PhaaNe commited on
Commit
9e5f341
·
verified ·
1 Parent(s): 31deea9

Add files using upload-large-folder tool

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-1274/special_tokens_map.json +15 -0
  2. ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-1274/tokenizer.json +0 -0
  3. ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-1274/tokenizer_config.json +56 -0
  4. ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-1274/trainer_state.json +470 -0
  5. ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-728/special_tokens_map.json +15 -0
  6. ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-728/tokenizer.json +0 -0
  7. ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-728/tokenizer_config.json +56 -0
  8. ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-728/trainer_state.json +290 -0
  9. ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-1710/special_tokens_map.json +15 -0
  10. ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-1710/tokenizer.json +0 -0
  11. ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-1710/tokenizer_config.json +56 -0
  12. ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-1710/trainer_state.json +629 -0
  13. ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-2223/special_tokens_map.json +15 -0
  14. ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-2223/tokenizer.json +0 -0
  15. ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-2223/tokenizer_config.json +56 -0
  16. ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-2223/trainer_state.json +802 -0
  17. ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-1267/special_tokens_map.json +15 -0
  18. ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-1267/tokenizer.json +0 -0
  19. ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-1267/tokenizer_config.json +56 -0
  20. ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-1267/trainer_state.json +470 -0
  21. ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-724/special_tokens_map.json +15 -0
  22. ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-724/tokenizer.json +0 -0
  23. ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-724/tokenizer_config.json +56 -0
  24. ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-724/trainer_state.json +283 -0
  25. ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-1281/special_tokens_map.json +15 -0
  26. ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-1281/tokenizer.json +0 -0
  27. ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-1281/tokenizer_config.json +56 -0
  28. ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-1281/trainer_state.json +477 -0
  29. ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-732/special_tokens_map.json +15 -0
  30. ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-732/tokenizer.json +0 -0
  31. ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-732/tokenizer_config.json +56 -0
  32. ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-732/trainer_state.json +290 -0
  33. ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1408/special_tokens_map.json +15 -0
  34. ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1408/tokenizer.json +0 -0
  35. ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1408/tokenizer_config.json +56 -0
  36. ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1408/trainer_state.json +523 -0
  37. ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1936/special_tokens_map.json +15 -0
  38. ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1936/tokenizer.json +0 -0
  39. ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1936/tokenizer_config.json +56 -0
  40. ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1936/trainer_state.json +703 -0
  41. ood-vifactcheck-visobert-source-source-05-s44-20260831T033516Z-59e46264/trainer/checkpoint-1080/special_tokens_map.json +15 -0
  42. ood-vifactcheck-visobert-source-source-05-s44-20260831T033516Z-59e46264/trainer/checkpoint-1080/tokenizer.json +0 -0
  43. ood-vifactcheck-visobert-source-source-05-s44-20260831T033516Z-59e46264/trainer/checkpoint-1080/tokenizer_config.json +56 -0
  44. ood-vifactcheck-visobert-source-source-06-s43-20260831T021725Z-1a169064/trainer/checkpoint-2590/special_tokens_map.json +15 -0
  45. ood-vifactcheck-visobert-source-source-06-s43-20260831T021725Z-1a169064/trainer/checkpoint-2590/tokenizer.json +0 -0
  46. ood-vifactcheck-visobert-source-source-06-s43-20260831T021725Z-1a169064/trainer/checkpoint-2590/tokenizer_config.json +56 -0
  47. ood-vifactcheck-visobert-source-source-06-s43-20260831T021725Z-1a169064/trainer/checkpoint-2590/trainer_state.json +918 -0
  48. ood-vifactcheck-visobert-source-source-06-s44-20260831T034444Z-59e46264/trainer/checkpoint-1850/special_tokens_map.json +15 -0
  49. ood-vifactcheck-visobert-source-source-06-s44-20260831T034444Z-59e46264/trainer/checkpoint-1850/tokenizer.json +0 -0
  50. ood-vifactcheck-visobert-source-source-06-s44-20260831T034444Z-59e46264/trainer/checkpoint-1850/tokenizer_config.json +56 -0
ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-1274/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-1274/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-1274/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-1274/trainer_state.json ADDED
@@ -0,0 +1,470 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 728,
3
+ "best_metric": 0.5657553897414734,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-728",
5
+ "epoch": 7.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1274,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13793103448275862,
14
+ "grad_norm": 20.15477180480957,
15
+ "learning_rate": 1.3186813186813187e-06,
16
+ "loss": 4.7668,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27586206896551724,
21
+ "grad_norm": 22.066551208496094,
22
+ "learning_rate": 2.6923076923076923e-06,
23
+ "loss": 4.5599,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.41379310344827586,
28
+ "grad_norm": 27.23032569885254,
29
+ "learning_rate": 4.065934065934066e-06,
30
+ "loss": 4.4862,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5517241379310345,
35
+ "grad_norm": 24.71661949157715,
36
+ "learning_rate": 5.43956043956044e-06,
37
+ "loss": 4.4246,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6896551724137931,
42
+ "grad_norm": 39.3474006652832,
43
+ "learning_rate": 6.813186813186814e-06,
44
+ "loss": 4.4258,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8275862068965517,
49
+ "grad_norm": 28.388126373291016,
50
+ "learning_rate": 8.186813186813188e-06,
51
+ "loss": 4.3926,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9655172413793104,
56
+ "grad_norm": 28.285808563232422,
57
+ "learning_rate": 9.560439560439562e-06,
58
+ "loss": 4.3772,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.44019138755980863,
64
+ "eval_loss": 1.0537093877792358,
65
+ "eval_macro_f1": 0.4034879274882082,
66
+ "eval_runtime": 1.2783,
67
+ "eval_samples_per_second": 490.486,
68
+ "eval_steps_per_second": 31.291,
69
+ "eval_weighted_f1": 0.40588775213089445,
70
+ "step": 182
71
+ },
72
+ {
73
+ "epoch": 1.0993103448275863,
74
+ "grad_norm": 16.964529037475586,
75
+ "learning_rate": 1.0934065934065933e-05,
76
+ "loss": 4.2272,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.237241379310345,
81
+ "grad_norm": 19.617334365844727,
82
+ "learning_rate": 1.230769230769231e-05,
83
+ "loss": 4.3645,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3751724137931034,
88
+ "grad_norm": 27.136455535888672,
89
+ "learning_rate": 1.3681318681318683e-05,
90
+ "loss": 4.2321,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.513103448275862,
95
+ "grad_norm": 28.211318969726562,
96
+ "learning_rate": 1.5054945054945056e-05,
97
+ "loss": 4.2452,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6510344827586207,
102
+ "grad_norm": 40.97351837158203,
103
+ "learning_rate": 1.642857142857143e-05,
104
+ "loss": 4.2935,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7889655172413792,
109
+ "grad_norm": 35.05915069580078,
110
+ "learning_rate": 1.78021978021978e-05,
111
+ "loss": 4.233,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.926896551724138,
116
+ "grad_norm": 38.13852310180664,
117
+ "learning_rate": 1.9175824175824178e-05,
118
+ "loss": 4.3812,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.4784688995215311,
124
+ "eval_loss": 1.0289137363433838,
125
+ "eval_macro_f1": 0.47078131767738834,
126
+ "eval_runtime": 1.345,
127
+ "eval_samples_per_second": 466.162,
128
+ "eval_steps_per_second": 29.739,
129
+ "eval_weighted_f1": 0.46964185719332235,
130
+ "step": 364
131
+ },
132
+ {
133
+ "epoch": 2.0606896551724136,
134
+ "grad_norm": 17.627437591552734,
135
+ "learning_rate": 1.993894993894994e-05,
136
+ "loss": 3.9478,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.1986206896551725,
141
+ "grad_norm": 36.87620162963867,
142
+ "learning_rate": 1.9786324786324788e-05,
143
+ "loss": 3.8288,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.336551724137931,
148
+ "grad_norm": 16.316394805908203,
149
+ "learning_rate": 1.9633699633699634e-05,
150
+ "loss": 3.8802,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.47448275862069,
155
+ "grad_norm": 21.271818161010742,
156
+ "learning_rate": 1.948107448107448e-05,
157
+ "loss": 3.8315,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.6124137931034483,
162
+ "grad_norm": 62.814430236816406,
163
+ "learning_rate": 1.932844932844933e-05,
164
+ "loss": 3.6702,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.750344827586207,
169
+ "grad_norm": 51.539852142333984,
170
+ "learning_rate": 1.9175824175824178e-05,
171
+ "loss": 3.8222,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.8882758620689657,
176
+ "grad_norm": 27.321670532226562,
177
+ "learning_rate": 1.9023199023199025e-05,
178
+ "loss": 3.6069,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5406698564593302,
184
+ "eval_loss": 0.9773996472358704,
185
+ "eval_macro_f1": 0.5348699968075374,
186
+ "eval_runtime": 1.4242,
187
+ "eval_samples_per_second": 440.262,
188
+ "eval_steps_per_second": 28.087,
189
+ "eval_weighted_f1": 0.5365183080948986,
190
+ "step": 546
191
+ },
192
+ {
193
+ "epoch": 3.0220689655172412,
194
+ "grad_norm": 28.224672317504883,
195
+ "learning_rate": 1.887057387057387e-05,
196
+ "loss": 3.3666,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.16,
201
+ "grad_norm": 24.172056198120117,
202
+ "learning_rate": 1.8717948717948718e-05,
203
+ "loss": 2.8604,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.2979310344827586,
208
+ "grad_norm": 35.04057693481445,
209
+ "learning_rate": 1.8565323565323568e-05,
210
+ "loss": 2.9409,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.435862068965517,
215
+ "grad_norm": 73.03572845458984,
216
+ "learning_rate": 1.8412698412698415e-05,
217
+ "loss": 2.9225,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.573793103448276,
222
+ "grad_norm": 41.814170837402344,
223
+ "learning_rate": 1.826007326007326e-05,
224
+ "loss": 2.9849,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.7117241379310344,
229
+ "grad_norm": 50.381431579589844,
230
+ "learning_rate": 1.8107448107448108e-05,
231
+ "loss": 2.9611,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.849655172413793,
236
+ "grad_norm": 25.989524841308594,
237
+ "learning_rate": 1.7954822954822958e-05,
238
+ "loss": 2.9334,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.9875862068965517,
243
+ "grad_norm": 34.98115921020508,
244
+ "learning_rate": 1.78021978021978e-05,
245
+ "loss": 2.721,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 4.0,
250
+ "eval_accuracy": 0.5677830940988836,
251
+ "eval_loss": 0.9596757888793945,
252
+ "eval_macro_f1": 0.5657553897414734,
253
+ "eval_runtime": 1.2648,
254
+ "eval_samples_per_second": 495.726,
255
+ "eval_steps_per_second": 31.625,
256
+ "eval_weighted_f1": 0.5652192754768104,
257
+ "step": 728
258
+ },
259
+ {
260
+ "epoch": 4.121379310344827,
261
+ "grad_norm": 28.530105590820312,
262
+ "learning_rate": 1.764957264957265e-05,
263
+ "loss": 1.893,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.259310344827586,
268
+ "grad_norm": 40.174560546875,
269
+ "learning_rate": 1.74969474969475e-05,
270
+ "loss": 1.7545,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.397241379310345,
275
+ "grad_norm": 50.635433197021484,
276
+ "learning_rate": 1.7344322344322345e-05,
277
+ "loss": 1.8194,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.535172413793103,
282
+ "grad_norm": 23.626415252685547,
283
+ "learning_rate": 1.7191697191697195e-05,
284
+ "loss": 2.085,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.673103448275862,
289
+ "grad_norm": 74.52599334716797,
290
+ "learning_rate": 1.703907203907204e-05,
291
+ "loss": 2.0802,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.811034482758621,
296
+ "grad_norm": 39.6097526550293,
297
+ "learning_rate": 1.688644688644689e-05,
298
+ "loss": 1.9657,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.94896551724138,
303
+ "grad_norm": 66.90764617919922,
304
+ "learning_rate": 1.6733821733821735e-05,
305
+ "loss": 1.9481,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 5.0,
310
+ "eval_accuracy": 0.5550239234449761,
311
+ "eval_loss": 1.1992511749267578,
312
+ "eval_macro_f1": 0.5569820230233465,
313
+ "eval_runtime": 1.3484,
314
+ "eval_samples_per_second": 465.01,
315
+ "eval_steps_per_second": 29.666,
316
+ "eval_weighted_f1": 0.5576508629062836,
317
+ "step": 910
318
+ },
319
+ {
320
+ "epoch": 5.082758620689655,
321
+ "grad_norm": 55.95634841918945,
322
+ "learning_rate": 1.6581196581196585e-05,
323
+ "loss": 1.4087,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.220689655172414,
328
+ "grad_norm": 80.93014526367188,
329
+ "learning_rate": 1.642857142857143e-05,
330
+ "loss": 1.1284,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.358620689655172,
335
+ "grad_norm": 44.51694107055664,
336
+ "learning_rate": 1.627594627594628e-05,
337
+ "loss": 1.1338,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.496551724137931,
342
+ "grad_norm": 34.489871978759766,
343
+ "learning_rate": 1.6123321123321125e-05,
344
+ "loss": 1.0865,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.63448275862069,
349
+ "grad_norm": 77.27289581298828,
350
+ "learning_rate": 1.5970695970695972e-05,
351
+ "loss": 1.1695,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.772413793103448,
356
+ "grad_norm": 57.632972717285156,
357
+ "learning_rate": 1.581807081807082e-05,
358
+ "loss": 1.3647,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.910344827586207,
363
+ "grad_norm": 62.253944396972656,
364
+ "learning_rate": 1.5665445665445665e-05,
365
+ "loss": 1.0863,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 6.0,
370
+ "eval_accuracy": 0.5406698564593302,
371
+ "eval_loss": 1.6473782062530518,
372
+ "eval_macro_f1": 0.5409866941843686,
373
+ "eval_runtime": 1.4929,
374
+ "eval_samples_per_second": 419.976,
375
+ "eval_steps_per_second": 26.793,
376
+ "eval_weighted_f1": 0.5422284750411489,
377
+ "step": 1092
378
+ },
379
+ {
380
+ "epoch": 6.0441379310344825,
381
+ "grad_norm": 46.98276138305664,
382
+ "learning_rate": 1.5512820512820516e-05,
383
+ "loss": 0.8465,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.182068965517241,
388
+ "grad_norm": 12.458864212036133,
389
+ "learning_rate": 1.5360195360195362e-05,
390
+ "loss": 0.5447,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.32,
395
+ "grad_norm": 61.23617935180664,
396
+ "learning_rate": 1.5207570207570207e-05,
397
+ "loss": 0.5712,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.457931034482758,
402
+ "grad_norm": 70.76924133300781,
403
+ "learning_rate": 1.5054945054945056e-05,
404
+ "loss": 0.7002,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.595862068965517,
409
+ "grad_norm": 56.29345703125,
410
+ "learning_rate": 1.4902319902319904e-05,
411
+ "loss": 0.7975,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.733793103448276,
416
+ "grad_norm": 25.182703018188477,
417
+ "learning_rate": 1.4749694749694752e-05,
418
+ "loss": 0.7455,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.871724137931034,
423
+ "grad_norm": 31.513938903808594,
424
+ "learning_rate": 1.4597069597069597e-05,
425
+ "loss": 0.779,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 7.0,
430
+ "eval_accuracy": 0.5598086124401914,
431
+ "eval_loss": 1.8176171779632568,
432
+ "eval_macro_f1": 0.5594915900632554,
433
+ "eval_runtime": 1.3247,
434
+ "eval_samples_per_second": 473.316,
435
+ "eval_steps_per_second": 30.196,
436
+ "eval_weighted_f1": 0.5588479559220413,
437
+ "step": 1274
438
+ }
439
+ ],
440
+ "logging_steps": 25,
441
+ "max_steps": 3640,
442
+ "num_input_tokens_seen": 0,
443
+ "num_train_epochs": 20,
444
+ "save_steps": 500,
445
+ "stateful_callbacks": {
446
+ "EarlyStoppingCallback": {
447
+ "args": {
448
+ "early_stopping_patience": 3,
449
+ "early_stopping_threshold": 0.0
450
+ },
451
+ "attributes": {
452
+ "early_stopping_patience_counter": 3
453
+ }
454
+ },
455
+ "TrainerControl": {
456
+ "args": {
457
+ "should_epoch_stop": false,
458
+ "should_evaluate": false,
459
+ "should_log": false,
460
+ "should_save": true,
461
+ "should_training_stop": true
462
+ },
463
+ "attributes": {}
464
+ }
465
+ },
466
+ "total_flos": 0.0,
467
+ "train_batch_size": 8,
468
+ "trial_name": null,
469
+ "trial_params": null
470
+ }
ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-728/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-728/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-728/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-728/trainer_state.json ADDED
@@ -0,0 +1,290 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 728,
3
+ "best_metric": 0.5657553897414734,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-00-s44-20260831T024900Z-59e46264/trainer/checkpoint-728",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 728,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13793103448275862,
14
+ "grad_norm": 20.15477180480957,
15
+ "learning_rate": 1.3186813186813187e-06,
16
+ "loss": 4.7668,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27586206896551724,
21
+ "grad_norm": 22.066551208496094,
22
+ "learning_rate": 2.6923076923076923e-06,
23
+ "loss": 4.5599,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.41379310344827586,
28
+ "grad_norm": 27.23032569885254,
29
+ "learning_rate": 4.065934065934066e-06,
30
+ "loss": 4.4862,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5517241379310345,
35
+ "grad_norm": 24.71661949157715,
36
+ "learning_rate": 5.43956043956044e-06,
37
+ "loss": 4.4246,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6896551724137931,
42
+ "grad_norm": 39.3474006652832,
43
+ "learning_rate": 6.813186813186814e-06,
44
+ "loss": 4.4258,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8275862068965517,
49
+ "grad_norm": 28.388126373291016,
50
+ "learning_rate": 8.186813186813188e-06,
51
+ "loss": 4.3926,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9655172413793104,
56
+ "grad_norm": 28.285808563232422,
57
+ "learning_rate": 9.560439560439562e-06,
58
+ "loss": 4.3772,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.44019138755980863,
64
+ "eval_loss": 1.0537093877792358,
65
+ "eval_macro_f1": 0.4034879274882082,
66
+ "eval_runtime": 1.2783,
67
+ "eval_samples_per_second": 490.486,
68
+ "eval_steps_per_second": 31.291,
69
+ "eval_weighted_f1": 0.40588775213089445,
70
+ "step": 182
71
+ },
72
+ {
73
+ "epoch": 1.0993103448275863,
74
+ "grad_norm": 16.964529037475586,
75
+ "learning_rate": 1.0934065934065933e-05,
76
+ "loss": 4.2272,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.237241379310345,
81
+ "grad_norm": 19.617334365844727,
82
+ "learning_rate": 1.230769230769231e-05,
83
+ "loss": 4.3645,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3751724137931034,
88
+ "grad_norm": 27.136455535888672,
89
+ "learning_rate": 1.3681318681318683e-05,
90
+ "loss": 4.2321,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.513103448275862,
95
+ "grad_norm": 28.211318969726562,
96
+ "learning_rate": 1.5054945054945056e-05,
97
+ "loss": 4.2452,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6510344827586207,
102
+ "grad_norm": 40.97351837158203,
103
+ "learning_rate": 1.642857142857143e-05,
104
+ "loss": 4.2935,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7889655172413792,
109
+ "grad_norm": 35.05915069580078,
110
+ "learning_rate": 1.78021978021978e-05,
111
+ "loss": 4.233,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.926896551724138,
116
+ "grad_norm": 38.13852310180664,
117
+ "learning_rate": 1.9175824175824178e-05,
118
+ "loss": 4.3812,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.4784688995215311,
124
+ "eval_loss": 1.0289137363433838,
125
+ "eval_macro_f1": 0.47078131767738834,
126
+ "eval_runtime": 1.345,
127
+ "eval_samples_per_second": 466.162,
128
+ "eval_steps_per_second": 29.739,
129
+ "eval_weighted_f1": 0.46964185719332235,
130
+ "step": 364
131
+ },
132
+ {
133
+ "epoch": 2.0606896551724136,
134
+ "grad_norm": 17.627437591552734,
135
+ "learning_rate": 1.993894993894994e-05,
136
+ "loss": 3.9478,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.1986206896551725,
141
+ "grad_norm": 36.87620162963867,
142
+ "learning_rate": 1.9786324786324788e-05,
143
+ "loss": 3.8288,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.336551724137931,
148
+ "grad_norm": 16.316394805908203,
149
+ "learning_rate": 1.9633699633699634e-05,
150
+ "loss": 3.8802,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.47448275862069,
155
+ "grad_norm": 21.271818161010742,
156
+ "learning_rate": 1.948107448107448e-05,
157
+ "loss": 3.8315,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.6124137931034483,
162
+ "grad_norm": 62.814430236816406,
163
+ "learning_rate": 1.932844932844933e-05,
164
+ "loss": 3.6702,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.750344827586207,
169
+ "grad_norm": 51.539852142333984,
170
+ "learning_rate": 1.9175824175824178e-05,
171
+ "loss": 3.8222,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.8882758620689657,
176
+ "grad_norm": 27.321670532226562,
177
+ "learning_rate": 1.9023199023199025e-05,
178
+ "loss": 3.6069,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5406698564593302,
184
+ "eval_loss": 0.9773996472358704,
185
+ "eval_macro_f1": 0.5348699968075374,
186
+ "eval_runtime": 1.4242,
187
+ "eval_samples_per_second": 440.262,
188
+ "eval_steps_per_second": 28.087,
189
+ "eval_weighted_f1": 0.5365183080948986,
190
+ "step": 546
191
+ },
192
+ {
193
+ "epoch": 3.0220689655172412,
194
+ "grad_norm": 28.224672317504883,
195
+ "learning_rate": 1.887057387057387e-05,
196
+ "loss": 3.3666,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.16,
201
+ "grad_norm": 24.172056198120117,
202
+ "learning_rate": 1.8717948717948718e-05,
203
+ "loss": 2.8604,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.2979310344827586,
208
+ "grad_norm": 35.04057693481445,
209
+ "learning_rate": 1.8565323565323568e-05,
210
+ "loss": 2.9409,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.435862068965517,
215
+ "grad_norm": 73.03572845458984,
216
+ "learning_rate": 1.8412698412698415e-05,
217
+ "loss": 2.9225,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.573793103448276,
222
+ "grad_norm": 41.814170837402344,
223
+ "learning_rate": 1.826007326007326e-05,
224
+ "loss": 2.9849,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.7117241379310344,
229
+ "grad_norm": 50.381431579589844,
230
+ "learning_rate": 1.8107448107448108e-05,
231
+ "loss": 2.9611,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.849655172413793,
236
+ "grad_norm": 25.989524841308594,
237
+ "learning_rate": 1.7954822954822958e-05,
238
+ "loss": 2.9334,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.9875862068965517,
243
+ "grad_norm": 34.98115921020508,
244
+ "learning_rate": 1.78021978021978e-05,
245
+ "loss": 2.721,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 4.0,
250
+ "eval_accuracy": 0.5677830940988836,
251
+ "eval_loss": 0.9596757888793945,
252
+ "eval_macro_f1": 0.5657553897414734,
253
+ "eval_runtime": 1.2648,
254
+ "eval_samples_per_second": 495.726,
255
+ "eval_steps_per_second": 31.625,
256
+ "eval_weighted_f1": 0.5652192754768104,
257
+ "step": 728
258
+ }
259
+ ],
260
+ "logging_steps": 25,
261
+ "max_steps": 3640,
262
+ "num_input_tokens_seen": 0,
263
+ "num_train_epochs": 20,
264
+ "save_steps": 500,
265
+ "stateful_callbacks": {
266
+ "EarlyStoppingCallback": {
267
+ "args": {
268
+ "early_stopping_patience": 3,
269
+ "early_stopping_threshold": 0.0
270
+ },
271
+ "attributes": {
272
+ "early_stopping_patience_counter": 0
273
+ }
274
+ },
275
+ "TrainerControl": {
276
+ "args": {
277
+ "should_epoch_stop": false,
278
+ "should_evaluate": false,
279
+ "should_log": false,
280
+ "should_save": true,
281
+ "should_training_stop": false
282
+ },
283
+ "attributes": {}
284
+ }
285
+ },
286
+ "total_flos": 0.0,
287
+ "train_batch_size": 8,
288
+ "trial_name": null,
289
+ "trial_params": null
290
+ }
ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-1710/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-1710/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-1710/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-1710/trainer_state.json ADDED
@@ -0,0 +1,629 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1710,
3
+ "best_metric": 0.580957527780125,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-1710",
5
+ "epoch": 10.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1710,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.14684287812041116,
14
+ "grad_norm": 19.353429794311523,
15
+ "learning_rate": 1.4035087719298246e-06,
16
+ "loss": 4.7879,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.2936857562408223,
21
+ "grad_norm": 30.09554100036621,
22
+ "learning_rate": 2.865497076023392e-06,
23
+ "loss": 4.549,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.44052863436123346,
28
+ "grad_norm": 31.329397201538086,
29
+ "learning_rate": 4.3274853801169596e-06,
30
+ "loss": 4.4843,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5873715124816447,
35
+ "grad_norm": 21.417451858520508,
36
+ "learning_rate": 5.789473684210527e-06,
37
+ "loss": 4.494,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.7342143906020558,
42
+ "grad_norm": 19.82093620300293,
43
+ "learning_rate": 7.251461988304094e-06,
44
+ "loss": 4.4666,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8810572687224669,
49
+ "grad_norm": 21.48238754272461,
50
+ "learning_rate": 8.713450292397661e-06,
51
+ "loss": 4.5115,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 1.0,
56
+ "eval_accuracy": 0.400355871886121,
57
+ "eval_loss": 1.0705939531326294,
58
+ "eval_macro_f1": 0.38519230084886114,
59
+ "eval_runtime": 1.2555,
60
+ "eval_samples_per_second": 447.648,
61
+ "eval_steps_per_second": 28.675,
62
+ "eval_weighted_f1": 0.38659017191156697,
63
+ "step": 171
64
+ },
65
+ {
66
+ "epoch": 1.0234948604992657,
67
+ "grad_norm": 18.355037689208984,
68
+ "learning_rate": 1.017543859649123e-05,
69
+ "loss": 4.2445,
70
+ "step": 175
71
+ },
72
+ {
73
+ "epoch": 1.170337738619677,
74
+ "grad_norm": 28.72773551940918,
75
+ "learning_rate": 1.1637426900584797e-05,
76
+ "loss": 4.2874,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.3171806167400881,
81
+ "grad_norm": 25.883697509765625,
82
+ "learning_rate": 1.3099415204678362e-05,
83
+ "loss": 4.2557,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.4640234948604993,
88
+ "grad_norm": 40.167694091796875,
89
+ "learning_rate": 1.4561403508771931e-05,
90
+ "loss": 4.2551,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.6108663729809103,
95
+ "grad_norm": 21.89159393310547,
96
+ "learning_rate": 1.60233918128655e-05,
97
+ "loss": 4.3497,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.7577092511013217,
102
+ "grad_norm": 26.441078186035156,
103
+ "learning_rate": 1.7485380116959064e-05,
104
+ "loss": 4.2409,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.9045521292217327,
109
+ "grad_norm": 22.421125411987305,
110
+ "learning_rate": 1.894736842105263e-05,
111
+ "loss": 4.2588,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 2.0,
116
+ "eval_accuracy": 0.46441281138790036,
117
+ "eval_loss": 1.044918417930603,
118
+ "eval_macro_f1": 0.46202289516495454,
119
+ "eval_runtime": 1.2659,
120
+ "eval_samples_per_second": 443.958,
121
+ "eval_steps_per_second": 28.439,
122
+ "eval_weighted_f1": 0.4614509909348169,
123
+ "step": 342
124
+ },
125
+ {
126
+ "epoch": 2.0469897209985315,
127
+ "grad_norm": 43.260337829589844,
128
+ "learning_rate": 1.99545159194282e-05,
129
+ "loss": 4.0338,
130
+ "step": 350
131
+ },
132
+ {
133
+ "epoch": 2.193832599118943,
134
+ "grad_norm": 32.08805465698242,
135
+ "learning_rate": 1.9792072774528916e-05,
136
+ "loss": 3.984,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.340675477239354,
141
+ "grad_norm": 26.372203826904297,
142
+ "learning_rate": 1.962962962962963e-05,
143
+ "loss": 3.8542,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.4875183553597653,
148
+ "grad_norm": 20.757434844970703,
149
+ "learning_rate": 1.9467186484730346e-05,
150
+ "loss": 3.7754,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.6343612334801763,
155
+ "grad_norm": 45.604854583740234,
156
+ "learning_rate": 1.930474333983106e-05,
157
+ "loss": 3.897,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.7812041116005872,
162
+ "grad_norm": 47.26052474975586,
163
+ "learning_rate": 1.9142300194931776e-05,
164
+ "loss": 3.7785,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.9280469897209986,
169
+ "grad_norm": 22.94167709350586,
170
+ "learning_rate": 1.897985705003249e-05,
171
+ "loss": 3.6098,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 3.0,
176
+ "eval_accuracy": 0.5088967971530249,
177
+ "eval_loss": 1.0127676725387573,
178
+ "eval_macro_f1": 0.5002107603877864,
179
+ "eval_runtime": 1.2283,
180
+ "eval_samples_per_second": 457.552,
181
+ "eval_steps_per_second": 29.309,
182
+ "eval_weighted_f1": 0.49992379157923295,
183
+ "step": 513
184
+ },
185
+ {
186
+ "epoch": 3.0704845814977975,
187
+ "grad_norm": 25.306499481201172,
188
+ "learning_rate": 1.8817413905133203e-05,
189
+ "loss": 3.1486,
190
+ "step": 525
191
+ },
192
+ {
193
+ "epoch": 3.2173274596182084,
194
+ "grad_norm": 51.927101135253906,
195
+ "learning_rate": 1.8654970760233918e-05,
196
+ "loss": 3.0286,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.36417033773862,
201
+ "grad_norm": 23.04163360595703,
202
+ "learning_rate": 1.8492527615334633e-05,
203
+ "loss": 3.0281,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.511013215859031,
208
+ "grad_norm": 31.153520584106445,
209
+ "learning_rate": 1.833008447043535e-05,
210
+ "loss": 2.888,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.657856093979442,
215
+ "grad_norm": 42.227622985839844,
216
+ "learning_rate": 1.8167641325536064e-05,
217
+ "loss": 2.8699,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.804698972099853,
222
+ "grad_norm": 36.88457489013672,
223
+ "learning_rate": 1.800519818063678e-05,
224
+ "loss": 2.9566,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.951541850220264,
229
+ "grad_norm": 30.99384117126465,
230
+ "learning_rate": 1.7842755035737494e-05,
231
+ "loss": 2.9823,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 4.0,
236
+ "eval_accuracy": 0.5658362989323843,
237
+ "eval_loss": 1.009939193725586,
238
+ "eval_macro_f1": 0.5582042983984731,
239
+ "eval_runtime": 1.1721,
240
+ "eval_samples_per_second": 479.484,
241
+ "eval_steps_per_second": 30.714,
242
+ "eval_weighted_f1": 0.5591711028839177,
243
+ "step": 684
244
+ },
245
+ {
246
+ "epoch": 4.093979441997063,
247
+ "grad_norm": 53.3286018371582,
248
+ "learning_rate": 1.768031189083821e-05,
249
+ "loss": 2.2471,
250
+ "step": 700
251
+ },
252
+ {
253
+ "epoch": 4.240822320117474,
254
+ "grad_norm": 62.07929992675781,
255
+ "learning_rate": 1.751786874593892e-05,
256
+ "loss": 1.8491,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.387665198237886,
261
+ "grad_norm": 46.8976936340332,
262
+ "learning_rate": 1.7355425601039636e-05,
263
+ "loss": 1.9902,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.534508076358296,
268
+ "grad_norm": 54.493446350097656,
269
+ "learning_rate": 1.719298245614035e-05,
270
+ "loss": 1.8621,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.681350954478708,
275
+ "grad_norm": 49.192718505859375,
276
+ "learning_rate": 1.7030539311241066e-05,
277
+ "loss": 1.8949,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.828193832599119,
282
+ "grad_norm": 68.6122817993164,
283
+ "learning_rate": 1.686809616634178e-05,
284
+ "loss": 1.9152,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.975036710719531,
289
+ "grad_norm": 64.56343078613281,
290
+ "learning_rate": 1.6705653021442497e-05,
291
+ "loss": 1.921,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 5.0,
296
+ "eval_accuracy": 0.5640569395017794,
297
+ "eval_loss": 1.111932396888733,
298
+ "eval_macro_f1": 0.5636776401175632,
299
+ "eval_runtime": 1.1624,
300
+ "eval_samples_per_second": 483.47,
301
+ "eval_steps_per_second": 30.97,
302
+ "eval_weighted_f1": 0.5636503993319255,
303
+ "step": 855
304
+ },
305
+ {
306
+ "epoch": 5.117474302496329,
307
+ "grad_norm": 122.34413146972656,
308
+ "learning_rate": 1.654320987654321e-05,
309
+ "loss": 1.3079,
310
+ "step": 875
311
+ },
312
+ {
313
+ "epoch": 5.26431718061674,
314
+ "grad_norm": 83.37774658203125,
315
+ "learning_rate": 1.6380766731643927e-05,
316
+ "loss": 1.0141,
317
+ "step": 900
318
+ },
319
+ {
320
+ "epoch": 5.411160058737151,
321
+ "grad_norm": 61.428306579589844,
322
+ "learning_rate": 1.621832358674464e-05,
323
+ "loss": 1.1481,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.558002936857562,
328
+ "grad_norm": 102.04503631591797,
329
+ "learning_rate": 1.6055880441845354e-05,
330
+ "loss": 1.0319,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.704845814977974,
335
+ "grad_norm": 35.03520202636719,
336
+ "learning_rate": 1.589343729694607e-05,
337
+ "loss": 1.1584,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.851688693098385,
342
+ "grad_norm": 43.3909797668457,
343
+ "learning_rate": 1.5730994152046787e-05,
344
+ "loss": 1.1633,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.998531571218796,
349
+ "grad_norm": 85.28243255615234,
350
+ "learning_rate": 1.55685510071475e-05,
351
+ "loss": 1.3961,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 6.0,
356
+ "eval_accuracy": 0.505338078291815,
357
+ "eval_loss": 1.8369455337524414,
358
+ "eval_macro_f1": 0.4815974670718708,
359
+ "eval_runtime": 1.1419,
360
+ "eval_samples_per_second": 492.162,
361
+ "eval_steps_per_second": 31.526,
362
+ "eval_weighted_f1": 0.4840301003544507,
363
+ "step": 1026
364
+ },
365
+ {
366
+ "epoch": 6.140969162995595,
367
+ "grad_norm": 63.258750915527344,
368
+ "learning_rate": 1.5406107862248214e-05,
369
+ "loss": 0.6735,
370
+ "step": 1050
371
+ },
372
+ {
373
+ "epoch": 6.2878120411160054,
374
+ "grad_norm": 35.072357177734375,
375
+ "learning_rate": 1.524366471734893e-05,
376
+ "loss": 0.7072,
377
+ "step": 1075
378
+ },
379
+ {
380
+ "epoch": 6.434654919236417,
381
+ "grad_norm": 29.91564178466797,
382
+ "learning_rate": 1.5081221572449645e-05,
383
+ "loss": 0.6603,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.581497797356828,
388
+ "grad_norm": 88.57781982421875,
389
+ "learning_rate": 1.4918778427550358e-05,
390
+ "loss": 0.5712,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.72834067547724,
395
+ "grad_norm": 47.00954055786133,
396
+ "learning_rate": 1.4756335282651073e-05,
397
+ "loss": 0.7554,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.87518355359765,
402
+ "grad_norm": 29.895748138427734,
403
+ "learning_rate": 1.4593892137751788e-05,
404
+ "loss": 0.6925,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 7.0,
409
+ "eval_accuracy": 0.5693950177935944,
410
+ "eval_loss": 1.8101662397384644,
411
+ "eval_macro_f1": 0.5657155822901679,
412
+ "eval_runtime": 1.171,
413
+ "eval_samples_per_second": 479.939,
414
+ "eval_steps_per_second": 30.743,
415
+ "eval_weighted_f1": 0.5661096303435228,
416
+ "step": 1197
417
+ },
418
+ {
419
+ "epoch": 7.0176211453744495,
420
+ "grad_norm": 74.3259506225586,
421
+ "learning_rate": 1.4431448992852503e-05,
422
+ "loss": 0.705,
423
+ "step": 1200
424
+ },
425
+ {
426
+ "epoch": 7.164464023494861,
427
+ "grad_norm": 198.61756896972656,
428
+ "learning_rate": 1.4269005847953217e-05,
429
+ "loss": 0.4091,
430
+ "step": 1225
431
+ },
432
+ {
433
+ "epoch": 7.311306901615271,
434
+ "grad_norm": 67.14990234375,
435
+ "learning_rate": 1.4106562703053932e-05,
436
+ "loss": 0.3649,
437
+ "step": 1250
438
+ },
439
+ {
440
+ "epoch": 7.458149779735683,
441
+ "grad_norm": 33.75697708129883,
442
+ "learning_rate": 1.3944119558154647e-05,
443
+ "loss": 0.4712,
444
+ "step": 1275
445
+ },
446
+ {
447
+ "epoch": 7.604992657856094,
448
+ "grad_norm": 66.35565948486328,
449
+ "learning_rate": 1.3781676413255362e-05,
450
+ "loss": 0.3817,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.751835535976505,
455
+ "grad_norm": 35.82437515258789,
456
+ "learning_rate": 1.3619233268356076e-05,
457
+ "loss": 0.4743,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.898678414096916,
462
+ "grad_norm": 28.671646118164062,
463
+ "learning_rate": 1.345679012345679e-05,
464
+ "loss": 0.333,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 8.0,
469
+ "eval_accuracy": 0.5729537366548043,
470
+ "eval_loss": 2.2073378562927246,
471
+ "eval_macro_f1": 0.5755855276422951,
472
+ "eval_runtime": 1.1967,
473
+ "eval_samples_per_second": 469.643,
474
+ "eval_steps_per_second": 30.084,
475
+ "eval_weighted_f1": 0.5754376022818657,
476
+ "step": 1368
477
+ },
478
+ {
479
+ "epoch": 8.041116005873715,
480
+ "grad_norm": 115.41302490234375,
481
+ "learning_rate": 1.3294346978557506e-05,
482
+ "loss": 0.4435,
483
+ "step": 1375
484
+ },
485
+ {
486
+ "epoch": 8.187958883994126,
487
+ "grad_norm": 10.865297317504883,
488
+ "learning_rate": 1.3131903833658221e-05,
489
+ "loss": 0.348,
490
+ "step": 1400
491
+ },
492
+ {
493
+ "epoch": 8.334801762114537,
494
+ "grad_norm": 77.6472396850586,
495
+ "learning_rate": 1.2969460688758935e-05,
496
+ "loss": 0.296,
497
+ "step": 1425
498
+ },
499
+ {
500
+ "epoch": 8.481644640234949,
501
+ "grad_norm": 7.557163715362549,
502
+ "learning_rate": 1.280701754385965e-05,
503
+ "loss": 0.2691,
504
+ "step": 1450
505
+ },
506
+ {
507
+ "epoch": 8.62848751835536,
508
+ "grad_norm": 52.351104736328125,
509
+ "learning_rate": 1.2644574398960365e-05,
510
+ "loss": 0.3538,
511
+ "step": 1475
512
+ },
513
+ {
514
+ "epoch": 8.775330396475772,
515
+ "grad_norm": 125.75761413574219,
516
+ "learning_rate": 1.248213125406108e-05,
517
+ "loss": 0.4217,
518
+ "step": 1500
519
+ },
520
+ {
521
+ "epoch": 8.922173274596181,
522
+ "grad_norm": 155.12820434570312,
523
+ "learning_rate": 1.2319688109161793e-05,
524
+ "loss": 0.3175,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 9.0,
529
+ "eval_accuracy": 0.5854092526690391,
530
+ "eval_loss": 2.4592270851135254,
531
+ "eval_macro_f1": 0.5778736448587275,
532
+ "eval_runtime": 1.2048,
533
+ "eval_samples_per_second": 466.472,
534
+ "eval_steps_per_second": 29.881,
535
+ "eval_weighted_f1": 0.5788739594109112,
536
+ "step": 1539
537
+ },
538
+ {
539
+ "epoch": 9.06461086637298,
540
+ "grad_norm": 97.55504608154297,
541
+ "learning_rate": 1.2157244964262509e-05,
542
+ "loss": 0.2786,
543
+ "step": 1550
544
+ },
545
+ {
546
+ "epoch": 9.211453744493392,
547
+ "grad_norm": 73.23235321044922,
548
+ "learning_rate": 1.1994801819363224e-05,
549
+ "loss": 0.1765,
550
+ "step": 1575
551
+ },
552
+ {
553
+ "epoch": 9.358296622613803,
554
+ "grad_norm": 46.12760543823242,
555
+ "learning_rate": 1.1832358674463939e-05,
556
+ "loss": 0.1052,
557
+ "step": 1600
558
+ },
559
+ {
560
+ "epoch": 9.505139500734215,
561
+ "grad_norm": 51.538185119628906,
562
+ "learning_rate": 1.1669915529564652e-05,
563
+ "loss": 0.2264,
564
+ "step": 1625
565
+ },
566
+ {
567
+ "epoch": 9.651982378854626,
568
+ "grad_norm": 73.48634338378906,
569
+ "learning_rate": 1.1507472384665367e-05,
570
+ "loss": 0.1375,
571
+ "step": 1650
572
+ },
573
+ {
574
+ "epoch": 9.798825256975036,
575
+ "grad_norm": 50.67729187011719,
576
+ "learning_rate": 1.1345029239766083e-05,
577
+ "loss": 0.204,
578
+ "step": 1675
579
+ },
580
+ {
581
+ "epoch": 9.945668135095447,
582
+ "grad_norm": 19.804401397705078,
583
+ "learning_rate": 1.11825860948668e-05,
584
+ "loss": 0.3012,
585
+ "step": 1700
586
+ },
587
+ {
588
+ "epoch": 10.0,
589
+ "eval_accuracy": 0.5765124555160143,
590
+ "eval_loss": 3.041280508041382,
591
+ "eval_macro_f1": 0.580957527780125,
592
+ "eval_runtime": 1.2149,
593
+ "eval_samples_per_second": 462.601,
594
+ "eval_steps_per_second": 29.633,
595
+ "eval_weighted_f1": 0.5803830508683689,
596
+ "step": 1710
597
+ }
598
+ ],
599
+ "logging_steps": 25,
600
+ "max_steps": 3420,
601
+ "num_input_tokens_seen": 0,
602
+ "num_train_epochs": 20,
603
+ "save_steps": 500,
604
+ "stateful_callbacks": {
605
+ "EarlyStoppingCallback": {
606
+ "args": {
607
+ "early_stopping_patience": 3,
608
+ "early_stopping_threshold": 0.0
609
+ },
610
+ "attributes": {
611
+ "early_stopping_patience_counter": 0
612
+ }
613
+ },
614
+ "TrainerControl": {
615
+ "args": {
616
+ "should_epoch_stop": false,
617
+ "should_evaluate": false,
618
+ "should_log": false,
619
+ "should_save": true,
620
+ "should_training_stop": false
621
+ },
622
+ "attributes": {}
623
+ }
624
+ },
625
+ "total_flos": 0.0,
626
+ "train_batch_size": 8,
627
+ "trial_name": null,
628
+ "trial_params": null
629
+ }
ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-2223/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-2223/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-2223/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-2223/trainer_state.json ADDED
@@ -0,0 +1,802 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1710,
3
+ "best_metric": 0.580957527780125,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-01-s44-20260831T025631Z-59e46264/trainer/checkpoint-1710",
5
+ "epoch": 13.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2223,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.14684287812041116,
14
+ "grad_norm": 19.353429794311523,
15
+ "learning_rate": 1.4035087719298246e-06,
16
+ "loss": 4.7879,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.2936857562408223,
21
+ "grad_norm": 30.09554100036621,
22
+ "learning_rate": 2.865497076023392e-06,
23
+ "loss": 4.549,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.44052863436123346,
28
+ "grad_norm": 31.329397201538086,
29
+ "learning_rate": 4.3274853801169596e-06,
30
+ "loss": 4.4843,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5873715124816447,
35
+ "grad_norm": 21.417451858520508,
36
+ "learning_rate": 5.789473684210527e-06,
37
+ "loss": 4.494,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.7342143906020558,
42
+ "grad_norm": 19.82093620300293,
43
+ "learning_rate": 7.251461988304094e-06,
44
+ "loss": 4.4666,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8810572687224669,
49
+ "grad_norm": 21.48238754272461,
50
+ "learning_rate": 8.713450292397661e-06,
51
+ "loss": 4.5115,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 1.0,
56
+ "eval_accuracy": 0.400355871886121,
57
+ "eval_loss": 1.0705939531326294,
58
+ "eval_macro_f1": 0.38519230084886114,
59
+ "eval_runtime": 1.2555,
60
+ "eval_samples_per_second": 447.648,
61
+ "eval_steps_per_second": 28.675,
62
+ "eval_weighted_f1": 0.38659017191156697,
63
+ "step": 171
64
+ },
65
+ {
66
+ "epoch": 1.0234948604992657,
67
+ "grad_norm": 18.355037689208984,
68
+ "learning_rate": 1.017543859649123e-05,
69
+ "loss": 4.2445,
70
+ "step": 175
71
+ },
72
+ {
73
+ "epoch": 1.170337738619677,
74
+ "grad_norm": 28.72773551940918,
75
+ "learning_rate": 1.1637426900584797e-05,
76
+ "loss": 4.2874,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.3171806167400881,
81
+ "grad_norm": 25.883697509765625,
82
+ "learning_rate": 1.3099415204678362e-05,
83
+ "loss": 4.2557,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.4640234948604993,
88
+ "grad_norm": 40.167694091796875,
89
+ "learning_rate": 1.4561403508771931e-05,
90
+ "loss": 4.2551,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.6108663729809103,
95
+ "grad_norm": 21.89159393310547,
96
+ "learning_rate": 1.60233918128655e-05,
97
+ "loss": 4.3497,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.7577092511013217,
102
+ "grad_norm": 26.441078186035156,
103
+ "learning_rate": 1.7485380116959064e-05,
104
+ "loss": 4.2409,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.9045521292217327,
109
+ "grad_norm": 22.421125411987305,
110
+ "learning_rate": 1.894736842105263e-05,
111
+ "loss": 4.2588,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 2.0,
116
+ "eval_accuracy": 0.46441281138790036,
117
+ "eval_loss": 1.044918417930603,
118
+ "eval_macro_f1": 0.46202289516495454,
119
+ "eval_runtime": 1.2659,
120
+ "eval_samples_per_second": 443.958,
121
+ "eval_steps_per_second": 28.439,
122
+ "eval_weighted_f1": 0.4614509909348169,
123
+ "step": 342
124
+ },
125
+ {
126
+ "epoch": 2.0469897209985315,
127
+ "grad_norm": 43.260337829589844,
128
+ "learning_rate": 1.99545159194282e-05,
129
+ "loss": 4.0338,
130
+ "step": 350
131
+ },
132
+ {
133
+ "epoch": 2.193832599118943,
134
+ "grad_norm": 32.08805465698242,
135
+ "learning_rate": 1.9792072774528916e-05,
136
+ "loss": 3.984,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.340675477239354,
141
+ "grad_norm": 26.372203826904297,
142
+ "learning_rate": 1.962962962962963e-05,
143
+ "loss": 3.8542,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.4875183553597653,
148
+ "grad_norm": 20.757434844970703,
149
+ "learning_rate": 1.9467186484730346e-05,
150
+ "loss": 3.7754,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.6343612334801763,
155
+ "grad_norm": 45.604854583740234,
156
+ "learning_rate": 1.930474333983106e-05,
157
+ "loss": 3.897,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.7812041116005872,
162
+ "grad_norm": 47.26052474975586,
163
+ "learning_rate": 1.9142300194931776e-05,
164
+ "loss": 3.7785,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.9280469897209986,
169
+ "grad_norm": 22.94167709350586,
170
+ "learning_rate": 1.897985705003249e-05,
171
+ "loss": 3.6098,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 3.0,
176
+ "eval_accuracy": 0.5088967971530249,
177
+ "eval_loss": 1.0127676725387573,
178
+ "eval_macro_f1": 0.5002107603877864,
179
+ "eval_runtime": 1.2283,
180
+ "eval_samples_per_second": 457.552,
181
+ "eval_steps_per_second": 29.309,
182
+ "eval_weighted_f1": 0.49992379157923295,
183
+ "step": 513
184
+ },
185
+ {
186
+ "epoch": 3.0704845814977975,
187
+ "grad_norm": 25.306499481201172,
188
+ "learning_rate": 1.8817413905133203e-05,
189
+ "loss": 3.1486,
190
+ "step": 525
191
+ },
192
+ {
193
+ "epoch": 3.2173274596182084,
194
+ "grad_norm": 51.927101135253906,
195
+ "learning_rate": 1.8654970760233918e-05,
196
+ "loss": 3.0286,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.36417033773862,
201
+ "grad_norm": 23.04163360595703,
202
+ "learning_rate": 1.8492527615334633e-05,
203
+ "loss": 3.0281,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.511013215859031,
208
+ "grad_norm": 31.153520584106445,
209
+ "learning_rate": 1.833008447043535e-05,
210
+ "loss": 2.888,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.657856093979442,
215
+ "grad_norm": 42.227622985839844,
216
+ "learning_rate": 1.8167641325536064e-05,
217
+ "loss": 2.8699,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.804698972099853,
222
+ "grad_norm": 36.88457489013672,
223
+ "learning_rate": 1.800519818063678e-05,
224
+ "loss": 2.9566,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.951541850220264,
229
+ "grad_norm": 30.99384117126465,
230
+ "learning_rate": 1.7842755035737494e-05,
231
+ "loss": 2.9823,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 4.0,
236
+ "eval_accuracy": 0.5658362989323843,
237
+ "eval_loss": 1.009939193725586,
238
+ "eval_macro_f1": 0.5582042983984731,
239
+ "eval_runtime": 1.1721,
240
+ "eval_samples_per_second": 479.484,
241
+ "eval_steps_per_second": 30.714,
242
+ "eval_weighted_f1": 0.5591711028839177,
243
+ "step": 684
244
+ },
245
+ {
246
+ "epoch": 4.093979441997063,
247
+ "grad_norm": 53.3286018371582,
248
+ "learning_rate": 1.768031189083821e-05,
249
+ "loss": 2.2471,
250
+ "step": 700
251
+ },
252
+ {
253
+ "epoch": 4.240822320117474,
254
+ "grad_norm": 62.07929992675781,
255
+ "learning_rate": 1.751786874593892e-05,
256
+ "loss": 1.8491,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.387665198237886,
261
+ "grad_norm": 46.8976936340332,
262
+ "learning_rate": 1.7355425601039636e-05,
263
+ "loss": 1.9902,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.534508076358296,
268
+ "grad_norm": 54.493446350097656,
269
+ "learning_rate": 1.719298245614035e-05,
270
+ "loss": 1.8621,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.681350954478708,
275
+ "grad_norm": 49.192718505859375,
276
+ "learning_rate": 1.7030539311241066e-05,
277
+ "loss": 1.8949,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.828193832599119,
282
+ "grad_norm": 68.6122817993164,
283
+ "learning_rate": 1.686809616634178e-05,
284
+ "loss": 1.9152,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.975036710719531,
289
+ "grad_norm": 64.56343078613281,
290
+ "learning_rate": 1.6705653021442497e-05,
291
+ "loss": 1.921,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 5.0,
296
+ "eval_accuracy": 0.5640569395017794,
297
+ "eval_loss": 1.111932396888733,
298
+ "eval_macro_f1": 0.5636776401175632,
299
+ "eval_runtime": 1.1624,
300
+ "eval_samples_per_second": 483.47,
301
+ "eval_steps_per_second": 30.97,
302
+ "eval_weighted_f1": 0.5636503993319255,
303
+ "step": 855
304
+ },
305
+ {
306
+ "epoch": 5.117474302496329,
307
+ "grad_norm": 122.34413146972656,
308
+ "learning_rate": 1.654320987654321e-05,
309
+ "loss": 1.3079,
310
+ "step": 875
311
+ },
312
+ {
313
+ "epoch": 5.26431718061674,
314
+ "grad_norm": 83.37774658203125,
315
+ "learning_rate": 1.6380766731643927e-05,
316
+ "loss": 1.0141,
317
+ "step": 900
318
+ },
319
+ {
320
+ "epoch": 5.411160058737151,
321
+ "grad_norm": 61.428306579589844,
322
+ "learning_rate": 1.621832358674464e-05,
323
+ "loss": 1.1481,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.558002936857562,
328
+ "grad_norm": 102.04503631591797,
329
+ "learning_rate": 1.6055880441845354e-05,
330
+ "loss": 1.0319,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.704845814977974,
335
+ "grad_norm": 35.03520202636719,
336
+ "learning_rate": 1.589343729694607e-05,
337
+ "loss": 1.1584,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.851688693098385,
342
+ "grad_norm": 43.3909797668457,
343
+ "learning_rate": 1.5730994152046787e-05,
344
+ "loss": 1.1633,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.998531571218796,
349
+ "grad_norm": 85.28243255615234,
350
+ "learning_rate": 1.55685510071475e-05,
351
+ "loss": 1.3961,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 6.0,
356
+ "eval_accuracy": 0.505338078291815,
357
+ "eval_loss": 1.8369455337524414,
358
+ "eval_macro_f1": 0.4815974670718708,
359
+ "eval_runtime": 1.1419,
360
+ "eval_samples_per_second": 492.162,
361
+ "eval_steps_per_second": 31.526,
362
+ "eval_weighted_f1": 0.4840301003544507,
363
+ "step": 1026
364
+ },
365
+ {
366
+ "epoch": 6.140969162995595,
367
+ "grad_norm": 63.258750915527344,
368
+ "learning_rate": 1.5406107862248214e-05,
369
+ "loss": 0.6735,
370
+ "step": 1050
371
+ },
372
+ {
373
+ "epoch": 6.2878120411160054,
374
+ "grad_norm": 35.072357177734375,
375
+ "learning_rate": 1.524366471734893e-05,
376
+ "loss": 0.7072,
377
+ "step": 1075
378
+ },
379
+ {
380
+ "epoch": 6.434654919236417,
381
+ "grad_norm": 29.91564178466797,
382
+ "learning_rate": 1.5081221572449645e-05,
383
+ "loss": 0.6603,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.581497797356828,
388
+ "grad_norm": 88.57781982421875,
389
+ "learning_rate": 1.4918778427550358e-05,
390
+ "loss": 0.5712,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.72834067547724,
395
+ "grad_norm": 47.00954055786133,
396
+ "learning_rate": 1.4756335282651073e-05,
397
+ "loss": 0.7554,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.87518355359765,
402
+ "grad_norm": 29.895748138427734,
403
+ "learning_rate": 1.4593892137751788e-05,
404
+ "loss": 0.6925,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 7.0,
409
+ "eval_accuracy": 0.5693950177935944,
410
+ "eval_loss": 1.8101662397384644,
411
+ "eval_macro_f1": 0.5657155822901679,
412
+ "eval_runtime": 1.171,
413
+ "eval_samples_per_second": 479.939,
414
+ "eval_steps_per_second": 30.743,
415
+ "eval_weighted_f1": 0.5661096303435228,
416
+ "step": 1197
417
+ },
418
+ {
419
+ "epoch": 7.0176211453744495,
420
+ "grad_norm": 74.3259506225586,
421
+ "learning_rate": 1.4431448992852503e-05,
422
+ "loss": 0.705,
423
+ "step": 1200
424
+ },
425
+ {
426
+ "epoch": 7.164464023494861,
427
+ "grad_norm": 198.61756896972656,
428
+ "learning_rate": 1.4269005847953217e-05,
429
+ "loss": 0.4091,
430
+ "step": 1225
431
+ },
432
+ {
433
+ "epoch": 7.311306901615271,
434
+ "grad_norm": 67.14990234375,
435
+ "learning_rate": 1.4106562703053932e-05,
436
+ "loss": 0.3649,
437
+ "step": 1250
438
+ },
439
+ {
440
+ "epoch": 7.458149779735683,
441
+ "grad_norm": 33.75697708129883,
442
+ "learning_rate": 1.3944119558154647e-05,
443
+ "loss": 0.4712,
444
+ "step": 1275
445
+ },
446
+ {
447
+ "epoch": 7.604992657856094,
448
+ "grad_norm": 66.35565948486328,
449
+ "learning_rate": 1.3781676413255362e-05,
450
+ "loss": 0.3817,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.751835535976505,
455
+ "grad_norm": 35.82437515258789,
456
+ "learning_rate": 1.3619233268356076e-05,
457
+ "loss": 0.4743,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.898678414096916,
462
+ "grad_norm": 28.671646118164062,
463
+ "learning_rate": 1.345679012345679e-05,
464
+ "loss": 0.333,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 8.0,
469
+ "eval_accuracy": 0.5729537366548043,
470
+ "eval_loss": 2.2073378562927246,
471
+ "eval_macro_f1": 0.5755855276422951,
472
+ "eval_runtime": 1.1967,
473
+ "eval_samples_per_second": 469.643,
474
+ "eval_steps_per_second": 30.084,
475
+ "eval_weighted_f1": 0.5754376022818657,
476
+ "step": 1368
477
+ },
478
+ {
479
+ "epoch": 8.041116005873715,
480
+ "grad_norm": 115.41302490234375,
481
+ "learning_rate": 1.3294346978557506e-05,
482
+ "loss": 0.4435,
483
+ "step": 1375
484
+ },
485
+ {
486
+ "epoch": 8.187958883994126,
487
+ "grad_norm": 10.865297317504883,
488
+ "learning_rate": 1.3131903833658221e-05,
489
+ "loss": 0.348,
490
+ "step": 1400
491
+ },
492
+ {
493
+ "epoch": 8.334801762114537,
494
+ "grad_norm": 77.6472396850586,
495
+ "learning_rate": 1.2969460688758935e-05,
496
+ "loss": 0.296,
497
+ "step": 1425
498
+ },
499
+ {
500
+ "epoch": 8.481644640234949,
501
+ "grad_norm": 7.557163715362549,
502
+ "learning_rate": 1.280701754385965e-05,
503
+ "loss": 0.2691,
504
+ "step": 1450
505
+ },
506
+ {
507
+ "epoch": 8.62848751835536,
508
+ "grad_norm": 52.351104736328125,
509
+ "learning_rate": 1.2644574398960365e-05,
510
+ "loss": 0.3538,
511
+ "step": 1475
512
+ },
513
+ {
514
+ "epoch": 8.775330396475772,
515
+ "grad_norm": 125.75761413574219,
516
+ "learning_rate": 1.248213125406108e-05,
517
+ "loss": 0.4217,
518
+ "step": 1500
519
+ },
520
+ {
521
+ "epoch": 8.922173274596181,
522
+ "grad_norm": 155.12820434570312,
523
+ "learning_rate": 1.2319688109161793e-05,
524
+ "loss": 0.3175,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 9.0,
529
+ "eval_accuracy": 0.5854092526690391,
530
+ "eval_loss": 2.4592270851135254,
531
+ "eval_macro_f1": 0.5778736448587275,
532
+ "eval_runtime": 1.2048,
533
+ "eval_samples_per_second": 466.472,
534
+ "eval_steps_per_second": 29.881,
535
+ "eval_weighted_f1": 0.5788739594109112,
536
+ "step": 1539
537
+ },
538
+ {
539
+ "epoch": 9.06461086637298,
540
+ "grad_norm": 97.55504608154297,
541
+ "learning_rate": 1.2157244964262509e-05,
542
+ "loss": 0.2786,
543
+ "step": 1550
544
+ },
545
+ {
546
+ "epoch": 9.211453744493392,
547
+ "grad_norm": 73.23235321044922,
548
+ "learning_rate": 1.1994801819363224e-05,
549
+ "loss": 0.1765,
550
+ "step": 1575
551
+ },
552
+ {
553
+ "epoch": 9.358296622613803,
554
+ "grad_norm": 46.12760543823242,
555
+ "learning_rate": 1.1832358674463939e-05,
556
+ "loss": 0.1052,
557
+ "step": 1600
558
+ },
559
+ {
560
+ "epoch": 9.505139500734215,
561
+ "grad_norm": 51.538185119628906,
562
+ "learning_rate": 1.1669915529564652e-05,
563
+ "loss": 0.2264,
564
+ "step": 1625
565
+ },
566
+ {
567
+ "epoch": 9.651982378854626,
568
+ "grad_norm": 73.48634338378906,
569
+ "learning_rate": 1.1507472384665367e-05,
570
+ "loss": 0.1375,
571
+ "step": 1650
572
+ },
573
+ {
574
+ "epoch": 9.798825256975036,
575
+ "grad_norm": 50.67729187011719,
576
+ "learning_rate": 1.1345029239766083e-05,
577
+ "loss": 0.204,
578
+ "step": 1675
579
+ },
580
+ {
581
+ "epoch": 9.945668135095447,
582
+ "grad_norm": 19.804401397705078,
583
+ "learning_rate": 1.11825860948668e-05,
584
+ "loss": 0.3012,
585
+ "step": 1700
586
+ },
587
+ {
588
+ "epoch": 10.0,
589
+ "eval_accuracy": 0.5765124555160143,
590
+ "eval_loss": 3.041280508041382,
591
+ "eval_macro_f1": 0.580957527780125,
592
+ "eval_runtime": 1.2149,
593
+ "eval_samples_per_second": 462.601,
594
+ "eval_steps_per_second": 29.633,
595
+ "eval_weighted_f1": 0.5803830508683689,
596
+ "step": 1710
597
+ },
598
+ {
599
+ "epoch": 10.088105726872246,
600
+ "grad_norm": 62.47627258300781,
601
+ "learning_rate": 1.1020142949967511e-05,
602
+ "loss": 0.101,
603
+ "step": 1725
604
+ },
605
+ {
606
+ "epoch": 10.234948604992658,
607
+ "grad_norm": 0.8165979981422424,
608
+ "learning_rate": 1.0857699805068226e-05,
609
+ "loss": 0.1177,
610
+ "step": 1750
611
+ },
612
+ {
613
+ "epoch": 10.38179148311307,
614
+ "grad_norm": 137.25111389160156,
615
+ "learning_rate": 1.0695256660168943e-05,
616
+ "loss": 0.128,
617
+ "step": 1775
618
+ },
619
+ {
620
+ "epoch": 10.52863436123348,
621
+ "grad_norm": 72.26915740966797,
622
+ "learning_rate": 1.0532813515269658e-05,
623
+ "loss": 0.1955,
624
+ "step": 1800
625
+ },
626
+ {
627
+ "epoch": 10.675477239353892,
628
+ "grad_norm": 25.298179626464844,
629
+ "learning_rate": 1.037037037037037e-05,
630
+ "loss": 0.21,
631
+ "step": 1825
632
+ },
633
+ {
634
+ "epoch": 10.822320117474302,
635
+ "grad_norm": 2.867316961288452,
636
+ "learning_rate": 1.0207927225471085e-05,
637
+ "loss": 0.1748,
638
+ "step": 1850
639
+ },
640
+ {
641
+ "epoch": 10.969162995594713,
642
+ "grad_norm": 9.12874984741211,
643
+ "learning_rate": 1.0045484080571802e-05,
644
+ "loss": 0.2161,
645
+ "step": 1875
646
+ },
647
+ {
648
+ "epoch": 11.0,
649
+ "eval_accuracy": 0.5729537366548043,
650
+ "eval_loss": 3.2233715057373047,
651
+ "eval_macro_f1": 0.5756900718966651,
652
+ "eval_runtime": 1.1998,
653
+ "eval_samples_per_second": 468.429,
654
+ "eval_steps_per_second": 30.006,
655
+ "eval_weighted_f1": 0.5756241335436229,
656
+ "step": 1881
657
+ },
658
+ {
659
+ "epoch": 11.111600587371512,
660
+ "grad_norm": 13.847013473510742,
661
+ "learning_rate": 9.883040935672515e-06,
662
+ "loss": 0.1083,
663
+ "step": 1900
664
+ },
665
+ {
666
+ "epoch": 11.258443465491924,
667
+ "grad_norm": 0.36310669779777527,
668
+ "learning_rate": 9.720597790773229e-06,
669
+ "loss": 0.0303,
670
+ "step": 1925
671
+ },
672
+ {
673
+ "epoch": 11.405286343612335,
674
+ "grad_norm": 0.1778726428747177,
675
+ "learning_rate": 9.558154645873946e-06,
676
+ "loss": 0.1078,
677
+ "step": 1950
678
+ },
679
+ {
680
+ "epoch": 11.552129221732747,
681
+ "grad_norm": 5.069540977478027,
682
+ "learning_rate": 9.395711500974659e-06,
683
+ "loss": 0.113,
684
+ "step": 1975
685
+ },
686
+ {
687
+ "epoch": 11.698972099853156,
688
+ "grad_norm": 0.1319819688796997,
689
+ "learning_rate": 9.233268356075374e-06,
690
+ "loss": 0.2165,
691
+ "step": 2000
692
+ },
693
+ {
694
+ "epoch": 11.845814977973568,
695
+ "grad_norm": 2.094083547592163,
696
+ "learning_rate": 9.07082521117609e-06,
697
+ "loss": 0.1849,
698
+ "step": 2025
699
+ },
700
+ {
701
+ "epoch": 11.99265785609398,
702
+ "grad_norm": 55.9804573059082,
703
+ "learning_rate": 8.908382066276805e-06,
704
+ "loss": 0.0822,
705
+ "step": 2050
706
+ },
707
+ {
708
+ "epoch": 12.0,
709
+ "eval_accuracy": 0.5587188612099644,
710
+ "eval_loss": 3.746398687362671,
711
+ "eval_macro_f1": 0.5618844911742946,
712
+ "eval_runtime": 1.1999,
713
+ "eval_samples_per_second": 468.364,
714
+ "eval_steps_per_second": 30.002,
715
+ "eval_weighted_f1": 0.5624085755713754,
716
+ "step": 2052
717
+ },
718
+ {
719
+ "epoch": 12.135095447870778,
720
+ "grad_norm": 0.03340785950422287,
721
+ "learning_rate": 8.745938921377518e-06,
722
+ "loss": 0.0232,
723
+ "step": 2075
724
+ },
725
+ {
726
+ "epoch": 12.28193832599119,
727
+ "grad_norm": 112.34405517578125,
728
+ "learning_rate": 8.583495776478233e-06,
729
+ "loss": 0.0708,
730
+ "step": 2100
731
+ },
732
+ {
733
+ "epoch": 12.428781204111601,
734
+ "grad_norm": 0.1899978667497635,
735
+ "learning_rate": 8.421052631578948e-06,
736
+ "loss": 0.0925,
737
+ "step": 2125
738
+ },
739
+ {
740
+ "epoch": 12.575624082232011,
741
+ "grad_norm": 3.4805235862731934,
742
+ "learning_rate": 8.258609486679663e-06,
743
+ "loss": 0.047,
744
+ "step": 2150
745
+ },
746
+ {
747
+ "epoch": 12.722466960352422,
748
+ "grad_norm": 0.1855817437171936,
749
+ "learning_rate": 8.096166341780377e-06,
750
+ "loss": 0.1449,
751
+ "step": 2175
752
+ },
753
+ {
754
+ "epoch": 12.869309838472834,
755
+ "grad_norm": 0.012470078654587269,
756
+ "learning_rate": 7.933723196881092e-06,
757
+ "loss": 0.037,
758
+ "step": 2200
759
+ },
760
+ {
761
+ "epoch": 13.0,
762
+ "eval_accuracy": 0.5782918149466192,
763
+ "eval_loss": 3.781280040740967,
764
+ "eval_macro_f1": 0.5758329304226294,
765
+ "eval_runtime": 1.1867,
766
+ "eval_samples_per_second": 473.578,
767
+ "eval_steps_per_second": 30.336,
768
+ "eval_weighted_f1": 0.5760590427541459,
769
+ "step": 2223
770
+ }
771
+ ],
772
+ "logging_steps": 25,
773
+ "max_steps": 3420,
774
+ "num_input_tokens_seen": 0,
775
+ "num_train_epochs": 20,
776
+ "save_steps": 500,
777
+ "stateful_callbacks": {
778
+ "EarlyStoppingCallback": {
779
+ "args": {
780
+ "early_stopping_patience": 3,
781
+ "early_stopping_threshold": 0.0
782
+ },
783
+ "attributes": {
784
+ "early_stopping_patience_counter": 3
785
+ }
786
+ },
787
+ "TrainerControl": {
788
+ "args": {
789
+ "should_epoch_stop": false,
790
+ "should_evaluate": false,
791
+ "should_log": false,
792
+ "should_save": true,
793
+ "should_training_stop": true
794
+ },
795
+ "attributes": {}
796
+ }
797
+ },
798
+ "total_flos": 0.0,
799
+ "train_batch_size": 8,
800
+ "trial_name": null,
801
+ "trial_params": null
802
+ }
ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-1267/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-1267/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-1267/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-1267/trainer_state.json ADDED
@@ -0,0 +1,470 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 724,
3
+ "best_metric": 0.581015262162336,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-724",
5
+ "epoch": 7.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1267,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13812154696132597,
14
+ "grad_norm": 26.694236755371094,
15
+ "learning_rate": 1.3259668508287293e-06,
16
+ "loss": 4.7484,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27624309392265195,
21
+ "grad_norm": 26.276744842529297,
22
+ "learning_rate": 2.707182320441989e-06,
23
+ "loss": 4.5513,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.4143646408839779,
28
+ "grad_norm": 24.02865219116211,
29
+ "learning_rate": 4.088397790055249e-06,
30
+ "loss": 4.4834,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5524861878453039,
35
+ "grad_norm": 23.19355010986328,
36
+ "learning_rate": 5.469613259668509e-06,
37
+ "loss": 4.4407,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6906077348066298,
42
+ "grad_norm": 19.562156677246094,
43
+ "learning_rate": 6.850828729281769e-06,
44
+ "loss": 4.447,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8287292817679558,
49
+ "grad_norm": 30.32309341430664,
50
+ "learning_rate": 8.232044198895029e-06,
51
+ "loss": 4.4195,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9668508287292817,
56
+ "grad_norm": 37.31390380859375,
57
+ "learning_rate": 9.613259668508288e-06,
58
+ "loss": 4.4054,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.47079037800687284,
64
+ "eval_loss": 1.0521403551101685,
65
+ "eval_macro_f1": 0.4445009430531239,
66
+ "eval_runtime": 1.2681,
67
+ "eval_samples_per_second": 458.953,
68
+ "eval_steps_per_second": 29.177,
69
+ "eval_weighted_f1": 0.4481873351422723,
70
+ "step": 181
71
+ },
72
+ {
73
+ "epoch": 1.1049723756906078,
74
+ "grad_norm": 34.274879455566406,
75
+ "learning_rate": 1.0994475138121547e-05,
76
+ "loss": 4.2683,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2430939226519337,
81
+ "grad_norm": 17.634672164916992,
82
+ "learning_rate": 1.2375690607734806e-05,
83
+ "loss": 4.3316,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3812154696132597,
88
+ "grad_norm": 25.136465072631836,
89
+ "learning_rate": 1.3756906077348067e-05,
90
+ "loss": 4.3891,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.5193370165745856,
95
+ "grad_norm": 42.85128402709961,
96
+ "learning_rate": 1.5138121546961326e-05,
97
+ "loss": 4.3788,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6574585635359116,
102
+ "grad_norm": 18.8840389251709,
103
+ "learning_rate": 1.6519337016574586e-05,
104
+ "loss": 4.2523,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7955801104972375,
109
+ "grad_norm": 23.44044303894043,
110
+ "learning_rate": 1.7900552486187846e-05,
111
+ "loss": 4.2698,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.9337016574585635,
116
+ "grad_norm": 25.44222068786621,
117
+ "learning_rate": 1.9281767955801107e-05,
118
+ "loss": 4.2717,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.4690721649484536,
124
+ "eval_loss": 1.0339078903198242,
125
+ "eval_macro_f1": 0.4063231112108802,
126
+ "eval_runtime": 1.2852,
127
+ "eval_samples_per_second": 452.862,
128
+ "eval_steps_per_second": 28.79,
129
+ "eval_weighted_f1": 0.4122189154907193,
130
+ "step": 362
131
+ },
132
+ {
133
+ "epoch": 2.0718232044198897,
134
+ "grad_norm": 15.546817779541016,
135
+ "learning_rate": 1.9926335174953962e-05,
136
+ "loss": 4.2522,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.2099447513812156,
141
+ "grad_norm": 51.74812316894531,
142
+ "learning_rate": 1.9772866789441377e-05,
143
+ "loss": 4.0238,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.3480662983425415,
148
+ "grad_norm": 41.77598571777344,
149
+ "learning_rate": 1.9619398403928794e-05,
150
+ "loss": 3.946,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.4861878453038675,
155
+ "grad_norm": 25.568912506103516,
156
+ "learning_rate": 1.946593001841621e-05,
157
+ "loss": 3.9322,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.6243093922651934,
162
+ "grad_norm": 25.970388412475586,
163
+ "learning_rate": 1.9312461632903623e-05,
164
+ "loss": 3.8189,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.7624309392265194,
169
+ "grad_norm": 48.910701751708984,
170
+ "learning_rate": 1.915899324739104e-05,
171
+ "loss": 3.7624,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.9005524861878453,
176
+ "grad_norm": 33.297969818115234,
177
+ "learning_rate": 1.9005524861878455e-05,
178
+ "loss": 3.7766,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5257731958762887,
184
+ "eval_loss": 0.9845371246337891,
185
+ "eval_macro_f1": 0.5044701213818862,
186
+ "eval_runtime": 1.2221,
187
+ "eval_samples_per_second": 476.237,
188
+ "eval_steps_per_second": 30.276,
189
+ "eval_weighted_f1": 0.5061129520536737,
190
+ "step": 543
191
+ },
192
+ {
193
+ "epoch": 3.0386740331491713,
194
+ "grad_norm": 28.731353759765625,
195
+ "learning_rate": 1.885205647636587e-05,
196
+ "loss": 3.5516,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.1767955801104972,
201
+ "grad_norm": 61.38904571533203,
202
+ "learning_rate": 1.8698588090853287e-05,
203
+ "loss": 3.0653,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.314917127071823,
208
+ "grad_norm": 38.65430450439453,
209
+ "learning_rate": 1.85451197053407e-05,
210
+ "loss": 3.1863,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.453038674033149,
215
+ "grad_norm": 49.679988861083984,
216
+ "learning_rate": 1.8391651319828116e-05,
217
+ "loss": 3.1589,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.591160220994475,
222
+ "grad_norm": 44.78519821166992,
223
+ "learning_rate": 1.823818293431553e-05,
224
+ "loss": 2.8768,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.729281767955801,
229
+ "grad_norm": 48.2161865234375,
230
+ "learning_rate": 1.8084714548802948e-05,
231
+ "loss": 2.8203,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.867403314917127,
236
+ "grad_norm": 66.51898956298828,
237
+ "learning_rate": 1.7931246163290362e-05,
238
+ "loss": 3.0759,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 4.0,
243
+ "eval_accuracy": 0.5773195876288659,
244
+ "eval_loss": 0.9527704119682312,
245
+ "eval_macro_f1": 0.581015262162336,
246
+ "eval_runtime": 1.2079,
247
+ "eval_samples_per_second": 481.827,
248
+ "eval_steps_per_second": 30.632,
249
+ "eval_weighted_f1": 0.5804185166178877,
250
+ "step": 724
251
+ },
252
+ {
253
+ "epoch": 4.005524861878453,
254
+ "grad_norm": 35.996734619140625,
255
+ "learning_rate": 1.7777777777777777e-05,
256
+ "loss": 2.8314,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.143646408839779,
261
+ "grad_norm": 77.58326721191406,
262
+ "learning_rate": 1.7624309392265194e-05,
263
+ "loss": 1.9896,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.281767955801105,
268
+ "grad_norm": 94.60843658447266,
269
+ "learning_rate": 1.747084100675261e-05,
270
+ "loss": 2.0467,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.419889502762431,
275
+ "grad_norm": 42.56327438354492,
276
+ "learning_rate": 1.7317372621240026e-05,
277
+ "loss": 1.8967,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.558011049723757,
282
+ "grad_norm": 67.6713638305664,
283
+ "learning_rate": 1.716390423572744e-05,
284
+ "loss": 1.642,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.696132596685083,
289
+ "grad_norm": 73.13631439208984,
290
+ "learning_rate": 1.701043585021486e-05,
291
+ "loss": 1.9882,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.834254143646409,
296
+ "grad_norm": 36.94429397583008,
297
+ "learning_rate": 1.6856967464702273e-05,
298
+ "loss": 2.0404,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.972375690607735,
303
+ "grad_norm": 110.7317886352539,
304
+ "learning_rate": 1.670349907918969e-05,
305
+ "loss": 1.9671,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 5.0,
310
+ "eval_accuracy": 0.5773195876288659,
311
+ "eval_loss": 1.1364641189575195,
312
+ "eval_macro_f1": 0.5594730958639235,
313
+ "eval_runtime": 1.2606,
314
+ "eval_samples_per_second": 461.674,
315
+ "eval_steps_per_second": 29.35,
316
+ "eval_weighted_f1": 0.5615914518238473,
317
+ "step": 905
318
+ },
319
+ {
320
+ "epoch": 5.110497237569061,
321
+ "grad_norm": 40.61456298828125,
322
+ "learning_rate": 1.6550030693677105e-05,
323
+ "loss": 1.3611,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.248618784530387,
328
+ "grad_norm": 41.819969177246094,
329
+ "learning_rate": 1.639656230816452e-05,
330
+ "loss": 1.0609,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.386740331491713,
335
+ "grad_norm": 106.5338363647461,
336
+ "learning_rate": 1.6243093922651937e-05,
337
+ "loss": 1.421,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.524861878453039,
342
+ "grad_norm": 44.064903259277344,
343
+ "learning_rate": 1.608962553713935e-05,
344
+ "loss": 0.9507,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.662983425414365,
349
+ "grad_norm": 68.2384262084961,
350
+ "learning_rate": 1.5936157151626766e-05,
351
+ "loss": 1.3316,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.801104972375691,
356
+ "grad_norm": 47.8666877746582,
357
+ "learning_rate": 1.5782688766114183e-05,
358
+ "loss": 1.0212,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.939226519337017,
363
+ "grad_norm": 45.24050521850586,
364
+ "learning_rate": 1.5629220380601598e-05,
365
+ "loss": 1.1752,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 6.0,
370
+ "eval_accuracy": 0.570446735395189,
371
+ "eval_loss": 1.4164072275161743,
372
+ "eval_macro_f1": 0.5593843071324701,
373
+ "eval_runtime": 1.2081,
374
+ "eval_samples_per_second": 481.754,
375
+ "eval_steps_per_second": 30.627,
376
+ "eval_weighted_f1": 0.560737315222414,
377
+ "step": 1086
378
+ },
379
+ {
380
+ "epoch": 6.077348066298343,
381
+ "grad_norm": 39.47881317138672,
382
+ "learning_rate": 1.5475751995089012e-05,
383
+ "loss": 0.8472,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.2154696132596685,
388
+ "grad_norm": 22.349708557128906,
389
+ "learning_rate": 1.532228360957643e-05,
390
+ "loss": 0.6577,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.3535911602209945,
395
+ "grad_norm": 28.993520736694336,
396
+ "learning_rate": 1.5168815224063844e-05,
397
+ "loss": 0.4534,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.49171270718232,
402
+ "grad_norm": 70.58302307128906,
403
+ "learning_rate": 1.5015346838551258e-05,
404
+ "loss": 0.6073,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.629834254143646,
409
+ "grad_norm": 39.072757720947266,
410
+ "learning_rate": 1.4861878453038676e-05,
411
+ "loss": 0.8806,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.767955801104972,
416
+ "grad_norm": 41.504207611083984,
417
+ "learning_rate": 1.470841006752609e-05,
418
+ "loss": 0.5984,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.906077348066298,
423
+ "grad_norm": 72.10482788085938,
424
+ "learning_rate": 1.4554941682013506e-05,
425
+ "loss": 0.6006,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 7.0,
430
+ "eval_accuracy": 0.5498281786941581,
431
+ "eval_loss": 1.9378756284713745,
432
+ "eval_macro_f1": 0.5328115558431438,
433
+ "eval_runtime": 1.2051,
434
+ "eval_samples_per_second": 482.94,
435
+ "eval_steps_per_second": 30.702,
436
+ "eval_weighted_f1": 0.5353102780018485,
437
+ "step": 1267
438
+ }
439
+ ],
440
+ "logging_steps": 25,
441
+ "max_steps": 3620,
442
+ "num_input_tokens_seen": 0,
443
+ "num_train_epochs": 20,
444
+ "save_steps": 500,
445
+ "stateful_callbacks": {
446
+ "EarlyStoppingCallback": {
447
+ "args": {
448
+ "early_stopping_patience": 3,
449
+ "early_stopping_threshold": 0.0
450
+ },
451
+ "attributes": {
452
+ "early_stopping_patience_counter": 3
453
+ }
454
+ },
455
+ "TrainerControl": {
456
+ "args": {
457
+ "should_epoch_stop": false,
458
+ "should_evaluate": false,
459
+ "should_log": false,
460
+ "should_save": true,
461
+ "should_training_stop": true
462
+ },
463
+ "attributes": {}
464
+ }
465
+ },
466
+ "total_flos": 0.0,
467
+ "train_batch_size": 8,
468
+ "trial_name": null,
469
+ "trial_params": null
470
+ }
ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-724/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-724/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-724/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-724/trainer_state.json ADDED
@@ -0,0 +1,283 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 724,
3
+ "best_metric": 0.581015262162336,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-02-s44-20260831T030909Z-59e46264/trainer/checkpoint-724",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 724,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13812154696132597,
14
+ "grad_norm": 26.694236755371094,
15
+ "learning_rate": 1.3259668508287293e-06,
16
+ "loss": 4.7484,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27624309392265195,
21
+ "grad_norm": 26.276744842529297,
22
+ "learning_rate": 2.707182320441989e-06,
23
+ "loss": 4.5513,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.4143646408839779,
28
+ "grad_norm": 24.02865219116211,
29
+ "learning_rate": 4.088397790055249e-06,
30
+ "loss": 4.4834,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5524861878453039,
35
+ "grad_norm": 23.19355010986328,
36
+ "learning_rate": 5.469613259668509e-06,
37
+ "loss": 4.4407,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6906077348066298,
42
+ "grad_norm": 19.562156677246094,
43
+ "learning_rate": 6.850828729281769e-06,
44
+ "loss": 4.447,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8287292817679558,
49
+ "grad_norm": 30.32309341430664,
50
+ "learning_rate": 8.232044198895029e-06,
51
+ "loss": 4.4195,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9668508287292817,
56
+ "grad_norm": 37.31390380859375,
57
+ "learning_rate": 9.613259668508288e-06,
58
+ "loss": 4.4054,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.47079037800687284,
64
+ "eval_loss": 1.0521403551101685,
65
+ "eval_macro_f1": 0.4445009430531239,
66
+ "eval_runtime": 1.2681,
67
+ "eval_samples_per_second": 458.953,
68
+ "eval_steps_per_second": 29.177,
69
+ "eval_weighted_f1": 0.4481873351422723,
70
+ "step": 181
71
+ },
72
+ {
73
+ "epoch": 1.1049723756906078,
74
+ "grad_norm": 34.274879455566406,
75
+ "learning_rate": 1.0994475138121547e-05,
76
+ "loss": 4.2683,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2430939226519337,
81
+ "grad_norm": 17.634672164916992,
82
+ "learning_rate": 1.2375690607734806e-05,
83
+ "loss": 4.3316,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3812154696132597,
88
+ "grad_norm": 25.136465072631836,
89
+ "learning_rate": 1.3756906077348067e-05,
90
+ "loss": 4.3891,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.5193370165745856,
95
+ "grad_norm": 42.85128402709961,
96
+ "learning_rate": 1.5138121546961326e-05,
97
+ "loss": 4.3788,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6574585635359116,
102
+ "grad_norm": 18.8840389251709,
103
+ "learning_rate": 1.6519337016574586e-05,
104
+ "loss": 4.2523,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7955801104972375,
109
+ "grad_norm": 23.44044303894043,
110
+ "learning_rate": 1.7900552486187846e-05,
111
+ "loss": 4.2698,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.9337016574585635,
116
+ "grad_norm": 25.44222068786621,
117
+ "learning_rate": 1.9281767955801107e-05,
118
+ "loss": 4.2717,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.4690721649484536,
124
+ "eval_loss": 1.0339078903198242,
125
+ "eval_macro_f1": 0.4063231112108802,
126
+ "eval_runtime": 1.2852,
127
+ "eval_samples_per_second": 452.862,
128
+ "eval_steps_per_second": 28.79,
129
+ "eval_weighted_f1": 0.4122189154907193,
130
+ "step": 362
131
+ },
132
+ {
133
+ "epoch": 2.0718232044198897,
134
+ "grad_norm": 15.546817779541016,
135
+ "learning_rate": 1.9926335174953962e-05,
136
+ "loss": 4.2522,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.2099447513812156,
141
+ "grad_norm": 51.74812316894531,
142
+ "learning_rate": 1.9772866789441377e-05,
143
+ "loss": 4.0238,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.3480662983425415,
148
+ "grad_norm": 41.77598571777344,
149
+ "learning_rate": 1.9619398403928794e-05,
150
+ "loss": 3.946,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.4861878453038675,
155
+ "grad_norm": 25.568912506103516,
156
+ "learning_rate": 1.946593001841621e-05,
157
+ "loss": 3.9322,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.6243093922651934,
162
+ "grad_norm": 25.970388412475586,
163
+ "learning_rate": 1.9312461632903623e-05,
164
+ "loss": 3.8189,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.7624309392265194,
169
+ "grad_norm": 48.910701751708984,
170
+ "learning_rate": 1.915899324739104e-05,
171
+ "loss": 3.7624,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.9005524861878453,
176
+ "grad_norm": 33.297969818115234,
177
+ "learning_rate": 1.9005524861878455e-05,
178
+ "loss": 3.7766,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5257731958762887,
184
+ "eval_loss": 0.9845371246337891,
185
+ "eval_macro_f1": 0.5044701213818862,
186
+ "eval_runtime": 1.2221,
187
+ "eval_samples_per_second": 476.237,
188
+ "eval_steps_per_second": 30.276,
189
+ "eval_weighted_f1": 0.5061129520536737,
190
+ "step": 543
191
+ },
192
+ {
193
+ "epoch": 3.0386740331491713,
194
+ "grad_norm": 28.731353759765625,
195
+ "learning_rate": 1.885205647636587e-05,
196
+ "loss": 3.5516,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.1767955801104972,
201
+ "grad_norm": 61.38904571533203,
202
+ "learning_rate": 1.8698588090853287e-05,
203
+ "loss": 3.0653,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.314917127071823,
208
+ "grad_norm": 38.65430450439453,
209
+ "learning_rate": 1.85451197053407e-05,
210
+ "loss": 3.1863,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.453038674033149,
215
+ "grad_norm": 49.679988861083984,
216
+ "learning_rate": 1.8391651319828116e-05,
217
+ "loss": 3.1589,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.591160220994475,
222
+ "grad_norm": 44.78519821166992,
223
+ "learning_rate": 1.823818293431553e-05,
224
+ "loss": 2.8768,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.729281767955801,
229
+ "grad_norm": 48.2161865234375,
230
+ "learning_rate": 1.8084714548802948e-05,
231
+ "loss": 2.8203,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.867403314917127,
236
+ "grad_norm": 66.51898956298828,
237
+ "learning_rate": 1.7931246163290362e-05,
238
+ "loss": 3.0759,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 4.0,
243
+ "eval_accuracy": 0.5773195876288659,
244
+ "eval_loss": 0.9527704119682312,
245
+ "eval_macro_f1": 0.581015262162336,
246
+ "eval_runtime": 1.2079,
247
+ "eval_samples_per_second": 481.827,
248
+ "eval_steps_per_second": 30.632,
249
+ "eval_weighted_f1": 0.5804185166178877,
250
+ "step": 724
251
+ }
252
+ ],
253
+ "logging_steps": 25,
254
+ "max_steps": 3620,
255
+ "num_input_tokens_seen": 0,
256
+ "num_train_epochs": 20,
257
+ "save_steps": 500,
258
+ "stateful_callbacks": {
259
+ "EarlyStoppingCallback": {
260
+ "args": {
261
+ "early_stopping_patience": 3,
262
+ "early_stopping_threshold": 0.0
263
+ },
264
+ "attributes": {
265
+ "early_stopping_patience_counter": 0
266
+ }
267
+ },
268
+ "TrainerControl": {
269
+ "args": {
270
+ "should_epoch_stop": false,
271
+ "should_evaluate": false,
272
+ "should_log": false,
273
+ "should_save": true,
274
+ "should_training_stop": false
275
+ },
276
+ "attributes": {}
277
+ }
278
+ },
279
+ "total_flos": 0.0,
280
+ "train_batch_size": 8,
281
+ "trial_name": null,
282
+ "trial_params": null
283
+ }
ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-1281/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-1281/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-1281/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-1281/trainer_state.json ADDED
@@ -0,0 +1,477 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 732,
3
+ "best_metric": 0.5672622157863422,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-732",
5
+ "epoch": 7.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1281,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13717421124828533,
14
+ "grad_norm": 23.982624053955078,
15
+ "learning_rate": 1.3114754098360657e-06,
16
+ "loss": 4.7545,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27434842249657065,
21
+ "grad_norm": 22.713624954223633,
22
+ "learning_rate": 2.677595628415301e-06,
23
+ "loss": 4.5062,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.411522633744856,
28
+ "grad_norm": 19.318227767944336,
29
+ "learning_rate": 4.043715846994536e-06,
30
+ "loss": 4.4396,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5486968449931413,
35
+ "grad_norm": 26.55528450012207,
36
+ "learning_rate": 5.409836065573772e-06,
37
+ "loss": 4.4558,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6858710562414266,
42
+ "grad_norm": 16.671968460083008,
43
+ "learning_rate": 6.775956284153006e-06,
44
+ "loss": 4.3536,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.823045267489712,
49
+ "grad_norm": 40.652828216552734,
50
+ "learning_rate": 8.14207650273224e-06,
51
+ "loss": 4.4056,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9602194787379973,
56
+ "grad_norm": 15.629646301269531,
57
+ "learning_rate": 9.508196721311476e-06,
58
+ "loss": 4.3376,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.3910149750415973,
64
+ "eval_loss": 1.0781054496765137,
65
+ "eval_macro_f1": 0.36967599249187333,
66
+ "eval_runtime": 1.3916,
67
+ "eval_samples_per_second": 431.865,
68
+ "eval_steps_per_second": 27.306,
69
+ "eval_weighted_f1": 0.3712625689976963,
70
+ "step": 183
71
+ },
72
+ {
73
+ "epoch": 1.093278463648834,
74
+ "grad_norm": 17.34412384033203,
75
+ "learning_rate": 1.0874316939890712e-05,
76
+ "loss": 4.1411,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2304526748971194,
81
+ "grad_norm": 19.5738582611084,
82
+ "learning_rate": 1.2240437158469946e-05,
83
+ "loss": 4.2919,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3676268861454046,
88
+ "grad_norm": 19.425687789916992,
89
+ "learning_rate": 1.3606557377049181e-05,
90
+ "loss": 4.2991,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.50480109739369,
95
+ "grad_norm": 38.90098190307617,
96
+ "learning_rate": 1.4972677595628417e-05,
97
+ "loss": 4.1817,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6419753086419753,
102
+ "grad_norm": 16.432270050048828,
103
+ "learning_rate": 1.633879781420765e-05,
104
+ "loss": 4.2399,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7791495198902605,
109
+ "grad_norm": 22.444599151611328,
110
+ "learning_rate": 1.7704918032786887e-05,
111
+ "loss": 4.2449,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.916323731138546,
116
+ "grad_norm": 16.111421585083008,
117
+ "learning_rate": 1.9071038251366122e-05,
118
+ "loss": 4.2752,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.46589018302828616,
124
+ "eval_loss": 1.0281691551208496,
125
+ "eval_macro_f1": 0.4635801781458577,
126
+ "eval_runtime": 1.239,
127
+ "eval_samples_per_second": 485.07,
128
+ "eval_steps_per_second": 30.67,
129
+ "eval_weighted_f1": 0.4648406141773152,
130
+ "step": 366
131
+ },
132
+ {
133
+ "epoch": 2.049382716049383,
134
+ "grad_norm": 29.3369083404541,
135
+ "learning_rate": 1.995142683667274e-05,
136
+ "loss": 3.9616,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.186556927297668,
141
+ "grad_norm": 45.82474899291992,
142
+ "learning_rate": 1.9799635701275045e-05,
143
+ "loss": 3.8487,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.3237311385459534,
148
+ "grad_norm": 27.18938636779785,
149
+ "learning_rate": 1.9647844565877354e-05,
150
+ "loss": 3.7389,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.460905349794239,
155
+ "grad_norm": 51.80466079711914,
156
+ "learning_rate": 1.9496053430479663e-05,
157
+ "loss": 3.7166,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.598079561042524,
162
+ "grad_norm": 26.696765899658203,
163
+ "learning_rate": 1.934426229508197e-05,
164
+ "loss": 3.7775,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.7352537722908092,
169
+ "grad_norm": 60.635826110839844,
170
+ "learning_rate": 1.9192471159684274e-05,
171
+ "loss": 3.599,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.8724279835390947,
176
+ "grad_norm": 33.839271545410156,
177
+ "learning_rate": 1.9040680024286583e-05,
178
+ "loss": 3.4726,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5507487520798668,
184
+ "eval_loss": 0.9921584725379944,
185
+ "eval_macro_f1": 0.5493051898295157,
186
+ "eval_runtime": 1.1676,
187
+ "eval_samples_per_second": 514.73,
188
+ "eval_steps_per_second": 32.545,
189
+ "eval_weighted_f1": 0.5504964602744651,
190
+ "step": 549
191
+ },
192
+ {
193
+ "epoch": 3.0054869684499312,
194
+ "grad_norm": 30.09143829345703,
195
+ "learning_rate": 1.888888888888889e-05,
196
+ "loss": 3.3354,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.1426611796982167,
201
+ "grad_norm": 110.91216278076172,
202
+ "learning_rate": 1.8737097753491198e-05,
203
+ "loss": 2.9417,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.279835390946502,
208
+ "grad_norm": 21.85003089904785,
209
+ "learning_rate": 1.8585306618093504e-05,
210
+ "loss": 2.939,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.4170096021947876,
215
+ "grad_norm": 71.07444763183594,
216
+ "learning_rate": 1.8433515482695813e-05,
217
+ "loss": 2.683,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.5541838134430725,
222
+ "grad_norm": 30.45505142211914,
223
+ "learning_rate": 1.828172434729812e-05,
224
+ "loss": 2.8627,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.691358024691358,
229
+ "grad_norm": 45.3055419921875,
230
+ "learning_rate": 1.8129933211900424e-05,
231
+ "loss": 2.7986,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.8285322359396434,
236
+ "grad_norm": 52.458465576171875,
237
+ "learning_rate": 1.7978142076502733e-05,
238
+ "loss": 2.8251,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.9657064471879284,
243
+ "grad_norm": 45.06843948364258,
244
+ "learning_rate": 1.7826350941105042e-05,
245
+ "loss": 2.6718,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 4.0,
250
+ "eval_accuracy": 0.5657237936772047,
251
+ "eval_loss": 1.0419025421142578,
252
+ "eval_macro_f1": 0.5672622157863422,
253
+ "eval_runtime": 1.1579,
254
+ "eval_samples_per_second": 519.024,
255
+ "eval_steps_per_second": 32.817,
256
+ "eval_weighted_f1": 0.5675058657304528,
257
+ "step": 732
258
+ },
259
+ {
260
+ "epoch": 4.098765432098766,
261
+ "grad_norm": 57.640052795410156,
262
+ "learning_rate": 1.7674559805707348e-05,
263
+ "loss": 2.1285,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.23593964334705,
268
+ "grad_norm": 73.37976837158203,
269
+ "learning_rate": 1.7522768670309657e-05,
270
+ "loss": 1.7606,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.373113854595336,
275
+ "grad_norm": 87.4377212524414,
276
+ "learning_rate": 1.7370977534911962e-05,
277
+ "loss": 1.8899,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.510288065843621,
282
+ "grad_norm": 70.71807098388672,
283
+ "learning_rate": 1.7219186399514268e-05,
284
+ "loss": 2.3208,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.647462277091907,
289
+ "grad_norm": 31.578956604003906,
290
+ "learning_rate": 1.7067395264116577e-05,
291
+ "loss": 1.9462,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.784636488340192,
296
+ "grad_norm": 55.558170318603516,
297
+ "learning_rate": 1.6915604128718883e-05,
298
+ "loss": 1.9213,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.921810699588478,
303
+ "grad_norm": 47.71686553955078,
304
+ "learning_rate": 1.6763812993321192e-05,
305
+ "loss": 1.975,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 5.0,
310
+ "eval_accuracy": 0.5457570715474209,
311
+ "eval_loss": 1.3320268392562866,
312
+ "eval_macro_f1": 0.5281132350464214,
313
+ "eval_runtime": 1.3396,
314
+ "eval_samples_per_second": 448.65,
315
+ "eval_steps_per_second": 28.367,
316
+ "eval_weighted_f1": 0.5292764093595372,
317
+ "step": 915
318
+ },
319
+ {
320
+ "epoch": 5.054869684499314,
321
+ "grad_norm": 44.092655181884766,
322
+ "learning_rate": 1.66120218579235e-05,
323
+ "loss": 1.4973,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.1920438957476,
328
+ "grad_norm": 85.77433776855469,
329
+ "learning_rate": 1.6460230722525807e-05,
330
+ "loss": 1.1583,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.329218106995885,
335
+ "grad_norm": 33.370086669921875,
336
+ "learning_rate": 1.6308439587128112e-05,
337
+ "loss": 1.134,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.4663923182441705,
342
+ "grad_norm": 21.9223690032959,
343
+ "learning_rate": 1.615664845173042e-05,
344
+ "loss": 1.0759,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.603566529492455,
349
+ "grad_norm": 65.0184555053711,
350
+ "learning_rate": 1.6004857316332727e-05,
351
+ "loss": 1.1336,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.7407407407407405,
356
+ "grad_norm": 56.2079963684082,
357
+ "learning_rate": 1.5853066180935036e-05,
358
+ "loss": 1.1638,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.877914951989026,
363
+ "grad_norm": 73.81262969970703,
364
+ "learning_rate": 1.570127504553734e-05,
365
+ "loss": 1.0278,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 6.0,
370
+ "eval_accuracy": 0.5607321131447587,
371
+ "eval_loss": 1.4035978317260742,
372
+ "eval_macro_f1": 0.5633862338120106,
373
+ "eval_runtime": 1.2319,
374
+ "eval_samples_per_second": 487.863,
375
+ "eval_steps_per_second": 30.847,
376
+ "eval_weighted_f1": 0.5620942934963105,
377
+ "step": 1098
378
+ },
379
+ {
380
+ "epoch": 6.0109739368998625,
381
+ "grad_norm": 21.437862396240234,
382
+ "learning_rate": 1.5549483910139647e-05,
383
+ "loss": 1.2338,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.148148148148148,
388
+ "grad_norm": 47.56119918823242,
389
+ "learning_rate": 1.5397692774741956e-05,
390
+ "loss": 0.5907,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.285322359396433,
395
+ "grad_norm": 19.970199584960938,
396
+ "learning_rate": 1.5245901639344264e-05,
397
+ "loss": 0.5607,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.422496570644719,
402
+ "grad_norm": 131.32518005371094,
403
+ "learning_rate": 1.5094110503946571e-05,
404
+ "loss": 0.7574,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.559670781893004,
409
+ "grad_norm": 13.367862701416016,
410
+ "learning_rate": 1.4942319368548878e-05,
411
+ "loss": 0.6082,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.69684499314129,
416
+ "grad_norm": 54.434593200683594,
417
+ "learning_rate": 1.4790528233151184e-05,
418
+ "loss": 0.8244,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.834019204389575,
423
+ "grad_norm": 59.72079849243164,
424
+ "learning_rate": 1.4638737097753493e-05,
425
+ "loss": 0.7563,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 6.97119341563786,
430
+ "grad_norm": 31.823333740234375,
431
+ "learning_rate": 1.44869459623558e-05,
432
+ "loss": 0.6605,
433
+ "step": 1275
434
+ },
435
+ {
436
+ "epoch": 7.0,
437
+ "eval_accuracy": 0.56738768718802,
438
+ "eval_loss": 2.001624822616577,
439
+ "eval_macro_f1": 0.5624050129716512,
440
+ "eval_runtime": 1.2542,
441
+ "eval_samples_per_second": 479.204,
442
+ "eval_steps_per_second": 30.299,
443
+ "eval_weighted_f1": 0.5633122153846144,
444
+ "step": 1281
445
+ }
446
+ ],
447
+ "logging_steps": 25,
448
+ "max_steps": 3660,
449
+ "num_input_tokens_seen": 0,
450
+ "num_train_epochs": 20,
451
+ "save_steps": 500,
452
+ "stateful_callbacks": {
453
+ "EarlyStoppingCallback": {
454
+ "args": {
455
+ "early_stopping_patience": 3,
456
+ "early_stopping_threshold": 0.0
457
+ },
458
+ "attributes": {
459
+ "early_stopping_patience_counter": 3
460
+ }
461
+ },
462
+ "TrainerControl": {
463
+ "args": {
464
+ "should_epoch_stop": false,
465
+ "should_evaluate": false,
466
+ "should_log": false,
467
+ "should_save": true,
468
+ "should_training_stop": true
469
+ },
470
+ "attributes": {}
471
+ }
472
+ },
473
+ "total_flos": 0.0,
474
+ "train_batch_size": 8,
475
+ "trial_name": null,
476
+ "trial_params": null
477
+ }
ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-732/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-732/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-732/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-732/trainer_state.json ADDED
@@ -0,0 +1,290 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 732,
3
+ "best_metric": 0.5672622157863422,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-03-s44-20260831T031636Z-59e46264/trainer/checkpoint-732",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 732,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13717421124828533,
14
+ "grad_norm": 23.982624053955078,
15
+ "learning_rate": 1.3114754098360657e-06,
16
+ "loss": 4.7545,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27434842249657065,
21
+ "grad_norm": 22.713624954223633,
22
+ "learning_rate": 2.677595628415301e-06,
23
+ "loss": 4.5062,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.411522633744856,
28
+ "grad_norm": 19.318227767944336,
29
+ "learning_rate": 4.043715846994536e-06,
30
+ "loss": 4.4396,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5486968449931413,
35
+ "grad_norm": 26.55528450012207,
36
+ "learning_rate": 5.409836065573772e-06,
37
+ "loss": 4.4558,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6858710562414266,
42
+ "grad_norm": 16.671968460083008,
43
+ "learning_rate": 6.775956284153006e-06,
44
+ "loss": 4.3536,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.823045267489712,
49
+ "grad_norm": 40.652828216552734,
50
+ "learning_rate": 8.14207650273224e-06,
51
+ "loss": 4.4056,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9602194787379973,
56
+ "grad_norm": 15.629646301269531,
57
+ "learning_rate": 9.508196721311476e-06,
58
+ "loss": 4.3376,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.3910149750415973,
64
+ "eval_loss": 1.0781054496765137,
65
+ "eval_macro_f1": 0.36967599249187333,
66
+ "eval_runtime": 1.3916,
67
+ "eval_samples_per_second": 431.865,
68
+ "eval_steps_per_second": 27.306,
69
+ "eval_weighted_f1": 0.3712625689976963,
70
+ "step": 183
71
+ },
72
+ {
73
+ "epoch": 1.093278463648834,
74
+ "grad_norm": 17.34412384033203,
75
+ "learning_rate": 1.0874316939890712e-05,
76
+ "loss": 4.1411,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2304526748971194,
81
+ "grad_norm": 19.5738582611084,
82
+ "learning_rate": 1.2240437158469946e-05,
83
+ "loss": 4.2919,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3676268861454046,
88
+ "grad_norm": 19.425687789916992,
89
+ "learning_rate": 1.3606557377049181e-05,
90
+ "loss": 4.2991,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.50480109739369,
95
+ "grad_norm": 38.90098190307617,
96
+ "learning_rate": 1.4972677595628417e-05,
97
+ "loss": 4.1817,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6419753086419753,
102
+ "grad_norm": 16.432270050048828,
103
+ "learning_rate": 1.633879781420765e-05,
104
+ "loss": 4.2399,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7791495198902605,
109
+ "grad_norm": 22.444599151611328,
110
+ "learning_rate": 1.7704918032786887e-05,
111
+ "loss": 4.2449,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.916323731138546,
116
+ "grad_norm": 16.111421585083008,
117
+ "learning_rate": 1.9071038251366122e-05,
118
+ "loss": 4.2752,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.46589018302828616,
124
+ "eval_loss": 1.0281691551208496,
125
+ "eval_macro_f1": 0.4635801781458577,
126
+ "eval_runtime": 1.239,
127
+ "eval_samples_per_second": 485.07,
128
+ "eval_steps_per_second": 30.67,
129
+ "eval_weighted_f1": 0.4648406141773152,
130
+ "step": 366
131
+ },
132
+ {
133
+ "epoch": 2.049382716049383,
134
+ "grad_norm": 29.3369083404541,
135
+ "learning_rate": 1.995142683667274e-05,
136
+ "loss": 3.9616,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.186556927297668,
141
+ "grad_norm": 45.82474899291992,
142
+ "learning_rate": 1.9799635701275045e-05,
143
+ "loss": 3.8487,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.3237311385459534,
148
+ "grad_norm": 27.18938636779785,
149
+ "learning_rate": 1.9647844565877354e-05,
150
+ "loss": 3.7389,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.460905349794239,
155
+ "grad_norm": 51.80466079711914,
156
+ "learning_rate": 1.9496053430479663e-05,
157
+ "loss": 3.7166,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.598079561042524,
162
+ "grad_norm": 26.696765899658203,
163
+ "learning_rate": 1.934426229508197e-05,
164
+ "loss": 3.7775,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.7352537722908092,
169
+ "grad_norm": 60.635826110839844,
170
+ "learning_rate": 1.9192471159684274e-05,
171
+ "loss": 3.599,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.8724279835390947,
176
+ "grad_norm": 33.839271545410156,
177
+ "learning_rate": 1.9040680024286583e-05,
178
+ "loss": 3.4726,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5507487520798668,
184
+ "eval_loss": 0.9921584725379944,
185
+ "eval_macro_f1": 0.5493051898295157,
186
+ "eval_runtime": 1.1676,
187
+ "eval_samples_per_second": 514.73,
188
+ "eval_steps_per_second": 32.545,
189
+ "eval_weighted_f1": 0.5504964602744651,
190
+ "step": 549
191
+ },
192
+ {
193
+ "epoch": 3.0054869684499312,
194
+ "grad_norm": 30.09143829345703,
195
+ "learning_rate": 1.888888888888889e-05,
196
+ "loss": 3.3354,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.1426611796982167,
201
+ "grad_norm": 110.91216278076172,
202
+ "learning_rate": 1.8737097753491198e-05,
203
+ "loss": 2.9417,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.279835390946502,
208
+ "grad_norm": 21.85003089904785,
209
+ "learning_rate": 1.8585306618093504e-05,
210
+ "loss": 2.939,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.4170096021947876,
215
+ "grad_norm": 71.07444763183594,
216
+ "learning_rate": 1.8433515482695813e-05,
217
+ "loss": 2.683,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.5541838134430725,
222
+ "grad_norm": 30.45505142211914,
223
+ "learning_rate": 1.828172434729812e-05,
224
+ "loss": 2.8627,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.691358024691358,
229
+ "grad_norm": 45.3055419921875,
230
+ "learning_rate": 1.8129933211900424e-05,
231
+ "loss": 2.7986,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.8285322359396434,
236
+ "grad_norm": 52.458465576171875,
237
+ "learning_rate": 1.7978142076502733e-05,
238
+ "loss": 2.8251,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.9657064471879284,
243
+ "grad_norm": 45.06843948364258,
244
+ "learning_rate": 1.7826350941105042e-05,
245
+ "loss": 2.6718,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 4.0,
250
+ "eval_accuracy": 0.5657237936772047,
251
+ "eval_loss": 1.0419025421142578,
252
+ "eval_macro_f1": 0.5672622157863422,
253
+ "eval_runtime": 1.1579,
254
+ "eval_samples_per_second": 519.024,
255
+ "eval_steps_per_second": 32.817,
256
+ "eval_weighted_f1": 0.5675058657304528,
257
+ "step": 732
258
+ }
259
+ ],
260
+ "logging_steps": 25,
261
+ "max_steps": 3660,
262
+ "num_input_tokens_seen": 0,
263
+ "num_train_epochs": 20,
264
+ "save_steps": 500,
265
+ "stateful_callbacks": {
266
+ "EarlyStoppingCallback": {
267
+ "args": {
268
+ "early_stopping_patience": 3,
269
+ "early_stopping_threshold": 0.0
270
+ },
271
+ "attributes": {
272
+ "early_stopping_patience_counter": 0
273
+ }
274
+ },
275
+ "TrainerControl": {
276
+ "args": {
277
+ "should_epoch_stop": false,
278
+ "should_evaluate": false,
279
+ "should_log": false,
280
+ "should_save": true,
281
+ "should_training_stop": false
282
+ },
283
+ "attributes": {}
284
+ }
285
+ },
286
+ "total_flos": 0.0,
287
+ "train_batch_size": 8,
288
+ "trial_name": null,
289
+ "trial_params": null
290
+ }
ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1408/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1408/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1408/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1408/trainer_state.json ADDED
@@ -0,0 +1,523 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1408,
3
+ "best_metric": 0.6019414574728712,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1408",
5
+ "epoch": 8.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1408,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.14245014245014245,
14
+ "grad_norm": 24.046409606933594,
15
+ "learning_rate": 1.3636363636363636e-06,
16
+ "loss": 4.7322,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.2849002849002849,
21
+ "grad_norm": 35.91823196411133,
22
+ "learning_rate": 2.784090909090909e-06,
23
+ "loss": 4.5454,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.42735042735042733,
28
+ "grad_norm": 26.33811378479004,
29
+ "learning_rate": 4.204545454545455e-06,
30
+ "loss": 4.4269,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5698005698005698,
35
+ "grad_norm": 33.35604476928711,
36
+ "learning_rate": 5.625e-06,
37
+ "loss": 4.423,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.7122507122507122,
42
+ "grad_norm": 24.652359008789062,
43
+ "learning_rate": 7.045454545454546e-06,
44
+ "loss": 4.4446,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8547008547008547,
49
+ "grad_norm": 21.109880447387695,
50
+ "learning_rate": 8.465909090909091e-06,
51
+ "loss": 4.383,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9971509971509972,
56
+ "grad_norm": 22.176088333129883,
57
+ "learning_rate": 9.886363636363637e-06,
58
+ "loss": 4.3153,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.44992743105950656,
64
+ "eval_loss": 1.0499252080917358,
65
+ "eval_macro_f1": 0.447329374957708,
66
+ "eval_runtime": 1.3891,
67
+ "eval_samples_per_second": 496.017,
68
+ "eval_steps_per_second": 31.676,
69
+ "eval_weighted_f1": 0.44867024259786953,
70
+ "step": 176
71
+ },
72
+ {
73
+ "epoch": 1.1367521367521367,
74
+ "grad_norm": 28.080364227294922,
75
+ "learning_rate": 1.1306818181818183e-05,
76
+ "loss": 4.2103,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2792022792022792,
81
+ "grad_norm": 30.090457916259766,
82
+ "learning_rate": 1.2727272727272728e-05,
83
+ "loss": 4.2303,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.4216524216524218,
88
+ "grad_norm": 28.65721893310547,
89
+ "learning_rate": 1.4147727272727274e-05,
90
+ "loss": 4.2216,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.564102564102564,
95
+ "grad_norm": 15.442717552185059,
96
+ "learning_rate": 1.5568181818181822e-05,
97
+ "loss": 4.2865,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.7065527065527064,
102
+ "grad_norm": 20.929706573486328,
103
+ "learning_rate": 1.6988636363636368e-05,
104
+ "loss": 4.1939,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.849002849002849,
109
+ "grad_norm": 20.27390480041504,
110
+ "learning_rate": 1.840909090909091e-05,
111
+ "loss": 4.2064,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.9914529914529915,
116
+ "grad_norm": 28.332134246826172,
117
+ "learning_rate": 1.9829545454545456e-05,
118
+ "loss": 4.1528,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.42670537010159654,
124
+ "eval_loss": 1.1095401048660278,
125
+ "eval_macro_f1": 0.3499269020562105,
126
+ "eval_runtime": 1.3146,
127
+ "eval_samples_per_second": 524.13,
128
+ "eval_steps_per_second": 33.471,
129
+ "eval_weighted_f1": 0.35553018850038615,
130
+ "step": 352
131
+ },
132
+ {
133
+ "epoch": 2.131054131054131,
134
+ "grad_norm": 22.95296859741211,
135
+ "learning_rate": 1.9861111111111114e-05,
136
+ "loss": 3.723,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.2735042735042734,
141
+ "grad_norm": 29.088937759399414,
142
+ "learning_rate": 1.9703282828282832e-05,
143
+ "loss": 3.725,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.415954415954416,
148
+ "grad_norm": 40.377933502197266,
149
+ "learning_rate": 1.9545454545454546e-05,
150
+ "loss": 3.6928,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.5584045584045585,
155
+ "grad_norm": 36.57823944091797,
156
+ "learning_rate": 1.9387626262626263e-05,
157
+ "loss": 3.9008,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.700854700854701,
162
+ "grad_norm": 24.683317184448242,
163
+ "learning_rate": 1.922979797979798e-05,
164
+ "loss": 3.7013,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.8433048433048436,
169
+ "grad_norm": 52.79389572143555,
170
+ "learning_rate": 1.90719696969697e-05,
171
+ "loss": 3.5975,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.9857549857549857,
176
+ "grad_norm": 53.46162796020508,
177
+ "learning_rate": 1.8914141414141416e-05,
178
+ "loss": 3.5438,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5355587808417998,
184
+ "eval_loss": 0.9561595916748047,
185
+ "eval_macro_f1": 0.5306337953646169,
186
+ "eval_runtime": 1.3614,
187
+ "eval_samples_per_second": 506.099,
188
+ "eval_steps_per_second": 32.32,
189
+ "eval_weighted_f1": 0.53180268253649,
190
+ "step": 528
191
+ },
192
+ {
193
+ "epoch": 3.1253561253561255,
194
+ "grad_norm": 54.70162582397461,
195
+ "learning_rate": 1.8756313131313133e-05,
196
+ "loss": 2.9987,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.267806267806268,
201
+ "grad_norm": 24.91816520690918,
202
+ "learning_rate": 1.859848484848485e-05,
203
+ "loss": 3.0628,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.41025641025641,
208
+ "grad_norm": 22.032522201538086,
209
+ "learning_rate": 1.844065656565657e-05,
210
+ "loss": 2.8255,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.5527065527065527,
215
+ "grad_norm": 55.444007873535156,
216
+ "learning_rate": 1.8282828282828286e-05,
217
+ "loss": 2.7891,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.695156695156695,
222
+ "grad_norm": 32.102508544921875,
223
+ "learning_rate": 1.8125e-05,
224
+ "loss": 3.0519,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.8376068376068377,
229
+ "grad_norm": 30.49248695373535,
230
+ "learning_rate": 1.7967171717171718e-05,
231
+ "loss": 2.9789,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.98005698005698,
236
+ "grad_norm": 78.2519760131836,
237
+ "learning_rate": 1.7809343434343435e-05,
238
+ "loss": 2.7117,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 4.0,
243
+ "eval_accuracy": 0.5341074020319303,
244
+ "eval_loss": 1.1482720375061035,
245
+ "eval_macro_f1": 0.5162401602808634,
246
+ "eval_runtime": 1.3158,
247
+ "eval_samples_per_second": 523.64,
248
+ "eval_steps_per_second": 33.44,
249
+ "eval_weighted_f1": 0.518799924209928,
250
+ "step": 704
251
+ },
252
+ {
253
+ "epoch": 4.119658119658119,
254
+ "grad_norm": 66.05480194091797,
255
+ "learning_rate": 1.7651515151515153e-05,
256
+ "loss": 2.3153,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.262108262108262,
261
+ "grad_norm": 39.6175422668457,
262
+ "learning_rate": 1.749368686868687e-05,
263
+ "loss": 2.2213,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.404558404558404,
268
+ "grad_norm": 47.88081359863281,
269
+ "learning_rate": 1.7335858585858588e-05,
270
+ "loss": 1.9793,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.547008547008547,
275
+ "grad_norm": 89.85733032226562,
276
+ "learning_rate": 1.7178030303030305e-05,
277
+ "loss": 1.9286,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.689458689458689,
282
+ "grad_norm": 31.8492374420166,
283
+ "learning_rate": 1.7020202020202023e-05,
284
+ "loss": 1.939,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.831908831908832,
289
+ "grad_norm": 83.04368591308594,
290
+ "learning_rate": 1.6862373737373737e-05,
291
+ "loss": 2.0258,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.9743589743589745,
296
+ "grad_norm": 44.39993667602539,
297
+ "learning_rate": 1.6704545454545454e-05,
298
+ "loss": 2.0225,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 5.0,
303
+ "eval_accuracy": 0.5674891146589259,
304
+ "eval_loss": 1.1989282369613647,
305
+ "eval_macro_f1": 0.568274105257835,
306
+ "eval_runtime": 1.3692,
307
+ "eval_samples_per_second": 503.227,
308
+ "eval_steps_per_second": 32.136,
309
+ "eval_weighted_f1": 0.5685611874763409,
310
+ "step": 880
311
+ },
312
+ {
313
+ "epoch": 5.113960113960114,
314
+ "grad_norm": 58.37739562988281,
315
+ "learning_rate": 1.6546717171717172e-05,
316
+ "loss": 1.2801,
317
+ "step": 900
318
+ },
319
+ {
320
+ "epoch": 5.256410256410256,
321
+ "grad_norm": 62.308109283447266,
322
+ "learning_rate": 1.638888888888889e-05,
323
+ "loss": 1.0192,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.398860398860399,
328
+ "grad_norm": 45.375205993652344,
329
+ "learning_rate": 1.6231060606060607e-05,
330
+ "loss": 1.2153,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.5413105413105415,
335
+ "grad_norm": 51.65726089477539,
336
+ "learning_rate": 1.6073232323232324e-05,
337
+ "loss": 1.2168,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.683760683760684,
342
+ "grad_norm": 55.187965393066406,
343
+ "learning_rate": 1.5915404040404042e-05,
344
+ "loss": 1.2306,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.8262108262108265,
349
+ "grad_norm": 85.41111755371094,
350
+ "learning_rate": 1.575757575757576e-05,
351
+ "loss": 1.506,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.968660968660968,
356
+ "grad_norm": 111.59111022949219,
357
+ "learning_rate": 1.5599747474747477e-05,
358
+ "loss": 1.2585,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 6.0,
363
+ "eval_accuracy": 0.5703918722786647,
364
+ "eval_loss": 1.3295668363571167,
365
+ "eval_macro_f1": 0.5733791811158245,
366
+ "eval_runtime": 1.3484,
367
+ "eval_samples_per_second": 510.972,
368
+ "eval_steps_per_second": 32.631,
369
+ "eval_weighted_f1": 0.5734591376305213,
370
+ "step": 1056
371
+ },
372
+ {
373
+ "epoch": 6.1082621082621085,
374
+ "grad_norm": 109.14195251464844,
375
+ "learning_rate": 1.544191919191919e-05,
376
+ "loss": 1.0153,
377
+ "step": 1075
378
+ },
379
+ {
380
+ "epoch": 6.250712250712251,
381
+ "grad_norm": 25.342824935913086,
382
+ "learning_rate": 1.528409090909091e-05,
383
+ "loss": 0.7578,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.3931623931623935,
388
+ "grad_norm": 61.64923095703125,
389
+ "learning_rate": 1.5126262626262626e-05,
390
+ "loss": 0.6751,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.535612535612536,
395
+ "grad_norm": 43.23941421508789,
396
+ "learning_rate": 1.4968434343434344e-05,
397
+ "loss": 0.7377,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.678062678062678,
402
+ "grad_norm": 72.0482177734375,
403
+ "learning_rate": 1.4810606060606061e-05,
404
+ "loss": 0.6318,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.82051282051282,
409
+ "grad_norm": 52.30923843383789,
410
+ "learning_rate": 1.4652777777777779e-05,
411
+ "loss": 0.7117,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.962962962962963,
416
+ "grad_norm": 51.14097213745117,
417
+ "learning_rate": 1.4494949494949494e-05,
418
+ "loss": 0.9345,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 7.0,
423
+ "eval_accuracy": 0.5645863570391872,
424
+ "eval_loss": 1.662731647491455,
425
+ "eval_macro_f1": 0.5670266886831596,
426
+ "eval_runtime": 1.3406,
427
+ "eval_samples_per_second": 513.959,
428
+ "eval_steps_per_second": 32.822,
429
+ "eval_weighted_f1": 0.567031343683912,
430
+ "step": 1232
431
+ },
432
+ {
433
+ "epoch": 7.102564102564102,
434
+ "grad_norm": 34.256412506103516,
435
+ "learning_rate": 1.4337121212121212e-05,
436
+ "loss": 0.5987,
437
+ "step": 1250
438
+ },
439
+ {
440
+ "epoch": 7.245014245014245,
441
+ "grad_norm": 54.797767639160156,
442
+ "learning_rate": 1.4179292929292931e-05,
443
+ "loss": 0.3503,
444
+ "step": 1275
445
+ },
446
+ {
447
+ "epoch": 7.387464387464387,
448
+ "grad_norm": 6.712471961975098,
449
+ "learning_rate": 1.4021464646464649e-05,
450
+ "loss": 0.4312,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.52991452991453,
455
+ "grad_norm": 45.92087936401367,
456
+ "learning_rate": 1.3863636363636366e-05,
457
+ "loss": 0.4082,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.672364672364672,
462
+ "grad_norm": 182.13890075683594,
463
+ "learning_rate": 1.3705808080808084e-05,
464
+ "loss": 0.4361,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.814814814814815,
469
+ "grad_norm": 90.058349609375,
470
+ "learning_rate": 1.35479797979798e-05,
471
+ "loss": 0.4911,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.957264957264957,
476
+ "grad_norm": 51.1805534362793,
477
+ "learning_rate": 1.3390151515151517e-05,
478
+ "loss": 0.526,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 8.0,
483
+ "eval_accuracy": 0.602322206095791,
484
+ "eval_loss": 2.0464773178100586,
485
+ "eval_macro_f1": 0.6019414574728712,
486
+ "eval_runtime": 1.3653,
487
+ "eval_samples_per_second": 504.636,
488
+ "eval_steps_per_second": 32.226,
489
+ "eval_weighted_f1": 0.6020224562932764,
490
+ "step": 1408
491
+ }
492
+ ],
493
+ "logging_steps": 25,
494
+ "max_steps": 3520,
495
+ "num_input_tokens_seen": 0,
496
+ "num_train_epochs": 20,
497
+ "save_steps": 500,
498
+ "stateful_callbacks": {
499
+ "EarlyStoppingCallback": {
500
+ "args": {
501
+ "early_stopping_patience": 3,
502
+ "early_stopping_threshold": 0.0
503
+ },
504
+ "attributes": {
505
+ "early_stopping_patience_counter": 0
506
+ }
507
+ },
508
+ "TrainerControl": {
509
+ "args": {
510
+ "should_epoch_stop": false,
511
+ "should_evaluate": false,
512
+ "should_log": false,
513
+ "should_save": true,
514
+ "should_training_stop": false
515
+ },
516
+ "attributes": {}
517
+ }
518
+ },
519
+ "total_flos": 0.0,
520
+ "train_batch_size": 8,
521
+ "trial_name": null,
522
+ "trial_params": null
523
+ }
ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1936/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1936/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1936/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1936/trainer_state.json ADDED
@@ -0,0 +1,703 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1408,
3
+ "best_metric": 0.6019414574728712,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-04-s44-20260831T032410Z-59e46264/trainer/checkpoint-1408",
5
+ "epoch": 11.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1936,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.14245014245014245,
14
+ "grad_norm": 24.046409606933594,
15
+ "learning_rate": 1.3636363636363636e-06,
16
+ "loss": 4.7322,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.2849002849002849,
21
+ "grad_norm": 35.91823196411133,
22
+ "learning_rate": 2.784090909090909e-06,
23
+ "loss": 4.5454,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.42735042735042733,
28
+ "grad_norm": 26.33811378479004,
29
+ "learning_rate": 4.204545454545455e-06,
30
+ "loss": 4.4269,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5698005698005698,
35
+ "grad_norm": 33.35604476928711,
36
+ "learning_rate": 5.625e-06,
37
+ "loss": 4.423,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.7122507122507122,
42
+ "grad_norm": 24.652359008789062,
43
+ "learning_rate": 7.045454545454546e-06,
44
+ "loss": 4.4446,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8547008547008547,
49
+ "grad_norm": 21.109880447387695,
50
+ "learning_rate": 8.465909090909091e-06,
51
+ "loss": 4.383,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9971509971509972,
56
+ "grad_norm": 22.176088333129883,
57
+ "learning_rate": 9.886363636363637e-06,
58
+ "loss": 4.3153,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.44992743105950656,
64
+ "eval_loss": 1.0499252080917358,
65
+ "eval_macro_f1": 0.447329374957708,
66
+ "eval_runtime": 1.3891,
67
+ "eval_samples_per_second": 496.017,
68
+ "eval_steps_per_second": 31.676,
69
+ "eval_weighted_f1": 0.44867024259786953,
70
+ "step": 176
71
+ },
72
+ {
73
+ "epoch": 1.1367521367521367,
74
+ "grad_norm": 28.080364227294922,
75
+ "learning_rate": 1.1306818181818183e-05,
76
+ "loss": 4.2103,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2792022792022792,
81
+ "grad_norm": 30.090457916259766,
82
+ "learning_rate": 1.2727272727272728e-05,
83
+ "loss": 4.2303,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.4216524216524218,
88
+ "grad_norm": 28.65721893310547,
89
+ "learning_rate": 1.4147727272727274e-05,
90
+ "loss": 4.2216,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.564102564102564,
95
+ "grad_norm": 15.442717552185059,
96
+ "learning_rate": 1.5568181818181822e-05,
97
+ "loss": 4.2865,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.7065527065527064,
102
+ "grad_norm": 20.929706573486328,
103
+ "learning_rate": 1.6988636363636368e-05,
104
+ "loss": 4.1939,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.849002849002849,
109
+ "grad_norm": 20.27390480041504,
110
+ "learning_rate": 1.840909090909091e-05,
111
+ "loss": 4.2064,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.9914529914529915,
116
+ "grad_norm": 28.332134246826172,
117
+ "learning_rate": 1.9829545454545456e-05,
118
+ "loss": 4.1528,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.42670537010159654,
124
+ "eval_loss": 1.1095401048660278,
125
+ "eval_macro_f1": 0.3499269020562105,
126
+ "eval_runtime": 1.3146,
127
+ "eval_samples_per_second": 524.13,
128
+ "eval_steps_per_second": 33.471,
129
+ "eval_weighted_f1": 0.35553018850038615,
130
+ "step": 352
131
+ },
132
+ {
133
+ "epoch": 2.131054131054131,
134
+ "grad_norm": 22.95296859741211,
135
+ "learning_rate": 1.9861111111111114e-05,
136
+ "loss": 3.723,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.2735042735042734,
141
+ "grad_norm": 29.088937759399414,
142
+ "learning_rate": 1.9703282828282832e-05,
143
+ "loss": 3.725,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.415954415954416,
148
+ "grad_norm": 40.377933502197266,
149
+ "learning_rate": 1.9545454545454546e-05,
150
+ "loss": 3.6928,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.5584045584045585,
155
+ "grad_norm": 36.57823944091797,
156
+ "learning_rate": 1.9387626262626263e-05,
157
+ "loss": 3.9008,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.700854700854701,
162
+ "grad_norm": 24.683317184448242,
163
+ "learning_rate": 1.922979797979798e-05,
164
+ "loss": 3.7013,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.8433048433048436,
169
+ "grad_norm": 52.79389572143555,
170
+ "learning_rate": 1.90719696969697e-05,
171
+ "loss": 3.5975,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.9857549857549857,
176
+ "grad_norm": 53.46162796020508,
177
+ "learning_rate": 1.8914141414141416e-05,
178
+ "loss": 3.5438,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5355587808417998,
184
+ "eval_loss": 0.9561595916748047,
185
+ "eval_macro_f1": 0.5306337953646169,
186
+ "eval_runtime": 1.3614,
187
+ "eval_samples_per_second": 506.099,
188
+ "eval_steps_per_second": 32.32,
189
+ "eval_weighted_f1": 0.53180268253649,
190
+ "step": 528
191
+ },
192
+ {
193
+ "epoch": 3.1253561253561255,
194
+ "grad_norm": 54.70162582397461,
195
+ "learning_rate": 1.8756313131313133e-05,
196
+ "loss": 2.9987,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.267806267806268,
201
+ "grad_norm": 24.91816520690918,
202
+ "learning_rate": 1.859848484848485e-05,
203
+ "loss": 3.0628,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.41025641025641,
208
+ "grad_norm": 22.032522201538086,
209
+ "learning_rate": 1.844065656565657e-05,
210
+ "loss": 2.8255,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.5527065527065527,
215
+ "grad_norm": 55.444007873535156,
216
+ "learning_rate": 1.8282828282828286e-05,
217
+ "loss": 2.7891,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.695156695156695,
222
+ "grad_norm": 32.102508544921875,
223
+ "learning_rate": 1.8125e-05,
224
+ "loss": 3.0519,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.8376068376068377,
229
+ "grad_norm": 30.49248695373535,
230
+ "learning_rate": 1.7967171717171718e-05,
231
+ "loss": 2.9789,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.98005698005698,
236
+ "grad_norm": 78.2519760131836,
237
+ "learning_rate": 1.7809343434343435e-05,
238
+ "loss": 2.7117,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 4.0,
243
+ "eval_accuracy": 0.5341074020319303,
244
+ "eval_loss": 1.1482720375061035,
245
+ "eval_macro_f1": 0.5162401602808634,
246
+ "eval_runtime": 1.3158,
247
+ "eval_samples_per_second": 523.64,
248
+ "eval_steps_per_second": 33.44,
249
+ "eval_weighted_f1": 0.518799924209928,
250
+ "step": 704
251
+ },
252
+ {
253
+ "epoch": 4.119658119658119,
254
+ "grad_norm": 66.05480194091797,
255
+ "learning_rate": 1.7651515151515153e-05,
256
+ "loss": 2.3153,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.262108262108262,
261
+ "grad_norm": 39.6175422668457,
262
+ "learning_rate": 1.749368686868687e-05,
263
+ "loss": 2.2213,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.404558404558404,
268
+ "grad_norm": 47.88081359863281,
269
+ "learning_rate": 1.7335858585858588e-05,
270
+ "loss": 1.9793,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.547008547008547,
275
+ "grad_norm": 89.85733032226562,
276
+ "learning_rate": 1.7178030303030305e-05,
277
+ "loss": 1.9286,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.689458689458689,
282
+ "grad_norm": 31.8492374420166,
283
+ "learning_rate": 1.7020202020202023e-05,
284
+ "loss": 1.939,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.831908831908832,
289
+ "grad_norm": 83.04368591308594,
290
+ "learning_rate": 1.6862373737373737e-05,
291
+ "loss": 2.0258,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.9743589743589745,
296
+ "grad_norm": 44.39993667602539,
297
+ "learning_rate": 1.6704545454545454e-05,
298
+ "loss": 2.0225,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 5.0,
303
+ "eval_accuracy": 0.5674891146589259,
304
+ "eval_loss": 1.1989282369613647,
305
+ "eval_macro_f1": 0.568274105257835,
306
+ "eval_runtime": 1.3692,
307
+ "eval_samples_per_second": 503.227,
308
+ "eval_steps_per_second": 32.136,
309
+ "eval_weighted_f1": 0.5685611874763409,
310
+ "step": 880
311
+ },
312
+ {
313
+ "epoch": 5.113960113960114,
314
+ "grad_norm": 58.37739562988281,
315
+ "learning_rate": 1.6546717171717172e-05,
316
+ "loss": 1.2801,
317
+ "step": 900
318
+ },
319
+ {
320
+ "epoch": 5.256410256410256,
321
+ "grad_norm": 62.308109283447266,
322
+ "learning_rate": 1.638888888888889e-05,
323
+ "loss": 1.0192,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.398860398860399,
328
+ "grad_norm": 45.375205993652344,
329
+ "learning_rate": 1.6231060606060607e-05,
330
+ "loss": 1.2153,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.5413105413105415,
335
+ "grad_norm": 51.65726089477539,
336
+ "learning_rate": 1.6073232323232324e-05,
337
+ "loss": 1.2168,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.683760683760684,
342
+ "grad_norm": 55.187965393066406,
343
+ "learning_rate": 1.5915404040404042e-05,
344
+ "loss": 1.2306,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.8262108262108265,
349
+ "grad_norm": 85.41111755371094,
350
+ "learning_rate": 1.575757575757576e-05,
351
+ "loss": 1.506,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.968660968660968,
356
+ "grad_norm": 111.59111022949219,
357
+ "learning_rate": 1.5599747474747477e-05,
358
+ "loss": 1.2585,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 6.0,
363
+ "eval_accuracy": 0.5703918722786647,
364
+ "eval_loss": 1.3295668363571167,
365
+ "eval_macro_f1": 0.5733791811158245,
366
+ "eval_runtime": 1.3484,
367
+ "eval_samples_per_second": 510.972,
368
+ "eval_steps_per_second": 32.631,
369
+ "eval_weighted_f1": 0.5734591376305213,
370
+ "step": 1056
371
+ },
372
+ {
373
+ "epoch": 6.1082621082621085,
374
+ "grad_norm": 109.14195251464844,
375
+ "learning_rate": 1.544191919191919e-05,
376
+ "loss": 1.0153,
377
+ "step": 1075
378
+ },
379
+ {
380
+ "epoch": 6.250712250712251,
381
+ "grad_norm": 25.342824935913086,
382
+ "learning_rate": 1.528409090909091e-05,
383
+ "loss": 0.7578,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.3931623931623935,
388
+ "grad_norm": 61.64923095703125,
389
+ "learning_rate": 1.5126262626262626e-05,
390
+ "loss": 0.6751,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.535612535612536,
395
+ "grad_norm": 43.23941421508789,
396
+ "learning_rate": 1.4968434343434344e-05,
397
+ "loss": 0.7377,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.678062678062678,
402
+ "grad_norm": 72.0482177734375,
403
+ "learning_rate": 1.4810606060606061e-05,
404
+ "loss": 0.6318,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.82051282051282,
409
+ "grad_norm": 52.30923843383789,
410
+ "learning_rate": 1.4652777777777779e-05,
411
+ "loss": 0.7117,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.962962962962963,
416
+ "grad_norm": 51.14097213745117,
417
+ "learning_rate": 1.4494949494949494e-05,
418
+ "loss": 0.9345,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 7.0,
423
+ "eval_accuracy": 0.5645863570391872,
424
+ "eval_loss": 1.662731647491455,
425
+ "eval_macro_f1": 0.5670266886831596,
426
+ "eval_runtime": 1.3406,
427
+ "eval_samples_per_second": 513.959,
428
+ "eval_steps_per_second": 32.822,
429
+ "eval_weighted_f1": 0.567031343683912,
430
+ "step": 1232
431
+ },
432
+ {
433
+ "epoch": 7.102564102564102,
434
+ "grad_norm": 34.256412506103516,
435
+ "learning_rate": 1.4337121212121212e-05,
436
+ "loss": 0.5987,
437
+ "step": 1250
438
+ },
439
+ {
440
+ "epoch": 7.245014245014245,
441
+ "grad_norm": 54.797767639160156,
442
+ "learning_rate": 1.4179292929292931e-05,
443
+ "loss": 0.3503,
444
+ "step": 1275
445
+ },
446
+ {
447
+ "epoch": 7.387464387464387,
448
+ "grad_norm": 6.712471961975098,
449
+ "learning_rate": 1.4021464646464649e-05,
450
+ "loss": 0.4312,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.52991452991453,
455
+ "grad_norm": 45.92087936401367,
456
+ "learning_rate": 1.3863636363636366e-05,
457
+ "loss": 0.4082,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.672364672364672,
462
+ "grad_norm": 182.13890075683594,
463
+ "learning_rate": 1.3705808080808084e-05,
464
+ "loss": 0.4361,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.814814814814815,
469
+ "grad_norm": 90.058349609375,
470
+ "learning_rate": 1.35479797979798e-05,
471
+ "loss": 0.4911,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.957264957264957,
476
+ "grad_norm": 51.1805534362793,
477
+ "learning_rate": 1.3390151515151517e-05,
478
+ "loss": 0.526,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 8.0,
483
+ "eval_accuracy": 0.602322206095791,
484
+ "eval_loss": 2.0464773178100586,
485
+ "eval_macro_f1": 0.6019414574728712,
486
+ "eval_runtime": 1.3653,
487
+ "eval_samples_per_second": 504.636,
488
+ "eval_steps_per_second": 32.226,
489
+ "eval_weighted_f1": 0.6020224562932764,
490
+ "step": 1408
491
+ },
492
+ {
493
+ "epoch": 8.096866096866098,
494
+ "grad_norm": 59.06166458129883,
495
+ "learning_rate": 1.3232323232323234e-05,
496
+ "loss": 0.2779,
497
+ "step": 1425
498
+ },
499
+ {
500
+ "epoch": 8.239316239316238,
501
+ "grad_norm": 94.11392974853516,
502
+ "learning_rate": 1.3074494949494952e-05,
503
+ "loss": 0.3237,
504
+ "step": 1450
505
+ },
506
+ {
507
+ "epoch": 8.381766381766381,
508
+ "grad_norm": 93.98200225830078,
509
+ "learning_rate": 1.2916666666666668e-05,
510
+ "loss": 0.2647,
511
+ "step": 1475
512
+ },
513
+ {
514
+ "epoch": 8.524216524216524,
515
+ "grad_norm": 83.4825439453125,
516
+ "learning_rate": 1.2758838383838385e-05,
517
+ "loss": 0.3104,
518
+ "step": 1500
519
+ },
520
+ {
521
+ "epoch": 8.666666666666666,
522
+ "grad_norm": 51.956661224365234,
523
+ "learning_rate": 1.2601010101010103e-05,
524
+ "loss": 0.4436,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 8.809116809116809,
529
+ "grad_norm": 93.28397369384766,
530
+ "learning_rate": 1.244318181818182e-05,
531
+ "loss": 0.4303,
532
+ "step": 1550
533
+ },
534
+ {
535
+ "epoch": 8.951566951566951,
536
+ "grad_norm": 144.07443237304688,
537
+ "learning_rate": 1.2285353535353536e-05,
538
+ "loss": 0.2386,
539
+ "step": 1575
540
+ },
541
+ {
542
+ "epoch": 9.0,
543
+ "eval_accuracy": 0.5515239477503628,
544
+ "eval_loss": 2.749403953552246,
545
+ "eval_macro_f1": 0.5505863893733007,
546
+ "eval_runtime": 1.471,
547
+ "eval_samples_per_second": 468.377,
548
+ "eval_steps_per_second": 29.911,
549
+ "eval_weighted_f1": 0.5516038613776756,
550
+ "step": 1584
551
+ },
552
+ {
553
+ "epoch": 9.091168091168091,
554
+ "grad_norm": 25.14110565185547,
555
+ "learning_rate": 1.2127525252525254e-05,
556
+ "loss": 0.2385,
557
+ "step": 1600
558
+ },
559
+ {
560
+ "epoch": 9.233618233618234,
561
+ "grad_norm": 68.24052429199219,
562
+ "learning_rate": 1.1969696969696971e-05,
563
+ "loss": 0.2022,
564
+ "step": 1625
565
+ },
566
+ {
567
+ "epoch": 9.376068376068377,
568
+ "grad_norm": 47.74794387817383,
569
+ "learning_rate": 1.1811868686868689e-05,
570
+ "loss": 0.138,
571
+ "step": 1650
572
+ },
573
+ {
574
+ "epoch": 9.518518518518519,
575
+ "grad_norm": 77.67135620117188,
576
+ "learning_rate": 1.1654040404040406e-05,
577
+ "loss": 0.212,
578
+ "step": 1675
579
+ },
580
+ {
581
+ "epoch": 9.660968660968662,
582
+ "grad_norm": 9.56105899810791,
583
+ "learning_rate": 1.1496212121212122e-05,
584
+ "loss": 0.4081,
585
+ "step": 1700
586
+ },
587
+ {
588
+ "epoch": 9.803418803418804,
589
+ "grad_norm": 0.683318018913269,
590
+ "learning_rate": 1.133838383838384e-05,
591
+ "loss": 0.1678,
592
+ "step": 1725
593
+ },
594
+ {
595
+ "epoch": 9.945868945868947,
596
+ "grad_norm": 6.2471818923950195,
597
+ "learning_rate": 1.1180555555555557e-05,
598
+ "loss": 0.2091,
599
+ "step": 1750
600
+ },
601
+ {
602
+ "epoch": 10.0,
603
+ "eval_accuracy": 0.5805515239477503,
604
+ "eval_loss": 3.001067876815796,
605
+ "eval_macro_f1": 0.56904222491609,
606
+ "eval_runtime": 1.3849,
607
+ "eval_samples_per_second": 497.522,
608
+ "eval_steps_per_second": 31.772,
609
+ "eval_weighted_f1": 0.5697014020293929,
610
+ "step": 1760
611
+ },
612
+ {
613
+ "epoch": 10.085470085470085,
614
+ "grad_norm": 208.6376495361328,
615
+ "learning_rate": 1.1022727272727275e-05,
616
+ "loss": 0.1453,
617
+ "step": 1775
618
+ },
619
+ {
620
+ "epoch": 10.227920227920228,
621
+ "grad_norm": 32.24196243286133,
622
+ "learning_rate": 1.086489898989899e-05,
623
+ "loss": 0.1171,
624
+ "step": 1800
625
+ },
626
+ {
627
+ "epoch": 10.37037037037037,
628
+ "grad_norm": 0.9490203857421875,
629
+ "learning_rate": 1.0707070707070708e-05,
630
+ "loss": 0.1071,
631
+ "step": 1825
632
+ },
633
+ {
634
+ "epoch": 10.512820512820513,
635
+ "grad_norm": 0.40106019377708435,
636
+ "learning_rate": 1.0549242424242425e-05,
637
+ "loss": 0.1375,
638
+ "step": 1850
639
+ },
640
+ {
641
+ "epoch": 10.655270655270655,
642
+ "grad_norm": 0.6680411696434021,
643
+ "learning_rate": 1.0391414141414143e-05,
644
+ "loss": 0.1711,
645
+ "step": 1875
646
+ },
647
+ {
648
+ "epoch": 10.797720797720798,
649
+ "grad_norm": 79.46979522705078,
650
+ "learning_rate": 1.0233585858585859e-05,
651
+ "loss": 0.2827,
652
+ "step": 1900
653
+ },
654
+ {
655
+ "epoch": 10.94017094017094,
656
+ "grad_norm": 101.4886474609375,
657
+ "learning_rate": 1.0075757575757576e-05,
658
+ "loss": 0.128,
659
+ "step": 1925
660
+ },
661
+ {
662
+ "epoch": 11.0,
663
+ "eval_accuracy": 0.5761973875181422,
664
+ "eval_loss": 3.298156261444092,
665
+ "eval_macro_f1": 0.5749744159808577,
666
+ "eval_runtime": 1.4412,
667
+ "eval_samples_per_second": 478.084,
668
+ "eval_steps_per_second": 30.531,
669
+ "eval_weighted_f1": 0.575826531223518,
670
+ "step": 1936
671
+ }
672
+ ],
673
+ "logging_steps": 25,
674
+ "max_steps": 3520,
675
+ "num_input_tokens_seen": 0,
676
+ "num_train_epochs": 20,
677
+ "save_steps": 500,
678
+ "stateful_callbacks": {
679
+ "EarlyStoppingCallback": {
680
+ "args": {
681
+ "early_stopping_patience": 3,
682
+ "early_stopping_threshold": 0.0
683
+ },
684
+ "attributes": {
685
+ "early_stopping_patience_counter": 3
686
+ }
687
+ },
688
+ "TrainerControl": {
689
+ "args": {
690
+ "should_epoch_stop": false,
691
+ "should_evaluate": false,
692
+ "should_log": false,
693
+ "should_save": true,
694
+ "should_training_stop": true
695
+ },
696
+ "attributes": {}
697
+ }
698
+ },
699
+ "total_flos": 0.0,
700
+ "train_batch_size": 8,
701
+ "trial_name": null,
702
+ "trial_params": null
703
+ }
ood-vifactcheck-visobert-source-source-05-s44-20260831T033516Z-59e46264/trainer/checkpoint-1080/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-05-s44-20260831T033516Z-59e46264/trainer/checkpoint-1080/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-05-s44-20260831T033516Z-59e46264/trainer/checkpoint-1080/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-06-s43-20260831T021725Z-1a169064/trainer/checkpoint-2590/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-06-s43-20260831T021725Z-1a169064/trainer/checkpoint-2590/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-06-s43-20260831T021725Z-1a169064/trainer/checkpoint-2590/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-06-s43-20260831T021725Z-1a169064/trainer/checkpoint-2590/trainer_state.json ADDED
@@ -0,0 +1,918 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2035,
3
+ "best_metric": 0.6008961707884494,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-06-s43-20260831T021725Z-1a169064/trainer/checkpoint-2035",
5
+ "epoch": 14.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2590,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13550135501355012,
14
+ "grad_norm": 23.644420623779297,
15
+ "learning_rate": 1.2972972972972974e-06,
16
+ "loss": 4.8413,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27100271002710025,
21
+ "grad_norm": 25.800350189208984,
22
+ "learning_rate": 2.648648648648649e-06,
23
+ "loss": 4.4473,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.4065040650406504,
28
+ "grad_norm": 19.654708862304688,
29
+ "learning_rate": 4.000000000000001e-06,
30
+ "loss": 4.4754,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5420054200542005,
35
+ "grad_norm": 27.55068588256836,
36
+ "learning_rate": 5.351351351351351e-06,
37
+ "loss": 4.3814,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6775067750677507,
42
+ "grad_norm": 18.95235824584961,
43
+ "learning_rate": 6.702702702702704e-06,
44
+ "loss": 4.4142,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8130081300813008,
49
+ "grad_norm": 32.0383415222168,
50
+ "learning_rate": 8.054054054054055e-06,
51
+ "loss": 4.3835,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.948509485094851,
56
+ "grad_norm": 27.09031867980957,
57
+ "learning_rate": 9.405405405405407e-06,
58
+ "loss": 4.3894,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.4051490514905149,
64
+ "eval_loss": 1.0821291208267212,
65
+ "eval_macro_f1": 0.3925480841139686,
66
+ "eval_runtime": 1.4265,
67
+ "eval_samples_per_second": 517.364,
68
+ "eval_steps_per_second": 32.949,
69
+ "eval_weighted_f1": 0.38991704668114896,
70
+ "step": 185
71
+ },
72
+ {
73
+ "epoch": 1.08130081300813,
74
+ "grad_norm": 37.34614944458008,
75
+ "learning_rate": 1.0756756756756757e-05,
76
+ "loss": 4.1795,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2168021680216803,
81
+ "grad_norm": 22.745908737182617,
82
+ "learning_rate": 1.210810810810811e-05,
83
+ "loss": 4.2444,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3523035230352303,
88
+ "grad_norm": 20.919368743896484,
89
+ "learning_rate": 1.3459459459459461e-05,
90
+ "loss": 4.2387,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.4878048780487805,
95
+ "grad_norm": 52.95161437988281,
96
+ "learning_rate": 1.4810810810810811e-05,
97
+ "loss": 4.2627,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6233062330623307,
102
+ "grad_norm": 27.904052734375,
103
+ "learning_rate": 1.6162162162162163e-05,
104
+ "loss": 4.0913,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.758807588075881,
109
+ "grad_norm": 29.80078125,
110
+ "learning_rate": 1.7513513513513517e-05,
111
+ "loss": 4.1792,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.8943089430894309,
116
+ "grad_norm": 22.177852630615234,
117
+ "learning_rate": 1.8864864864864866e-05,
118
+ "loss": 4.2457,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.4959349593495935,
124
+ "eval_loss": 1.0042661428451538,
125
+ "eval_macro_f1": 0.49460544237685083,
126
+ "eval_runtime": 1.7122,
127
+ "eval_samples_per_second": 431.03,
128
+ "eval_steps_per_second": 27.45,
129
+ "eval_weighted_f1": 0.4934151033840644,
130
+ "step": 370
131
+ },
132
+ {
133
+ "epoch": 2.02710027100271,
134
+ "grad_norm": 20.84690284729004,
135
+ "learning_rate": 1.9975975975975978e-05,
136
+ "loss": 3.9331,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.16260162601626,
141
+ "grad_norm": 36.183631896972656,
142
+ "learning_rate": 1.9825825825825828e-05,
143
+ "loss": 3.7031,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.2981029810298104,
148
+ "grad_norm": 42.469383239746094,
149
+ "learning_rate": 1.967567567567568e-05,
150
+ "loss": 3.5926,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.4336043360433606,
155
+ "grad_norm": 58.60617446899414,
156
+ "learning_rate": 1.9525525525525525e-05,
157
+ "loss": 3.6548,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.569105691056911,
162
+ "grad_norm": 35.66259765625,
163
+ "learning_rate": 1.9375375375375376e-05,
164
+ "loss": 3.7605,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.7046070460704605,
169
+ "grad_norm": 34.129878997802734,
170
+ "learning_rate": 1.9225225225225226e-05,
171
+ "loss": 3.6567,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.8401084010840107,
176
+ "grad_norm": 28.44754409790039,
177
+ "learning_rate": 1.9075075075075076e-05,
178
+ "loss": 3.5784,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 2.975609756097561,
183
+ "grad_norm": 21.952425003051758,
184
+ "learning_rate": 1.8924924924924926e-05,
185
+ "loss": 3.6628,
186
+ "step": 550
187
+ },
188
+ {
189
+ "epoch": 3.0,
190
+ "eval_accuracy": 0.5271002710027101,
191
+ "eval_loss": 0.9627034068107605,
192
+ "eval_macro_f1": 0.5204525414299935,
193
+ "eval_runtime": 1.477,
194
+ "eval_samples_per_second": 499.675,
195
+ "eval_steps_per_second": 31.822,
196
+ "eval_weighted_f1": 0.5175579336276049,
197
+ "step": 555
198
+ },
199
+ {
200
+ "epoch": 3.1084010840108403,
201
+ "grad_norm": 28.471473693847656,
202
+ "learning_rate": 1.8774774774774776e-05,
203
+ "loss": 2.8768,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.2439024390243905,
208
+ "grad_norm": 29.604429244995117,
209
+ "learning_rate": 1.8624624624624627e-05,
210
+ "loss": 2.8584,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.3794037940379402,
215
+ "grad_norm": 43.4071044921875,
216
+ "learning_rate": 1.8474474474474477e-05,
217
+ "loss": 2.5758,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.5149051490514904,
222
+ "grad_norm": 47.71430587768555,
223
+ "learning_rate": 1.8324324324324324e-05,
224
+ "loss": 2.8767,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.6504065040650406,
229
+ "grad_norm": 37.3945198059082,
230
+ "learning_rate": 1.8174174174174174e-05,
231
+ "loss": 2.6165,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.785907859078591,
236
+ "grad_norm": 27.95343589782715,
237
+ "learning_rate": 1.8024024024024024e-05,
238
+ "loss": 2.7598,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.921409214092141,
243
+ "grad_norm": 41.22407913208008,
244
+ "learning_rate": 1.7873873873873874e-05,
245
+ "loss": 2.8942,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 4.0,
250
+ "eval_accuracy": 0.5772357723577236,
251
+ "eval_loss": 1.0029407739639282,
252
+ "eval_macro_f1": 0.5682174009734623,
253
+ "eval_runtime": 1.4061,
254
+ "eval_samples_per_second": 524.86,
255
+ "eval_steps_per_second": 33.426,
256
+ "eval_weighted_f1": 0.5694122373702232,
257
+ "step": 740
258
+ },
259
+ {
260
+ "epoch": 4.05420054200542,
261
+ "grad_norm": 26.70307731628418,
262
+ "learning_rate": 1.7723723723723724e-05,
263
+ "loss": 2.57,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.1897018970189706,
268
+ "grad_norm": 41.035099029541016,
269
+ "learning_rate": 1.7573573573573575e-05,
270
+ "loss": 1.7102,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.32520325203252,
275
+ "grad_norm": 47.82585525512695,
276
+ "learning_rate": 1.7423423423423425e-05,
277
+ "loss": 1.869,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.46070460704607,
282
+ "grad_norm": 53.93787384033203,
283
+ "learning_rate": 1.7273273273273275e-05,
284
+ "loss": 1.8035,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.596205962059621,
289
+ "grad_norm": 27.55320930480957,
290
+ "learning_rate": 1.7123123123123125e-05,
291
+ "loss": 1.7709,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.7317073170731705,
296
+ "grad_norm": 65.54625701904297,
297
+ "learning_rate": 1.6972972972972975e-05,
298
+ "loss": 1.7673,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.867208672086721,
303
+ "grad_norm": 66.76729583740234,
304
+ "learning_rate": 1.6822822822822826e-05,
305
+ "loss": 1.7912,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 5.0,
310
+ "grad_norm": 109.89456939697266,
311
+ "learning_rate": 1.6672672672672673e-05,
312
+ "loss": 1.9093,
313
+ "step": 925
314
+ },
315
+ {
316
+ "epoch": 5.0,
317
+ "eval_accuracy": 0.5636856368563685,
318
+ "eval_loss": 1.1756083965301514,
319
+ "eval_macro_f1": 0.5673510528846676,
320
+ "eval_runtime": 1.3549,
321
+ "eval_samples_per_second": 544.689,
322
+ "eval_steps_per_second": 34.689,
323
+ "eval_weighted_f1": 0.5668583332524822,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.13550135501355,
328
+ "grad_norm": 66.03388214111328,
329
+ "learning_rate": 1.6522522522522523e-05,
330
+ "loss": 0.992,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.2710027100271,
335
+ "grad_norm": 67.87275695800781,
336
+ "learning_rate": 1.6372372372372373e-05,
337
+ "loss": 0.951,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.40650406504065,
342
+ "grad_norm": 116.9877700805664,
343
+ "learning_rate": 1.6222222222222223e-05,
344
+ "loss": 1.1113,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.542005420054201,
349
+ "grad_norm": 69.57601165771484,
350
+ "learning_rate": 1.6072072072072073e-05,
351
+ "loss": 1.305,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.677506775067751,
356
+ "grad_norm": 57.67194747924805,
357
+ "learning_rate": 1.5921921921921924e-05,
358
+ "loss": 1.5146,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.8130081300813,
363
+ "grad_norm": 26.43688201904297,
364
+ "learning_rate": 1.5771771771771774e-05,
365
+ "loss": 1.0235,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 5.948509485094851,
370
+ "grad_norm": 42.5482063293457,
371
+ "learning_rate": 1.5621621621621624e-05,
372
+ "loss": 1.0116,
373
+ "step": 1100
374
+ },
375
+ {
376
+ "epoch": 6.0,
377
+ "eval_accuracy": 0.55420054200542,
378
+ "eval_loss": 1.4322667121887207,
379
+ "eval_macro_f1": 0.5567958398020552,
380
+ "eval_runtime": 1.4799,
381
+ "eval_samples_per_second": 498.668,
382
+ "eval_steps_per_second": 31.758,
383
+ "eval_weighted_f1": 0.5555285916479377,
384
+ "step": 1110
385
+ },
386
+ {
387
+ "epoch": 6.08130081300813,
388
+ "grad_norm": 24.943260192871094,
389
+ "learning_rate": 1.5471471471471474e-05,
390
+ "loss": 0.6804,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.2168021680216805,
395
+ "grad_norm": 57.205467224121094,
396
+ "learning_rate": 1.5321321321321324e-05,
397
+ "loss": 0.6917,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.35230352303523,
402
+ "grad_norm": 60.98979568481445,
403
+ "learning_rate": 1.5171171171171173e-05,
404
+ "loss": 0.6462,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.487804878048781,
409
+ "grad_norm": 42.11968994140625,
410
+ "learning_rate": 1.5021021021021023e-05,
411
+ "loss": 0.729,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.623306233062331,
416
+ "grad_norm": 31.105592727661133,
417
+ "learning_rate": 1.4870870870870873e-05,
418
+ "loss": 0.5225,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.7588075880758804,
423
+ "grad_norm": 46.184783935546875,
424
+ "learning_rate": 1.4720720720720722e-05,
425
+ "loss": 0.6345,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 6.894308943089431,
430
+ "grad_norm": 104.41244506835938,
431
+ "learning_rate": 1.4570570570570572e-05,
432
+ "loss": 0.5265,
433
+ "step": 1275
434
+ },
435
+ {
436
+ "epoch": 7.0,
437
+ "eval_accuracy": 0.5650406504065041,
438
+ "eval_loss": 1.9319794178009033,
439
+ "eval_macro_f1": 0.5684701440187893,
440
+ "eval_runtime": 1.525,
441
+ "eval_samples_per_second": 483.942,
442
+ "eval_steps_per_second": 30.82,
443
+ "eval_weighted_f1": 0.5683702527633752,
444
+ "step": 1295
445
+ },
446
+ {
447
+ "epoch": 7.02710027100271,
448
+ "grad_norm": 3.5306365489959717,
449
+ "learning_rate": 1.4420420420420422e-05,
450
+ "loss": 0.58,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.16260162601626,
455
+ "grad_norm": 28.436826705932617,
456
+ "learning_rate": 1.4270270270270272e-05,
457
+ "loss": 0.3672,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.29810298102981,
462
+ "grad_norm": 41.26507568359375,
463
+ "learning_rate": 1.4120120120120123e-05,
464
+ "loss": 0.3083,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.43360433604336,
469
+ "grad_norm": 43.98733901977539,
470
+ "learning_rate": 1.3969969969969971e-05,
471
+ "loss": 0.4188,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.569105691056911,
476
+ "grad_norm": 80.8287124633789,
477
+ "learning_rate": 1.3819819819819821e-05,
478
+ "loss": 0.3759,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 7.7046070460704605,
483
+ "grad_norm": 36.36510467529297,
484
+ "learning_rate": 1.3669669669669672e-05,
485
+ "loss": 0.4157,
486
+ "step": 1425
487
+ },
488
+ {
489
+ "epoch": 7.840108401084011,
490
+ "grad_norm": 55.13428497314453,
491
+ "learning_rate": 1.3519519519519522e-05,
492
+ "loss": 0.2661,
493
+ "step": 1450
494
+ },
495
+ {
496
+ "epoch": 7.975609756097561,
497
+ "grad_norm": 49.47197723388672,
498
+ "learning_rate": 1.3369369369369369e-05,
499
+ "loss": 0.4562,
500
+ "step": 1475
501
+ },
502
+ {
503
+ "epoch": 8.0,
504
+ "eval_accuracy": 0.5691056910569106,
505
+ "eval_loss": 2.113414764404297,
506
+ "eval_macro_f1": 0.5698669130203521,
507
+ "eval_runtime": 1.5495,
508
+ "eval_samples_per_second": 476.273,
509
+ "eval_steps_per_second": 30.332,
510
+ "eval_weighted_f1": 0.5693848500221795,
511
+ "step": 1480
512
+ },
513
+ {
514
+ "epoch": 8.10840108401084,
515
+ "grad_norm": 76.07450103759766,
516
+ "learning_rate": 1.3219219219219219e-05,
517
+ "loss": 0.3138,
518
+ "step": 1500
519
+ },
520
+ {
521
+ "epoch": 8.24390243902439,
522
+ "grad_norm": 50.969268798828125,
523
+ "learning_rate": 1.3069069069069069e-05,
524
+ "loss": 0.2577,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 8.379403794037941,
529
+ "grad_norm": 92.61463165283203,
530
+ "learning_rate": 1.291891891891892e-05,
531
+ "loss": 0.1471,
532
+ "step": 1550
533
+ },
534
+ {
535
+ "epoch": 8.51490514905149,
536
+ "grad_norm": 38.85557556152344,
537
+ "learning_rate": 1.276876876876877e-05,
538
+ "loss": 0.3627,
539
+ "step": 1575
540
+ },
541
+ {
542
+ "epoch": 8.65040650406504,
543
+ "grad_norm": 2.0656871795654297,
544
+ "learning_rate": 1.2618618618618618e-05,
545
+ "loss": 0.4193,
546
+ "step": 1600
547
+ },
548
+ {
549
+ "epoch": 8.785907859078591,
550
+ "grad_norm": 23.408620834350586,
551
+ "learning_rate": 1.2468468468468468e-05,
552
+ "loss": 0.2464,
553
+ "step": 1625
554
+ },
555
+ {
556
+ "epoch": 8.92140921409214,
557
+ "grad_norm": 141.25840759277344,
558
+ "learning_rate": 1.2318318318318318e-05,
559
+ "loss": 0.4415,
560
+ "step": 1650
561
+ },
562
+ {
563
+ "epoch": 9.0,
564
+ "eval_accuracy": 0.5934959349593496,
565
+ "eval_loss": 2.5610709190368652,
566
+ "eval_macro_f1": 0.5907776037613098,
567
+ "eval_runtime": 1.6025,
568
+ "eval_samples_per_second": 460.52,
569
+ "eval_steps_per_second": 29.328,
570
+ "eval_weighted_f1": 0.5913238827137199,
571
+ "step": 1665
572
+ },
573
+ {
574
+ "epoch": 9.05420054200542,
575
+ "grad_norm": 0.8425450921058655,
576
+ "learning_rate": 1.2168168168168169e-05,
577
+ "loss": 0.2424,
578
+ "step": 1675
579
+ },
580
+ {
581
+ "epoch": 9.18970189701897,
582
+ "grad_norm": 5.634618759155273,
583
+ "learning_rate": 1.2018018018018019e-05,
584
+ "loss": 0.3127,
585
+ "step": 1700
586
+ },
587
+ {
588
+ "epoch": 9.32520325203252,
589
+ "grad_norm": 16.855060577392578,
590
+ "learning_rate": 1.1867867867867869e-05,
591
+ "loss": 0.1878,
592
+ "step": 1725
593
+ },
594
+ {
595
+ "epoch": 9.46070460704607,
596
+ "grad_norm": 91.97650146484375,
597
+ "learning_rate": 1.1717717717717718e-05,
598
+ "loss": 0.2588,
599
+ "step": 1750
600
+ },
601
+ {
602
+ "epoch": 9.59620596205962,
603
+ "grad_norm": 30.13783073425293,
604
+ "learning_rate": 1.1567567567567568e-05,
605
+ "loss": 0.2151,
606
+ "step": 1775
607
+ },
608
+ {
609
+ "epoch": 9.731707317073171,
610
+ "grad_norm": 40.8541259765625,
611
+ "learning_rate": 1.1417417417417418e-05,
612
+ "loss": 0.2247,
613
+ "step": 1800
614
+ },
615
+ {
616
+ "epoch": 9.86720867208672,
617
+ "grad_norm": 0.22134290635585785,
618
+ "learning_rate": 1.1267267267267268e-05,
619
+ "loss": 0.1759,
620
+ "step": 1825
621
+ },
622
+ {
623
+ "epoch": 10.0,
624
+ "grad_norm": 39.33653259277344,
625
+ "learning_rate": 1.1117117117117118e-05,
626
+ "loss": 0.1725,
627
+ "step": 1850
628
+ },
629
+ {
630
+ "epoch": 10.0,
631
+ "eval_accuracy": 0.5745257452574526,
632
+ "eval_loss": 3.2890443801879883,
633
+ "eval_macro_f1": 0.5663151864992984,
634
+ "eval_runtime": 1.4142,
635
+ "eval_samples_per_second": 521.855,
636
+ "eval_steps_per_second": 33.235,
637
+ "eval_weighted_f1": 0.5672364398217392,
638
+ "step": 1850
639
+ },
640
+ {
641
+ "epoch": 10.13550135501355,
642
+ "grad_norm": 0.9082435965538025,
643
+ "learning_rate": 1.0966966966966967e-05,
644
+ "loss": 0.1883,
645
+ "step": 1875
646
+ },
647
+ {
648
+ "epoch": 10.2710027100271,
649
+ "grad_norm": 4.879505157470703,
650
+ "learning_rate": 1.0816816816816817e-05,
651
+ "loss": 0.1127,
652
+ "step": 1900
653
+ },
654
+ {
655
+ "epoch": 10.40650406504065,
656
+ "grad_norm": 0.7202223539352417,
657
+ "learning_rate": 1.0666666666666667e-05,
658
+ "loss": 0.2144,
659
+ "step": 1925
660
+ },
661
+ {
662
+ "epoch": 10.5420054200542,
663
+ "grad_norm": 10.110024452209473,
664
+ "learning_rate": 1.0516516516516518e-05,
665
+ "loss": 0.2009,
666
+ "step": 1950
667
+ },
668
+ {
669
+ "epoch": 10.677506775067751,
670
+ "grad_norm": 142.56756591796875,
671
+ "learning_rate": 1.0366366366366368e-05,
672
+ "loss": 0.0916,
673
+ "step": 1975
674
+ },
675
+ {
676
+ "epoch": 10.8130081300813,
677
+ "grad_norm": 18.083961486816406,
678
+ "learning_rate": 1.0216216216216216e-05,
679
+ "loss": 0.0476,
680
+ "step": 2000
681
+ },
682
+ {
683
+ "epoch": 10.948509485094851,
684
+ "grad_norm": 92.09305572509766,
685
+ "learning_rate": 1.0066066066066066e-05,
686
+ "loss": 0.0713,
687
+ "step": 2025
688
+ },
689
+ {
690
+ "epoch": 11.0,
691
+ "eval_accuracy": 0.6002710027100271,
692
+ "eval_loss": 3.4880893230438232,
693
+ "eval_macro_f1": 0.6008961707884494,
694
+ "eval_runtime": 1.4136,
695
+ "eval_samples_per_second": 522.079,
696
+ "eval_steps_per_second": 33.249,
697
+ "eval_weighted_f1": 0.6011004064149246,
698
+ "step": 2035
699
+ },
700
+ {
701
+ "epoch": 11.08130081300813,
702
+ "grad_norm": 0.09992863982915878,
703
+ "learning_rate": 9.915915915915917e-06,
704
+ "loss": 0.1147,
705
+ "step": 2050
706
+ },
707
+ {
708
+ "epoch": 11.21680216802168,
709
+ "grad_norm": 2.103893280029297,
710
+ "learning_rate": 9.765765765765767e-06,
711
+ "loss": 0.0703,
712
+ "step": 2075
713
+ },
714
+ {
715
+ "epoch": 11.35230352303523,
716
+ "grad_norm": 65.16277313232422,
717
+ "learning_rate": 9.615615615615617e-06,
718
+ "loss": 0.0652,
719
+ "step": 2100
720
+ },
721
+ {
722
+ "epoch": 11.487804878048781,
723
+ "grad_norm": 0.2217359095811844,
724
+ "learning_rate": 9.465465465465467e-06,
725
+ "loss": 0.0783,
726
+ "step": 2125
727
+ },
728
+ {
729
+ "epoch": 11.62330623306233,
730
+ "grad_norm": 0.5729576349258423,
731
+ "learning_rate": 9.315315315315316e-06,
732
+ "loss": 0.1287,
733
+ "step": 2150
734
+ },
735
+ {
736
+ "epoch": 11.75880758807588,
737
+ "grad_norm": 0.2788489758968353,
738
+ "learning_rate": 9.165165165165166e-06,
739
+ "loss": 0.0184,
740
+ "step": 2175
741
+ },
742
+ {
743
+ "epoch": 11.894308943089431,
744
+ "grad_norm": 0.0746355876326561,
745
+ "learning_rate": 9.015015015015016e-06,
746
+ "loss": 0.0843,
747
+ "step": 2200
748
+ },
749
+ {
750
+ "epoch": 12.0,
751
+ "eval_accuracy": 0.5853658536585366,
752
+ "eval_loss": 4.011200904846191,
753
+ "eval_macro_f1": 0.5801614469217244,
754
+ "eval_runtime": 1.443,
755
+ "eval_samples_per_second": 511.435,
756
+ "eval_steps_per_second": 32.571,
757
+ "eval_weighted_f1": 0.5804715792495382,
758
+ "step": 2220
759
+ },
760
+ {
761
+ "epoch": 12.02710027100271,
762
+ "grad_norm": 0.008277347311377525,
763
+ "learning_rate": 8.864864864864866e-06,
764
+ "loss": 0.1007,
765
+ "step": 2225
766
+ },
767
+ {
768
+ "epoch": 12.16260162601626,
769
+ "grad_norm": 5.9232282638549805,
770
+ "learning_rate": 8.714714714714717e-06,
771
+ "loss": 0.064,
772
+ "step": 2250
773
+ },
774
+ {
775
+ "epoch": 12.29810298102981,
776
+ "grad_norm": 0.029273850843310356,
777
+ "learning_rate": 8.564564564564565e-06,
778
+ "loss": 0.0043,
779
+ "step": 2275
780
+ },
781
+ {
782
+ "epoch": 12.433604336043361,
783
+ "grad_norm": 1.6714510917663574,
784
+ "learning_rate": 8.414414414414415e-06,
785
+ "loss": 0.034,
786
+ "step": 2300
787
+ },
788
+ {
789
+ "epoch": 12.56910569105691,
790
+ "grad_norm": 1.2065329551696777,
791
+ "learning_rate": 8.264264264264264e-06,
792
+ "loss": 0.0222,
793
+ "step": 2325
794
+ },
795
+ {
796
+ "epoch": 12.70460704607046,
797
+ "grad_norm": 5.032736778259277,
798
+ "learning_rate": 8.114114114114114e-06,
799
+ "loss": 0.0161,
800
+ "step": 2350
801
+ },
802
+ {
803
+ "epoch": 12.840108401084011,
804
+ "grad_norm": 0.6025921106338501,
805
+ "learning_rate": 7.963963963963964e-06,
806
+ "loss": 0.0901,
807
+ "step": 2375
808
+ },
809
+ {
810
+ "epoch": 12.975609756097562,
811
+ "grad_norm": 133.40142822265625,
812
+ "learning_rate": 7.813813813813815e-06,
813
+ "loss": 0.0559,
814
+ "step": 2400
815
+ },
816
+ {
817
+ "epoch": 13.0,
818
+ "eval_accuracy": 0.5799457994579946,
819
+ "eval_loss": 4.267383575439453,
820
+ "eval_macro_f1": 0.5780554156433554,
821
+ "eval_runtime": 1.429,
822
+ "eval_samples_per_second": 516.438,
823
+ "eval_steps_per_second": 32.89,
824
+ "eval_weighted_f1": 0.5782149929014333,
825
+ "step": 2405
826
+ },
827
+ {
828
+ "epoch": 13.10840108401084,
829
+ "grad_norm": 0.09415595978498459,
830
+ "learning_rate": 7.663663663663665e-06,
831
+ "loss": 0.1452,
832
+ "step": 2425
833
+ },
834
+ {
835
+ "epoch": 13.24390243902439,
836
+ "grad_norm": 83.01631164550781,
837
+ "learning_rate": 7.513513513513514e-06,
838
+ "loss": 0.0419,
839
+ "step": 2450
840
+ },
841
+ {
842
+ "epoch": 13.379403794037941,
843
+ "grad_norm": 0.024820571765303612,
844
+ "learning_rate": 7.363363363363364e-06,
845
+ "loss": 0.0786,
846
+ "step": 2475
847
+ },
848
+ {
849
+ "epoch": 13.51490514905149,
850
+ "grad_norm": 0.11721368134021759,
851
+ "learning_rate": 7.213213213213214e-06,
852
+ "loss": 0.1145,
853
+ "step": 2500
854
+ },
855
+ {
856
+ "epoch": 13.65040650406504,
857
+ "grad_norm": 19.64776039123535,
858
+ "learning_rate": 7.063063063063064e-06,
859
+ "loss": 0.015,
860
+ "step": 2525
861
+ },
862
+ {
863
+ "epoch": 13.785907859078591,
864
+ "grad_norm": 0.01341403741389513,
865
+ "learning_rate": 6.912912912912913e-06,
866
+ "loss": 0.0553,
867
+ "step": 2550
868
+ },
869
+ {
870
+ "epoch": 13.92140921409214,
871
+ "grad_norm": 0.015896497294306755,
872
+ "learning_rate": 6.7627627627627634e-06,
873
+ "loss": 0.1157,
874
+ "step": 2575
875
+ },
876
+ {
877
+ "epoch": 14.0,
878
+ "eval_accuracy": 0.5799457994579946,
879
+ "eval_loss": 4.190284252166748,
880
+ "eval_macro_f1": 0.5819918629118254,
881
+ "eval_runtime": 1.4449,
882
+ "eval_samples_per_second": 510.775,
883
+ "eval_steps_per_second": 32.529,
884
+ "eval_weighted_f1": 0.5818788609783218,
885
+ "step": 2590
886
+ }
887
+ ],
888
+ "logging_steps": 25,
889
+ "max_steps": 3700,
890
+ "num_input_tokens_seen": 0,
891
+ "num_train_epochs": 20,
892
+ "save_steps": 500,
893
+ "stateful_callbacks": {
894
+ "EarlyStoppingCallback": {
895
+ "args": {
896
+ "early_stopping_patience": 3,
897
+ "early_stopping_threshold": 0.0
898
+ },
899
+ "attributes": {
900
+ "early_stopping_patience_counter": 3
901
+ }
902
+ },
903
+ "TrainerControl": {
904
+ "args": {
905
+ "should_epoch_stop": false,
906
+ "should_evaluate": false,
907
+ "should_log": false,
908
+ "should_save": true,
909
+ "should_training_stop": true
910
+ },
911
+ "attributes": {}
912
+ }
913
+ },
914
+ "total_flos": 0.0,
915
+ "train_batch_size": 8,
916
+ "trial_name": null,
917
+ "trial_params": null
918
+ }
ood-vifactcheck-visobert-source-source-06-s44-20260831T034444Z-59e46264/trainer/checkpoint-1850/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-06-s44-20260831T034444Z-59e46264/trainer/checkpoint-1850/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-06-s44-20260831T034444Z-59e46264/trainer/checkpoint-1850/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }