,
782
+ "step": 5300
783
+ },
784
+ {
785
+ "epoch": 3.573814295257181,
786
+ "grad_norm": 0.009357443079352379,
787
+ "learning_rate": 5.426290760869566e-06,
788
+ "loss": 0.0363,
789
+ "step": 5350
790
+ },
791
+ {
792
+ "epoch": 3.6072144288577155,
793
+ "grad_norm": 14.05784797668457,
794
+ "learning_rate": 5.001698369565218e-06,
795
+ "loss": 0.0227,
796
+ "step": 5400
797
+ },
798
+ {
799
+ "epoch": 3.64061456245825,
800
+ "grad_norm": 0.07442031055688858,
801
+ "learning_rate": 4.57710597826087e-06,
802
+ "loss": 0.005,
803
+ "step": 5450
804
+ },
805
+ {
806
+ "epoch": 3.6740146960587845,
807
+ "grad_norm": 0.007993385195732117,
808
+ "learning_rate": 4.152513586956522e-06,
809
+ "loss": 0.0121,
810
+ "step": 5500
811
+ },
812
+ {
813
+ "epoch": 3.7074148296593186,
814
+ "grad_norm": 0.020146405324339867,
815
+ "learning_rate": 3.7279211956521737e-06,
816
+ "loss": 0.0092,
817
+ "step": 5550
818
+ },
819
+ {
820
+ "epoch": 3.740814963259853,
821
+ "grad_norm": 0.0066772401332855225,
822
+ "learning_rate": 3.3033288043478265e-06,
823
+ "loss": 0.0322,
824
+ "step": 5600
825
+ },
826
+ {
827
+ "epoch": 3.7742150968603876,
828
+ "grad_norm": 0.007795071694999933,
829
+ "learning_rate": 2.8787364130434785e-06,
830
+ "loss": 0.0018,
831
+ "step": 5650
832
+ },
833
+ {
834
+ "epoch": 3.8076152304609217,
835
+ "grad_norm": 0.008354957215487957,
836
+ "learning_rate": 2.4541440217391305e-06,
837
+ "loss": 0.0221,
838
+ "step": 5700
839
+ },
840
+ {
841
+ "epoch": 3.841015364061456,
842
+ "grad_norm": 1.7550073862075806,
843
+ "learning_rate": 2.029551630434783e-06,
844
+ "loss": 0.0331,
845
+ "step": 5750
846
+ },
847
+ {
848
+ "epoch": 3.8744154976619907,
849
+ "grad_norm": 0.03482593223452568,
850
+ "learning_rate": 1.604959239130435e-06,
851
+ "loss": 0.0316,
852
+ "step": 5800
853
+ },
854
+ {
855
+ "epoch": 3.907815631262525,
856
+ "grad_norm": 0.0070787896402180195,
857
+ "learning_rate": 1.180366847826087e-06,
858
+ "loss": 0.0354,
859
+ "step": 5850
860
+ },
861
+ {
862
+ "epoch": 3.9412157648630597,
863
+ "grad_norm": 0.00712177250534296,
864
+ "learning_rate": 7.557744565217392e-07,
865
+ "loss": 0.0014,
866
+ "step": 5900
867
+ },
868
+ {
869
+ "epoch": 3.9746158984635938,
870
+ "grad_norm": 0.008186141029000282,
871
+ "learning_rate": 3.3118206521739133e-07,
872
+ "loss": 0.006,
873
+ "step": 5950
874
+ },
875
+ {
876
+ "epoch": 4.0,
877
+ "eval_accuracy": 0.9891150023663038,
878
+ "eval_f1": 0.9053497942386831,
879
+ "eval_loss": 0.054513588547706604,
880
+ "eval_runtime": 8.2838,
881
+ "eval_samples_per_second": 255.077,
882
+ "eval_steps_per_second": 31.99,
883
+ "step": 5988
884
+ }
885
+ ],
886
+ "logging_steps": 50,
887
+ "max_steps": 5988,
888
+ "num_input_tokens_seen": 0,
889
+ "num_train_epochs": 4,
890
+ "save_steps": 500,
891
+ "stateful_callbacks": {
892
+ "TrainerControl": {
893
+ "args": {
894
+ "should_epoch_stop": false,
895
+ "should_evaluate": false,
896
+ "should_log": false,
897
+ "should_save": true,
898
+ "should_training_stop": true
899
+ },
900
+ "attributes": {}
901
+ }
902
+ },
903
+ "total_flos": 1408847760162816.0,
904
+ "train_batch_size": 8,
905
+ "trial_name": null,
906
+ "trial_params": null
907
+ }
checkpoint-5988/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e03a8046e1b1da0fe3b9b933f56a9589c401ad31953c08b2477ec29e5fff7bf3
3
+ size 5713
checkpoint-5988/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ElectraForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "embedding_size": 128,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 256,
11
+ "id2label": {
12
+ "0": "NOT_HEADING",
13
+ "1": "HEADING"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 1024,
17
+ "label2id": {
18
+ "HEADING": 1,
19
+ "NOT_HEADING": 0
20
+ },
21
+ "layer_norm_eps": 1e-12,
22
+ "max_position_embeddings": 512,
23
+ "model_type": "electra",
24
+ "num_attention_heads": 4,
25
+ "num_hidden_layers": 12,
26
+ "pad_token_id": 0,
27
+ "position_embedding_type": "absolute",
28
+ "problem_type": "single_label_classification",
29
+ "summary_activation": "gelu",
30
+ "summary_last_dropout": 0.1,
31
+ "summary_type": "first",
32
+ "summary_use_proj": true,
33
+ "torch_dtype": "float32",
34
+ "transformers_version": "4.53.3",
35
+ "type_vocab_size": 2,
36
+ "use_cache": true,
37
+ "vocab_size": 30522
38
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd44b6dd249aee7ff0f6ea19dc5b6a1da7bf4fd922e025166e1fbfa57ee6da3f
3
+ size 54221200
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": "ElectraTokenizer",
55
+ "unk_token": "[UNK]"
56
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e03a8046e1b1da0fe3b9b933f56a9589c401ad31953c08b2477ec29e5fff7bf3
3
+ size 5713