Hugging Face's logo

heavyhelium
/
roberta-large-touche-base-binary

Text Classification
Transformers
Safetensors
roberta
Generated from Trainer
text-embeddings-inference
Model card Files
xet
 Community
heavyhelium commited on
Commit
726077d
·
verified ·
1 Parent(s): a9d550f

Training in progress, epoch 3, checkpoint

Browse files
Files changed (10) hide show
  1. checkpoint-279/config.json +37 -0
  2. checkpoint-279/model.safetensors +3 -0
  3. checkpoint-279/optimizer.pt +3 -0
  4. checkpoint-279/rng_state.pth +3 -0
  5. checkpoint-279/scaler.pt +3 -0
  6. checkpoint-279/scheduler.pt +3 -0
  7. checkpoint-279/tokenizer.json +0 -0
  8. checkpoint-279/tokenizer_config.json +17 -0
  9. checkpoint-279/trainer_state.json +256 -0
  10. checkpoint-279/training_args.bin +3 -0
checkpoint-279/config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "RobertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "dtype": "float32",
10
+ "eos_token_id": 2,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 1024,
14
+ "id2label": {
15
+ "0": "valid",
16
+ "1": "fallacy"
17
+ },
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 4096,
20
+ "is_decoder": false,
21
+ "label2id": {
22
+ "fallacy": 1,
23
+ "valid": 0
24
+ },
25
+ "layer_norm_eps": 1e-05,
26
+ "max_position_embeddings": 514,
27
+ "model_type": "roberta",
28
+ "num_attention_heads": 16,
29
+ "num_hidden_layers": 24,
30
+ "pad_token_id": 1,
31
+ "problem_type": "single_label_classification",
32
+ "tie_word_embeddings": true,
33
+ "transformers_version": "5.9.0",
34
+ "type_vocab_size": 1,
35
+ "use_cache": false,
36
+ "vocab_size": 50265
37
+ }
checkpoint-279/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0c2ea7f5abd2d36286e09e471f677bc2317da87a8f106814c282ef1a71e86e1
3
+ size 1421495416
checkpoint-279/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb6bfb6fa01a48fa245c3e19f8206c32e4a9d5570b921a740819ac33741108f5
3
+ size 2843229182
checkpoint-279/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:093012ed868f6851a9a2c31ed6e517aab2eaaf0713db7d293add670edb9d4b2a
3
+ size 14645
checkpoint-279/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18be69e97022417922b3271a0187967232ec090112ac65dd403e6ecdc524311b
3
+ size 1383
checkpoint-279/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3c666bb83b189d0abd750461a0a6c85b7889bee03060cdf301ab639ff3aad06
3
+ size 1465
checkpoint-279/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-279/tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "errors": "replace",
8
+ "is_local": false,
9
+ "local_files_only": false,
10
+ "mask_token": "<mask>",
11
+ "model_max_length": 512,
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "tokenizer_class": "RobertaTokenizer",
15
+ "trim_offsets": true,
16
+ "unk_token": "<unk>"
17
+ }
checkpoint-279/trainer_state.json ADDED
@@ -0,0 +1,256 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 186,
3
+ "best_metric": 0.64996499649965,
4
+ "best_model_checkpoint": "models/roberta-large-touche-base-binary/trainer/checkpoint-186",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 279,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.10810810810810811,
14
+ "grad_norm": 12.057156562805176,
15
+ "learning_rate": 6.4285714285714295e-06,
16
+ "loss": 1.51865234375,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.21621621621621623,
21
+ "grad_norm": 10.9874906539917,
22
+ "learning_rate": 1.3571428571428574e-05,
23
+ "loss": 1.37591552734375,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.32432432432432434,
28
+ "grad_norm": 10.673126220703125,
29
+ "learning_rate": 1.9920318725099602e-05,
30
+ "loss": 1.4565673828125,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.43243243243243246,
35
+ "grad_norm": 17.499927520751953,
36
+ "learning_rate": 1.912350597609562e-05,
37
+ "loss": 1.4800537109375,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.5405405405405406,
42
+ "grad_norm": 15.72847843170166,
43
+ "learning_rate": 1.8326693227091633e-05,
44
+ "loss": 1.447882080078125,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.6486486486486487,
49
+ "grad_norm": 11.744357109069824,
50
+ "learning_rate": 1.752988047808765e-05,
51
+ "loss": 1.3298416137695312,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.7567567567567568,
56
+ "grad_norm": 8.649158477783203,
57
+ "learning_rate": 1.6733067729083667e-05,
58
+ "loss": 1.40760498046875,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.8648648648648649,
63
+ "grad_norm": 15.254891395568848,
64
+ "learning_rate": 1.593625498007968e-05,
65
+ "loss": 1.349920654296875,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.972972972972973,
70
+ "grad_norm": 21.086435317993164,
71
+ "learning_rate": 1.5139442231075698e-05,
72
+ "loss": 1.3678375244140626,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 1.0,
77
+ "eval_accuracy": 0.515,
78
+ "eval_fallacy_f1": 0.6666666666666666,
79
+ "eval_loss": 0.6907617449760437,
80
+ "eval_macro_f1": 0.38837920489296635,
81
+ "eval_runtime": 1.539,
82
+ "eval_samples_per_second": 129.957,
83
+ "eval_steps_per_second": 8.447,
84
+ "step": 93
85
+ },
86
+ {
87
+ "epoch": 1.0756756756756758,
88
+ "grad_norm": 15.82499885559082,
89
+ "learning_rate": 1.4342629482071715e-05,
90
+ "loss": 1.250738525390625,
91
+ "step": 100
92
+ },
93
+ {
94
+ "epoch": 1.1837837837837837,
95
+ "grad_norm": 18.66545867919922,
96
+ "learning_rate": 1.354581673306773e-05,
97
+ "loss": 1.2540924072265625,
98
+ "step": 110
99
+ },
100
+ {
101
+ "epoch": 1.291891891891892,
102
+ "grad_norm": 26.059511184692383,
103
+ "learning_rate": 1.2749003984063746e-05,
104
+ "loss": 1.3781219482421876,
105
+ "step": 120
106
+ },
107
+ {
108
+ "epoch": 1.4,
109
+ "grad_norm": 16.263402938842773,
110
+ "learning_rate": 1.1952191235059762e-05,
111
+ "loss": 1.4477294921875,
112
+ "step": 130
113
+ },
114
+ {
115
+ "epoch": 1.5081081081081082,
116
+ "grad_norm": 14.684833526611328,
117
+ "learning_rate": 1.1155378486055777e-05,
118
+ "loss": 1.3233001708984375,
119
+ "step": 140
120
+ },
121
+ {
122
+ "epoch": 1.6162162162162161,
123
+ "grad_norm": 45.42232894897461,
124
+ "learning_rate": 1.0358565737051793e-05,
125
+ "loss": 1.227001953125,
126
+ "step": 150
127
+ },
128
+ {
129
+ "epoch": 1.7243243243243245,
130
+ "grad_norm": 61.13242721557617,
131
+ "learning_rate": 9.56175298804781e-06,
132
+ "loss": 1.3158493041992188,
133
+ "step": 160
134
+ },
135
+ {
136
+ "epoch": 1.8324324324324324,
137
+ "grad_norm": 18.299968719482422,
138
+ "learning_rate": 8.764940239043825e-06,
139
+ "loss": 1.3390182495117187,
140
+ "step": 170
141
+ },
142
+ {
143
+ "epoch": 1.9405405405405407,
144
+ "grad_norm": 26.040592193603516,
145
+ "learning_rate": 7.96812749003984e-06,
146
+ "loss": 1.277142333984375,
147
+ "step": 180
148
+ },
149
+ {
150
+ "epoch": 2.0,
151
+ "eval_accuracy": 0.65,
152
+ "eval_fallacy_f1": 0.6464646464646465,
153
+ "eval_loss": 0.6374948024749756,
154
+ "eval_macro_f1": 0.64996499649965,
155
+ "eval_runtime": 1.507,
156
+ "eval_samples_per_second": 132.715,
157
+ "eval_steps_per_second": 8.626,
158
+ "step": 186
159
+ },
160
+ {
161
+ "epoch": 2.0432432432432432,
162
+ "grad_norm": 36.60226821899414,
163
+ "learning_rate": 7.171314741035858e-06,
164
+ "loss": 1.2425552368164063,
165
+ "step": 190
166
+ },
167
+ {
168
+ "epoch": 2.1513513513513516,
169
+ "grad_norm": 31.960996627807617,
170
+ "learning_rate": 6.374501992031873e-06,
171
+ "loss": 1.236676025390625,
172
+ "step": 200
173
+ },
174
+ {
175
+ "epoch": 2.2594594594594595,
176
+ "grad_norm": 24.7769718170166,
177
+ "learning_rate": 5.577689243027889e-06,
178
+ "loss": 1.2561294555664062,
179
+ "step": 210
180
+ },
181
+ {
182
+ "epoch": 2.3675675675675674,
183
+ "grad_norm": 46.840301513671875,
184
+ "learning_rate": 4.780876494023905e-06,
185
+ "loss": 1.2476821899414063,
186
+ "step": 220
187
+ },
188
+ {
189
+ "epoch": 2.4756756756756757,
190
+ "grad_norm": 19.01693344116211,
191
+ "learning_rate": 3.98406374501992e-06,
192
+ "loss": 1.1008167266845703,
193
+ "step": 230
194
+ },
195
+ {
196
+ "epoch": 2.583783783783784,
197
+ "grad_norm": 36.20149230957031,
198
+ "learning_rate": 3.1872509960159366e-06,
199
+ "loss": 1.090567398071289,
200
+ "step": 240
201
+ },
202
+ {
203
+ "epoch": 2.691891891891892,
204
+ "grad_norm": 64.47439575195312,
205
+ "learning_rate": 2.3904382470119524e-06,
206
+ "loss": 0.8782524108886719,
207
+ "step": 250
208
+ },
209
+ {
210
+ "epoch": 2.8,
211
+ "grad_norm": 28.88818359375,
212
+ "learning_rate": 1.5936254980079683e-06,
213
+ "loss": 0.7962211608886719,
214
+ "step": 260
215
+ },
216
+ {
217
+ "epoch": 2.908108108108108,
218
+ "grad_norm": 19.417667388916016,
219
+ "learning_rate": 7.968127490039841e-07,
220
+ "loss": 1.3786026000976563,
221
+ "step": 270
222
+ },
223
+ {
224
+ "epoch": 3.0,
225
+ "eval_accuracy": 0.645,
226
+ "eval_fallacy_f1": 0.6697674418604651,
227
+ "eval_loss": 0.7070519328117371,
228
+ "eval_macro_f1": 0.6429918290383407,
229
+ "eval_runtime": 1.4407,
230
+ "eval_samples_per_second": 138.819,
231
+ "eval_steps_per_second": 9.023,
232
+ "step": 279
233
+ }
234
+ ],
235
+ "logging_steps": 10,
236
+ "max_steps": 279,
237
+ "num_input_tokens_seen": 0,
238
+ "num_train_epochs": 3,
239
+ "save_steps": 500,
240
+ "stateful_callbacks": {
241
+ "TrainerControl": {
242
+ "args": {
243
+ "should_epoch_stop": false,
244
+ "should_evaluate": false,
245
+ "should_log": false,
246
+ "should_save": true,
247
+ "should_training_stop": true
248
+ },
249
+ "attributes": {}
250
+ }
251
+ },
252
+ "total_flos": 506948819509104.0,
253
+ "train_batch_size": 4,
254
+ "trial_name": null,
255
+ "trial_params": null
256
+ }
checkpoint-279/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b57361bd3df136d106452d8a34502b0e6fe2a74596feef14daa5dc9d043af2d9
3
+ size 5393