PhaaNe commited on
Commit
904a87b
·
verified ·
1 Parent(s): afaf89c

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-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-1638/special_tokens_map.json +15 -0
  2. ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-1638/tokenizer.json +0 -0
  3. ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-1638/tokenizer_config.json +56 -0
  4. ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-1638/trainer_state.json +597 -0
  5. ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-2184/special_tokens_map.json +15 -0
  6. ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-2184/tokenizer.json +0 -0
  7. ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-2184/tokenizer_config.json +56 -0
  8. ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-2184/trainer_state.json +784 -0
  9. ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1026/special_tokens_map.json +15 -0
  10. ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1026/tokenizer.json +0 -0
  11. ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1026/tokenizer_config.json +56 -0
  12. ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1026/trainer_state.json +396 -0
  13. ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1539/special_tokens_map.json +15 -0
  14. ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1539/tokenizer.json +0 -0
  15. ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1539/tokenizer_config.json +56 -0
  16. ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1539/trainer_state.json +569 -0
  17. ood-vifactcheck-visobert-source-source-02-s42-20260830T234935Z-f21e2f7e/trainer/checkpoint-1267/special_tokens_map.json +15 -0
  18. ood-vifactcheck-visobert-source-source-02-s42-20260830T234935Z-f21e2f7e/trainer/checkpoint-1267/tokenizer.json +0 -0
  19. ood-vifactcheck-visobert-source-source-02-s42-20260830T234935Z-f21e2f7e/trainer/checkpoint-1267/tokenizer_config.json +56 -0
  20. ood-vifactcheck-visobert-source-source-02-s42-20260830T234935Z-f21e2f7e/trainer/checkpoint-1267/trainer_state.json +470 -0
  21. ood-vifactcheck-visobert-source-source-02-s42-20260830T234935Z-f21e2f7e/trainer/checkpoint-724/special_tokens_map.json +15 -0
  22. ood-vifactcheck-visobert-source-source-02-s42-20260830T234935Z-f21e2f7e/trainer/checkpoint-724/tokenizer.json +0 -0
  23. ood-vifactcheck-visobert-source-source-02-s42-20260830T234935Z-f21e2f7e/trainer/checkpoint-724/tokenizer_config.json +56 -0
  24. ood-vifactcheck-visobert-source-source-02-s42-20260830T234935Z-f21e2f7e/trainer/checkpoint-724/trainer_state.json +283 -0
  25. ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2013/special_tokens_map.json +15 -0
  26. ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2013/tokenizer.json +0 -0
  27. ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2013/tokenizer_config.json +56 -0
  28. ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2013/trainer_state.json +724 -0
  29. ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2562/special_tokens_map.json +15 -0
  30. ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2562/tokenizer.json +0 -0
  31. ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2562/tokenizer_config.json +56 -0
  32. ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2562/trainer_state.json +911 -0
  33. ood-vifactcheck-visobert-source-source-04-s42-20260831T001152Z-f21e2f7e/trainer/checkpoint-1584/special_tokens_map.json +15 -0
  34. ood-vifactcheck-visobert-source-source-04-s42-20260831T001152Z-f21e2f7e/trainer/checkpoint-1584/tokenizer.json +0 -0
  35. ood-vifactcheck-visobert-source-source-04-s42-20260831T001152Z-f21e2f7e/trainer/checkpoint-1584/tokenizer_config.json +56 -0
  36. ood-vifactcheck-visobert-source-source-04-s42-20260831T001152Z-f21e2f7e/trainer/checkpoint-1584/trainer_state.json +583 -0
  37. ood-vifactcheck-visobert-source-source-04-s42-20260831T001152Z-f21e2f7e/trainer/checkpoint-2112/tokenizer_config.json +56 -0
  38. ood-vifactcheck-visobert-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-1504/special_tokens_map.json +15 -0
  39. ood-vifactcheck-visobert-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-1504/tokenizer.json +0 -0
  40. ood-vifactcheck-visobert-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-1504/trainer_state.json +551 -0
  41. ood-vifactcheck-visobert-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-2068/special_tokens_map.json +15 -0
  42. ood-vifactcheck-visobert-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-2068/tokenizer.json +0 -0
  43. ood-vifactcheck-visobert-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-2068/tokenizer_config.json +56 -0
  44. ood-vifactcheck-visobert-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-2068/trainer_state.json +738 -0
  45. ood-vifactcheck-visobert-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1050/special_tokens_map.json +15 -0
  46. ood-vifactcheck-visobert-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1050/tokenizer.json +0 -0
  47. ood-vifactcheck-visobert-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1050/tokenizer_config.json +56 -0
  48. ood-vifactcheck-visobert-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1050/trainer_state.json +403 -0
  49. ood-vifactcheck-visobert-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1575/special_tokens_map.json +15 -0
  50. ood-vifactcheck-visobert-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1575/tokenizer.json +0 -0
ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-1638/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-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-1638/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-1638/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-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-1638/trainer_state.json ADDED
@@ -0,0 +1,597 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1638,
3
+ "best_metric": 0.5726049605469152,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-1638",
5
+ "epoch": 9.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1638,
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": 37.184391021728516,
15
+ "learning_rate": 1.3186813186813187e-06,
16
+ "loss": 4.7284,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27586206896551724,
21
+ "grad_norm": 31.35881805419922,
22
+ "learning_rate": 2.6923076923076923e-06,
23
+ "loss": 4.5436,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.41379310344827586,
28
+ "grad_norm": 40.47816467285156,
29
+ "learning_rate": 4.065934065934066e-06,
30
+ "loss": 4.5382,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5517241379310345,
35
+ "grad_norm": 28.50191307067871,
36
+ "learning_rate": 5.43956043956044e-06,
37
+ "loss": 4.4636,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6896551724137931,
42
+ "grad_norm": 29.587844848632812,
43
+ "learning_rate": 6.813186813186814e-06,
44
+ "loss": 4.4057,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8275862068965517,
49
+ "grad_norm": 19.32939338684082,
50
+ "learning_rate": 8.186813186813188e-06,
51
+ "loss": 4.3842,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9655172413793104,
56
+ "grad_norm": 30.56195831298828,
57
+ "learning_rate": 9.560439560439562e-06,
58
+ "loss": 4.387,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.40669856459330145,
64
+ "eval_loss": 1.0807603597640991,
65
+ "eval_macro_f1": 0.3250454610810856,
66
+ "eval_runtime": 1.3826,
67
+ "eval_samples_per_second": 453.479,
68
+ "eval_steps_per_second": 28.93,
69
+ "eval_weighted_f1": 0.3338543425140942,
70
+ "step": 182
71
+ },
72
+ {
73
+ "epoch": 1.0993103448275863,
74
+ "grad_norm": 45.888633728027344,
75
+ "learning_rate": 1.0934065934065933e-05,
76
+ "loss": 4.2564,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.237241379310345,
81
+ "grad_norm": 16.613557815551758,
82
+ "learning_rate": 1.230769230769231e-05,
83
+ "loss": 4.2484,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3751724137931034,
88
+ "grad_norm": 28.772672653198242,
89
+ "learning_rate": 1.3681318681318683e-05,
90
+ "loss": 4.24,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.513103448275862,
95
+ "grad_norm": 27.18781852722168,
96
+ "learning_rate": 1.5054945054945056e-05,
97
+ "loss": 4.2858,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6510344827586207,
102
+ "grad_norm": 19.768953323364258,
103
+ "learning_rate": 1.642857142857143e-05,
104
+ "loss": 4.2016,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7889655172413792,
109
+ "grad_norm": 20.3443660736084,
110
+ "learning_rate": 1.78021978021978e-05,
111
+ "loss": 4.2666,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.926896551724138,
116
+ "grad_norm": 20.553537368774414,
117
+ "learning_rate": 1.9175824175824178e-05,
118
+ "loss": 4.2036,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.42902711323763953,
124
+ "eval_loss": 1.118101716041565,
125
+ "eval_macro_f1": 0.36056831206341905,
126
+ "eval_runtime": 1.2323,
127
+ "eval_samples_per_second": 508.803,
128
+ "eval_steps_per_second": 32.46,
129
+ "eval_weighted_f1": 0.3488534356774562,
130
+ "step": 364
131
+ },
132
+ {
133
+ "epoch": 2.0606896551724136,
134
+ "grad_norm": 34.58537292480469,
135
+ "learning_rate": 1.993894993894994e-05,
136
+ "loss": 4.0065,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.1986206896551725,
141
+ "grad_norm": 25.910797119140625,
142
+ "learning_rate": 1.9786324786324788e-05,
143
+ "loss": 3.8394,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.336551724137931,
148
+ "grad_norm": 40.513763427734375,
149
+ "learning_rate": 1.9633699633699634e-05,
150
+ "loss": 3.7494,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.47448275862069,
155
+ "grad_norm": 36.02470397949219,
156
+ "learning_rate": 1.948107448107448e-05,
157
+ "loss": 3.8057,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.6124137931034483,
162
+ "grad_norm": 45.625755310058594,
163
+ "learning_rate": 1.932844932844933e-05,
164
+ "loss": 3.754,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.750344827586207,
169
+ "grad_norm": 32.18267059326172,
170
+ "learning_rate": 1.9175824175824178e-05,
171
+ "loss": 3.7035,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.8882758620689657,
176
+ "grad_norm": 32.876625061035156,
177
+ "learning_rate": 1.9023199023199025e-05,
178
+ "loss": 3.6018,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5007974481658692,
184
+ "eval_loss": 0.984271228313446,
185
+ "eval_macro_f1": 0.49151323654018847,
186
+ "eval_runtime": 1.221,
187
+ "eval_samples_per_second": 513.52,
188
+ "eval_steps_per_second": 32.76,
189
+ "eval_weighted_f1": 0.48633333130427775,
190
+ "step": 546
191
+ },
192
+ {
193
+ "epoch": 3.0220689655172412,
194
+ "grad_norm": 50.10889434814453,
195
+ "learning_rate": 1.887057387057387e-05,
196
+ "loss": 3.3932,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.16,
201
+ "grad_norm": 56.40572738647461,
202
+ "learning_rate": 1.8717948717948718e-05,
203
+ "loss": 2.9721,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.2979310344827586,
208
+ "grad_norm": 36.12623596191406,
209
+ "learning_rate": 1.8565323565323568e-05,
210
+ "loss": 2.5898,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.435862068965517,
215
+ "grad_norm": 28.53684425354004,
216
+ "learning_rate": 1.8412698412698415e-05,
217
+ "loss": 2.7856,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.573793103448276,
222
+ "grad_norm": 29.99932098388672,
223
+ "learning_rate": 1.826007326007326e-05,
224
+ "loss": 2.9761,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.7117241379310344,
229
+ "grad_norm": 38.28874206542969,
230
+ "learning_rate": 1.8107448107448108e-05,
231
+ "loss": 2.9072,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.849655172413793,
236
+ "grad_norm": 76.56749725341797,
237
+ "learning_rate": 1.7954822954822958e-05,
238
+ "loss": 3.0844,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.9875862068965517,
243
+ "grad_norm": 40.25591278076172,
244
+ "learning_rate": 1.78021978021978e-05,
245
+ "loss": 3.0495,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 4.0,
250
+ "eval_accuracy": 0.5311004784688995,
251
+ "eval_loss": 1.0743650197982788,
252
+ "eval_macro_f1": 0.5205468214965933,
253
+ "eval_runtime": 1.208,
254
+ "eval_samples_per_second": 519.056,
255
+ "eval_steps_per_second": 33.114,
256
+ "eval_weighted_f1": 0.5175166903640859,
257
+ "step": 728
258
+ },
259
+ {
260
+ "epoch": 4.121379310344827,
261
+ "grad_norm": 41.000728607177734,
262
+ "learning_rate": 1.764957264957265e-05,
263
+ "loss": 1.8659,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.259310344827586,
268
+ "grad_norm": 29.559438705444336,
269
+ "learning_rate": 1.74969474969475e-05,
270
+ "loss": 1.9239,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.397241379310345,
275
+ "grad_norm": 38.09396743774414,
276
+ "learning_rate": 1.7344322344322345e-05,
277
+ "loss": 1.9678,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.535172413793103,
282
+ "grad_norm": 50.2789192199707,
283
+ "learning_rate": 1.7191697191697195e-05,
284
+ "loss": 2.0875,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.673103448275862,
289
+ "grad_norm": 30.099040985107422,
290
+ "learning_rate": 1.703907203907204e-05,
291
+ "loss": 1.8674,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.811034482758621,
296
+ "grad_norm": 75.6994400024414,
297
+ "learning_rate": 1.688644688644689e-05,
298
+ "loss": 1.9013,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.94896551724138,
303
+ "grad_norm": 40.34619903564453,
304
+ "learning_rate": 1.6733821733821735e-05,
305
+ "loss": 1.9484,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 5.0,
310
+ "eval_accuracy": 0.5390749601275917,
311
+ "eval_loss": 1.274722695350647,
312
+ "eval_macro_f1": 0.5310477318972339,
313
+ "eval_runtime": 1.3175,
314
+ "eval_samples_per_second": 475.899,
315
+ "eval_steps_per_second": 30.36,
316
+ "eval_weighted_f1": 0.5321559967817984,
317
+ "step": 910
318
+ },
319
+ {
320
+ "epoch": 5.082758620689655,
321
+ "grad_norm": 62.8062858581543,
322
+ "learning_rate": 1.6581196581196585e-05,
323
+ "loss": 1.5826,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.220689655172414,
328
+ "grad_norm": 92.9386215209961,
329
+ "learning_rate": 1.642857142857143e-05,
330
+ "loss": 1.2958,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.358620689655172,
335
+ "grad_norm": 40.67005157470703,
336
+ "learning_rate": 1.627594627594628e-05,
337
+ "loss": 1.0679,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.496551724137931,
342
+ "grad_norm": 35.90666198730469,
343
+ "learning_rate": 1.6123321123321125e-05,
344
+ "loss": 1.4217,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.63448275862069,
349
+ "grad_norm": 74.77809143066406,
350
+ "learning_rate": 1.5970695970695972e-05,
351
+ "loss": 1.2528,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.772413793103448,
356
+ "grad_norm": 32.45465850830078,
357
+ "learning_rate": 1.581807081807082e-05,
358
+ "loss": 1.0701,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.910344827586207,
363
+ "grad_norm": 57.50745391845703,
364
+ "learning_rate": 1.5665445665445665e-05,
365
+ "loss": 0.983,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 6.0,
370
+ "eval_accuracy": 0.5502392344497608,
371
+ "eval_loss": 1.5191906690597534,
372
+ "eval_macro_f1": 0.5442008282447773,
373
+ "eval_runtime": 1.3112,
374
+ "eval_samples_per_second": 478.194,
375
+ "eval_steps_per_second": 30.507,
376
+ "eval_weighted_f1": 0.5438193474963617,
377
+ "step": 1092
378
+ },
379
+ {
380
+ "epoch": 6.0441379310344825,
381
+ "grad_norm": 21.811845779418945,
382
+ "learning_rate": 1.5512820512820516e-05,
383
+ "loss": 1.0928,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.182068965517241,
388
+ "grad_norm": 49.65191650390625,
389
+ "learning_rate": 1.5360195360195362e-05,
390
+ "loss": 0.6468,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.32,
395
+ "grad_norm": 55.17049789428711,
396
+ "learning_rate": 1.5207570207570207e-05,
397
+ "loss": 0.6764,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.457931034482758,
402
+ "grad_norm": 50.09459686279297,
403
+ "learning_rate": 1.5054945054945056e-05,
404
+ "loss": 0.6792,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.595862068965517,
409
+ "grad_norm": 78.72917938232422,
410
+ "learning_rate": 1.4902319902319904e-05,
411
+ "loss": 0.5457,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.733793103448276,
416
+ "grad_norm": 55.84168243408203,
417
+ "learning_rate": 1.4749694749694752e-05,
418
+ "loss": 0.699,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.871724137931034,
423
+ "grad_norm": 89.6448745727539,
424
+ "learning_rate": 1.4597069597069597e-05,
425
+ "loss": 0.5713,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 7.0,
430
+ "eval_accuracy": 0.5422647527910686,
431
+ "eval_loss": 1.9264899492263794,
432
+ "eval_macro_f1": 0.5445446062391962,
433
+ "eval_runtime": 1.3129,
434
+ "eval_samples_per_second": 477.584,
435
+ "eval_steps_per_second": 30.468,
436
+ "eval_weighted_f1": 0.5432596186525799,
437
+ "step": 1274
438
+ },
439
+ {
440
+ "epoch": 7.0055172413793105,
441
+ "grad_norm": 14.94321346282959,
442
+ "learning_rate": 1.4444444444444446e-05,
443
+ "loss": 0.5426,
444
+ "step": 1275
445
+ },
446
+ {
447
+ "epoch": 7.1434482758620685,
448
+ "grad_norm": 19.083919525146484,
449
+ "learning_rate": 1.4291819291819293e-05,
450
+ "loss": 0.5013,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.281379310344827,
455
+ "grad_norm": 41.64915466308594,
456
+ "learning_rate": 1.4139194139194141e-05,
457
+ "loss": 0.4538,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.419310344827586,
462
+ "grad_norm": 92.08172607421875,
463
+ "learning_rate": 1.3986568986568988e-05,
464
+ "loss": 0.3412,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.557241379310344,
469
+ "grad_norm": 36.61614990234375,
470
+ "learning_rate": 1.3833943833943834e-05,
471
+ "loss": 0.4431,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.695172413793103,
476
+ "grad_norm": 30.905656814575195,
477
+ "learning_rate": 1.3681318681318683e-05,
478
+ "loss": 0.2944,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 7.833103448275862,
483
+ "grad_norm": 6.589932918548584,
484
+ "learning_rate": 1.3528693528693531e-05,
485
+ "loss": 0.3515,
486
+ "step": 1425
487
+ },
488
+ {
489
+ "epoch": 7.971034482758621,
490
+ "grad_norm": 21.302804946899414,
491
+ "learning_rate": 1.3376068376068376e-05,
492
+ "loss": 0.4887,
493
+ "step": 1450
494
+ },
495
+ {
496
+ "epoch": 8.0,
497
+ "eval_accuracy": 0.5247208931419458,
498
+ "eval_loss": 2.6525657176971436,
499
+ "eval_macro_f1": 0.525906212470859,
500
+ "eval_runtime": 1.2638,
501
+ "eval_samples_per_second": 496.117,
502
+ "eval_steps_per_second": 31.65,
503
+ "eval_weighted_f1": 0.5222064951790744,
504
+ "step": 1456
505
+ },
506
+ {
507
+ "epoch": 8.104827586206897,
508
+ "grad_norm": 8.977629661560059,
509
+ "learning_rate": 1.3223443223443224e-05,
510
+ "loss": 0.49,
511
+ "step": 1475
512
+ },
513
+ {
514
+ "epoch": 8.242758620689655,
515
+ "grad_norm": 54.29226303100586,
516
+ "learning_rate": 1.3070818070818073e-05,
517
+ "loss": 0.1912,
518
+ "step": 1500
519
+ },
520
+ {
521
+ "epoch": 8.380689655172414,
522
+ "grad_norm": 47.832183837890625,
523
+ "learning_rate": 1.291819291819292e-05,
524
+ "loss": 0.3511,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 8.518620689655172,
529
+ "grad_norm": 3.3483729362487793,
530
+ "learning_rate": 1.2765567765567766e-05,
531
+ "loss": 0.2213,
532
+ "step": 1550
533
+ },
534
+ {
535
+ "epoch": 8.65655172413793,
536
+ "grad_norm": 38.751686096191406,
537
+ "learning_rate": 1.2612942612942613e-05,
538
+ "loss": 0.2785,
539
+ "step": 1575
540
+ },
541
+ {
542
+ "epoch": 8.79448275862069,
543
+ "grad_norm": 55.66695785522461,
544
+ "learning_rate": 1.2460317460317461e-05,
545
+ "loss": 0.2519,
546
+ "step": 1600
547
+ },
548
+ {
549
+ "epoch": 8.932413793103448,
550
+ "grad_norm": 16.75074577331543,
551
+ "learning_rate": 1.230769230769231e-05,
552
+ "loss": 0.2295,
553
+ "step": 1625
554
+ },
555
+ {
556
+ "epoch": 9.0,
557
+ "eval_accuracy": 0.5709728867623605,
558
+ "eval_loss": 2.661311626434326,
559
+ "eval_macro_f1": 0.5726049605469152,
560
+ "eval_runtime": 1.2867,
561
+ "eval_samples_per_second": 487.281,
562
+ "eval_steps_per_second": 31.087,
563
+ "eval_weighted_f1": 0.5721256126906006,
564
+ "step": 1638
565
+ }
566
+ ],
567
+ "logging_steps": 25,
568
+ "max_steps": 3640,
569
+ "num_input_tokens_seen": 0,
570
+ "num_train_epochs": 20,
571
+ "save_steps": 500,
572
+ "stateful_callbacks": {
573
+ "EarlyStoppingCallback": {
574
+ "args": {
575
+ "early_stopping_patience": 3,
576
+ "early_stopping_threshold": 0.0
577
+ },
578
+ "attributes": {
579
+ "early_stopping_patience_counter": 0
580
+ }
581
+ },
582
+ "TrainerControl": {
583
+ "args": {
584
+ "should_epoch_stop": false,
585
+ "should_evaluate": false,
586
+ "should_log": false,
587
+ "should_save": true,
588
+ "should_training_stop": false
589
+ },
590
+ "attributes": {}
591
+ }
592
+ },
593
+ "total_flos": 0.0,
594
+ "train_batch_size": 8,
595
+ "trial_name": null,
596
+ "trial_params": null
597
+ }
ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-2184/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-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-2184/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-2184/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-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-2184/trainer_state.json ADDED
@@ -0,0 +1,784 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1638,
3
+ "best_metric": 0.5726049605469152,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-00-s42-20260830T232831Z-f21e2f7e/trainer/checkpoint-1638",
5
+ "epoch": 12.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2184,
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": 37.184391021728516,
15
+ "learning_rate": 1.3186813186813187e-06,
16
+ "loss": 4.7284,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27586206896551724,
21
+ "grad_norm": 31.35881805419922,
22
+ "learning_rate": 2.6923076923076923e-06,
23
+ "loss": 4.5436,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.41379310344827586,
28
+ "grad_norm": 40.47816467285156,
29
+ "learning_rate": 4.065934065934066e-06,
30
+ "loss": 4.5382,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5517241379310345,
35
+ "grad_norm": 28.50191307067871,
36
+ "learning_rate": 5.43956043956044e-06,
37
+ "loss": 4.4636,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6896551724137931,
42
+ "grad_norm": 29.587844848632812,
43
+ "learning_rate": 6.813186813186814e-06,
44
+ "loss": 4.4057,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8275862068965517,
49
+ "grad_norm": 19.32939338684082,
50
+ "learning_rate": 8.186813186813188e-06,
51
+ "loss": 4.3842,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9655172413793104,
56
+ "grad_norm": 30.56195831298828,
57
+ "learning_rate": 9.560439560439562e-06,
58
+ "loss": 4.387,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.40669856459330145,
64
+ "eval_loss": 1.0807603597640991,
65
+ "eval_macro_f1": 0.3250454610810856,
66
+ "eval_runtime": 1.3826,
67
+ "eval_samples_per_second": 453.479,
68
+ "eval_steps_per_second": 28.93,
69
+ "eval_weighted_f1": 0.3338543425140942,
70
+ "step": 182
71
+ },
72
+ {
73
+ "epoch": 1.0993103448275863,
74
+ "grad_norm": 45.888633728027344,
75
+ "learning_rate": 1.0934065934065933e-05,
76
+ "loss": 4.2564,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.237241379310345,
81
+ "grad_norm": 16.613557815551758,
82
+ "learning_rate": 1.230769230769231e-05,
83
+ "loss": 4.2484,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3751724137931034,
88
+ "grad_norm": 28.772672653198242,
89
+ "learning_rate": 1.3681318681318683e-05,
90
+ "loss": 4.24,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.513103448275862,
95
+ "grad_norm": 27.18781852722168,
96
+ "learning_rate": 1.5054945054945056e-05,
97
+ "loss": 4.2858,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6510344827586207,
102
+ "grad_norm": 19.768953323364258,
103
+ "learning_rate": 1.642857142857143e-05,
104
+ "loss": 4.2016,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7889655172413792,
109
+ "grad_norm": 20.3443660736084,
110
+ "learning_rate": 1.78021978021978e-05,
111
+ "loss": 4.2666,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.926896551724138,
116
+ "grad_norm": 20.553537368774414,
117
+ "learning_rate": 1.9175824175824178e-05,
118
+ "loss": 4.2036,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.42902711323763953,
124
+ "eval_loss": 1.118101716041565,
125
+ "eval_macro_f1": 0.36056831206341905,
126
+ "eval_runtime": 1.2323,
127
+ "eval_samples_per_second": 508.803,
128
+ "eval_steps_per_second": 32.46,
129
+ "eval_weighted_f1": 0.3488534356774562,
130
+ "step": 364
131
+ },
132
+ {
133
+ "epoch": 2.0606896551724136,
134
+ "grad_norm": 34.58537292480469,
135
+ "learning_rate": 1.993894993894994e-05,
136
+ "loss": 4.0065,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.1986206896551725,
141
+ "grad_norm": 25.910797119140625,
142
+ "learning_rate": 1.9786324786324788e-05,
143
+ "loss": 3.8394,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.336551724137931,
148
+ "grad_norm": 40.513763427734375,
149
+ "learning_rate": 1.9633699633699634e-05,
150
+ "loss": 3.7494,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.47448275862069,
155
+ "grad_norm": 36.02470397949219,
156
+ "learning_rate": 1.948107448107448e-05,
157
+ "loss": 3.8057,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.6124137931034483,
162
+ "grad_norm": 45.625755310058594,
163
+ "learning_rate": 1.932844932844933e-05,
164
+ "loss": 3.754,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.750344827586207,
169
+ "grad_norm": 32.18267059326172,
170
+ "learning_rate": 1.9175824175824178e-05,
171
+ "loss": 3.7035,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.8882758620689657,
176
+ "grad_norm": 32.876625061035156,
177
+ "learning_rate": 1.9023199023199025e-05,
178
+ "loss": 3.6018,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5007974481658692,
184
+ "eval_loss": 0.984271228313446,
185
+ "eval_macro_f1": 0.49151323654018847,
186
+ "eval_runtime": 1.221,
187
+ "eval_samples_per_second": 513.52,
188
+ "eval_steps_per_second": 32.76,
189
+ "eval_weighted_f1": 0.48633333130427775,
190
+ "step": 546
191
+ },
192
+ {
193
+ "epoch": 3.0220689655172412,
194
+ "grad_norm": 50.10889434814453,
195
+ "learning_rate": 1.887057387057387e-05,
196
+ "loss": 3.3932,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.16,
201
+ "grad_norm": 56.40572738647461,
202
+ "learning_rate": 1.8717948717948718e-05,
203
+ "loss": 2.9721,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.2979310344827586,
208
+ "grad_norm": 36.12623596191406,
209
+ "learning_rate": 1.8565323565323568e-05,
210
+ "loss": 2.5898,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.435862068965517,
215
+ "grad_norm": 28.53684425354004,
216
+ "learning_rate": 1.8412698412698415e-05,
217
+ "loss": 2.7856,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.573793103448276,
222
+ "grad_norm": 29.99932098388672,
223
+ "learning_rate": 1.826007326007326e-05,
224
+ "loss": 2.9761,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.7117241379310344,
229
+ "grad_norm": 38.28874206542969,
230
+ "learning_rate": 1.8107448107448108e-05,
231
+ "loss": 2.9072,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.849655172413793,
236
+ "grad_norm": 76.56749725341797,
237
+ "learning_rate": 1.7954822954822958e-05,
238
+ "loss": 3.0844,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.9875862068965517,
243
+ "grad_norm": 40.25591278076172,
244
+ "learning_rate": 1.78021978021978e-05,
245
+ "loss": 3.0495,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 4.0,
250
+ "eval_accuracy": 0.5311004784688995,
251
+ "eval_loss": 1.0743650197982788,
252
+ "eval_macro_f1": 0.5205468214965933,
253
+ "eval_runtime": 1.208,
254
+ "eval_samples_per_second": 519.056,
255
+ "eval_steps_per_second": 33.114,
256
+ "eval_weighted_f1": 0.5175166903640859,
257
+ "step": 728
258
+ },
259
+ {
260
+ "epoch": 4.121379310344827,
261
+ "grad_norm": 41.000728607177734,
262
+ "learning_rate": 1.764957264957265e-05,
263
+ "loss": 1.8659,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.259310344827586,
268
+ "grad_norm": 29.559438705444336,
269
+ "learning_rate": 1.74969474969475e-05,
270
+ "loss": 1.9239,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.397241379310345,
275
+ "grad_norm": 38.09396743774414,
276
+ "learning_rate": 1.7344322344322345e-05,
277
+ "loss": 1.9678,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.535172413793103,
282
+ "grad_norm": 50.2789192199707,
283
+ "learning_rate": 1.7191697191697195e-05,
284
+ "loss": 2.0875,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.673103448275862,
289
+ "grad_norm": 30.099040985107422,
290
+ "learning_rate": 1.703907203907204e-05,
291
+ "loss": 1.8674,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.811034482758621,
296
+ "grad_norm": 75.6994400024414,
297
+ "learning_rate": 1.688644688644689e-05,
298
+ "loss": 1.9013,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.94896551724138,
303
+ "grad_norm": 40.34619903564453,
304
+ "learning_rate": 1.6733821733821735e-05,
305
+ "loss": 1.9484,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 5.0,
310
+ "eval_accuracy": 0.5390749601275917,
311
+ "eval_loss": 1.274722695350647,
312
+ "eval_macro_f1": 0.5310477318972339,
313
+ "eval_runtime": 1.3175,
314
+ "eval_samples_per_second": 475.899,
315
+ "eval_steps_per_second": 30.36,
316
+ "eval_weighted_f1": 0.5321559967817984,
317
+ "step": 910
318
+ },
319
+ {
320
+ "epoch": 5.082758620689655,
321
+ "grad_norm": 62.8062858581543,
322
+ "learning_rate": 1.6581196581196585e-05,
323
+ "loss": 1.5826,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.220689655172414,
328
+ "grad_norm": 92.9386215209961,
329
+ "learning_rate": 1.642857142857143e-05,
330
+ "loss": 1.2958,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.358620689655172,
335
+ "grad_norm": 40.67005157470703,
336
+ "learning_rate": 1.627594627594628e-05,
337
+ "loss": 1.0679,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.496551724137931,
342
+ "grad_norm": 35.90666198730469,
343
+ "learning_rate": 1.6123321123321125e-05,
344
+ "loss": 1.4217,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.63448275862069,
349
+ "grad_norm": 74.77809143066406,
350
+ "learning_rate": 1.5970695970695972e-05,
351
+ "loss": 1.2528,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.772413793103448,
356
+ "grad_norm": 32.45465850830078,
357
+ "learning_rate": 1.581807081807082e-05,
358
+ "loss": 1.0701,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.910344827586207,
363
+ "grad_norm": 57.50745391845703,
364
+ "learning_rate": 1.5665445665445665e-05,
365
+ "loss": 0.983,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 6.0,
370
+ "eval_accuracy": 0.5502392344497608,
371
+ "eval_loss": 1.5191906690597534,
372
+ "eval_macro_f1": 0.5442008282447773,
373
+ "eval_runtime": 1.3112,
374
+ "eval_samples_per_second": 478.194,
375
+ "eval_steps_per_second": 30.507,
376
+ "eval_weighted_f1": 0.5438193474963617,
377
+ "step": 1092
378
+ },
379
+ {
380
+ "epoch": 6.0441379310344825,
381
+ "grad_norm": 21.811845779418945,
382
+ "learning_rate": 1.5512820512820516e-05,
383
+ "loss": 1.0928,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.182068965517241,
388
+ "grad_norm": 49.65191650390625,
389
+ "learning_rate": 1.5360195360195362e-05,
390
+ "loss": 0.6468,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.32,
395
+ "grad_norm": 55.17049789428711,
396
+ "learning_rate": 1.5207570207570207e-05,
397
+ "loss": 0.6764,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.457931034482758,
402
+ "grad_norm": 50.09459686279297,
403
+ "learning_rate": 1.5054945054945056e-05,
404
+ "loss": 0.6792,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.595862068965517,
409
+ "grad_norm": 78.72917938232422,
410
+ "learning_rate": 1.4902319902319904e-05,
411
+ "loss": 0.5457,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.733793103448276,
416
+ "grad_norm": 55.84168243408203,
417
+ "learning_rate": 1.4749694749694752e-05,
418
+ "loss": 0.699,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.871724137931034,
423
+ "grad_norm": 89.6448745727539,
424
+ "learning_rate": 1.4597069597069597e-05,
425
+ "loss": 0.5713,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 7.0,
430
+ "eval_accuracy": 0.5422647527910686,
431
+ "eval_loss": 1.9264899492263794,
432
+ "eval_macro_f1": 0.5445446062391962,
433
+ "eval_runtime": 1.3129,
434
+ "eval_samples_per_second": 477.584,
435
+ "eval_steps_per_second": 30.468,
436
+ "eval_weighted_f1": 0.5432596186525799,
437
+ "step": 1274
438
+ },
439
+ {
440
+ "epoch": 7.0055172413793105,
441
+ "grad_norm": 14.94321346282959,
442
+ "learning_rate": 1.4444444444444446e-05,
443
+ "loss": 0.5426,
444
+ "step": 1275
445
+ },
446
+ {
447
+ "epoch": 7.1434482758620685,
448
+ "grad_norm": 19.083919525146484,
449
+ "learning_rate": 1.4291819291819293e-05,
450
+ "loss": 0.5013,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.281379310344827,
455
+ "grad_norm": 41.64915466308594,
456
+ "learning_rate": 1.4139194139194141e-05,
457
+ "loss": 0.4538,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.419310344827586,
462
+ "grad_norm": 92.08172607421875,
463
+ "learning_rate": 1.3986568986568988e-05,
464
+ "loss": 0.3412,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.557241379310344,
469
+ "grad_norm": 36.61614990234375,
470
+ "learning_rate": 1.3833943833943834e-05,
471
+ "loss": 0.4431,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.695172413793103,
476
+ "grad_norm": 30.905656814575195,
477
+ "learning_rate": 1.3681318681318683e-05,
478
+ "loss": 0.2944,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 7.833103448275862,
483
+ "grad_norm": 6.589932918548584,
484
+ "learning_rate": 1.3528693528693531e-05,
485
+ "loss": 0.3515,
486
+ "step": 1425
487
+ },
488
+ {
489
+ "epoch": 7.971034482758621,
490
+ "grad_norm": 21.302804946899414,
491
+ "learning_rate": 1.3376068376068376e-05,
492
+ "loss": 0.4887,
493
+ "step": 1450
494
+ },
495
+ {
496
+ "epoch": 8.0,
497
+ "eval_accuracy": 0.5247208931419458,
498
+ "eval_loss": 2.6525657176971436,
499
+ "eval_macro_f1": 0.525906212470859,
500
+ "eval_runtime": 1.2638,
501
+ "eval_samples_per_second": 496.117,
502
+ "eval_steps_per_second": 31.65,
503
+ "eval_weighted_f1": 0.5222064951790744,
504
+ "step": 1456
505
+ },
506
+ {
507
+ "epoch": 8.104827586206897,
508
+ "grad_norm": 8.977629661560059,
509
+ "learning_rate": 1.3223443223443224e-05,
510
+ "loss": 0.49,
511
+ "step": 1475
512
+ },
513
+ {
514
+ "epoch": 8.242758620689655,
515
+ "grad_norm": 54.29226303100586,
516
+ "learning_rate": 1.3070818070818073e-05,
517
+ "loss": 0.1912,
518
+ "step": 1500
519
+ },
520
+ {
521
+ "epoch": 8.380689655172414,
522
+ "grad_norm": 47.832183837890625,
523
+ "learning_rate": 1.291819291819292e-05,
524
+ "loss": 0.3511,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 8.518620689655172,
529
+ "grad_norm": 3.3483729362487793,
530
+ "learning_rate": 1.2765567765567766e-05,
531
+ "loss": 0.2213,
532
+ "step": 1550
533
+ },
534
+ {
535
+ "epoch": 8.65655172413793,
536
+ "grad_norm": 38.751686096191406,
537
+ "learning_rate": 1.2612942612942613e-05,
538
+ "loss": 0.2785,
539
+ "step": 1575
540
+ },
541
+ {
542
+ "epoch": 8.79448275862069,
543
+ "grad_norm": 55.66695785522461,
544
+ "learning_rate": 1.2460317460317461e-05,
545
+ "loss": 0.2519,
546
+ "step": 1600
547
+ },
548
+ {
549
+ "epoch": 8.932413793103448,
550
+ "grad_norm": 16.75074577331543,
551
+ "learning_rate": 1.230769230769231e-05,
552
+ "loss": 0.2295,
553
+ "step": 1625
554
+ },
555
+ {
556
+ "epoch": 9.0,
557
+ "eval_accuracy": 0.5709728867623605,
558
+ "eval_loss": 2.661311626434326,
559
+ "eval_macro_f1": 0.5726049605469152,
560
+ "eval_runtime": 1.2867,
561
+ "eval_samples_per_second": 487.281,
562
+ "eval_steps_per_second": 31.087,
563
+ "eval_weighted_f1": 0.5721256126906006,
564
+ "step": 1638
565
+ },
566
+ {
567
+ "epoch": 9.066206896551725,
568
+ "grad_norm": 61.49134063720703,
569
+ "learning_rate": 1.2155067155067155e-05,
570
+ "loss": 0.3155,
571
+ "step": 1650
572
+ },
573
+ {
574
+ "epoch": 9.204137931034483,
575
+ "grad_norm": 5.640354156494141,
576
+ "learning_rate": 1.2002442002442003e-05,
577
+ "loss": 0.1485,
578
+ "step": 1675
579
+ },
580
+ {
581
+ "epoch": 9.34206896551724,
582
+ "grad_norm": 84.32801818847656,
583
+ "learning_rate": 1.1849816849816852e-05,
584
+ "loss": 0.3296,
585
+ "step": 1700
586
+ },
587
+ {
588
+ "epoch": 9.48,
589
+ "grad_norm": 6.50032377243042,
590
+ "learning_rate": 1.1697191697191698e-05,
591
+ "loss": 0.2997,
592
+ "step": 1725
593
+ },
594
+ {
595
+ "epoch": 9.617931034482758,
596
+ "grad_norm": 61.773460388183594,
597
+ "learning_rate": 1.1544566544566545e-05,
598
+ "loss": 0.2432,
599
+ "step": 1750
600
+ },
601
+ {
602
+ "epoch": 9.755862068965516,
603
+ "grad_norm": 78.97508239746094,
604
+ "learning_rate": 1.1391941391941392e-05,
605
+ "loss": 0.1147,
606
+ "step": 1775
607
+ },
608
+ {
609
+ "epoch": 9.893793103448276,
610
+ "grad_norm": 160.76438903808594,
611
+ "learning_rate": 1.123931623931624e-05,
612
+ "loss": 0.1133,
613
+ "step": 1800
614
+ },
615
+ {
616
+ "epoch": 10.0,
617
+ "eval_accuracy": 0.5486443381180224,
618
+ "eval_loss": 3.2329728603363037,
619
+ "eval_macro_f1": 0.5470959178853272,
620
+ "eval_runtime": 1.2727,
621
+ "eval_samples_per_second": 492.653,
622
+ "eval_steps_per_second": 31.429,
623
+ "eval_weighted_f1": 0.5447599054372868,
624
+ "step": 1820
625
+ },
626
+ {
627
+ "epoch": 10.027586206896551,
628
+ "grad_norm": 16.743375778198242,
629
+ "learning_rate": 1.1086691086691088e-05,
630
+ "loss": 0.253,
631
+ "step": 1825
632
+ },
633
+ {
634
+ "epoch": 10.16551724137931,
635
+ "grad_norm": 11.530265808105469,
636
+ "learning_rate": 1.0934065934065933e-05,
637
+ "loss": 0.0636,
638
+ "step": 1850
639
+ },
640
+ {
641
+ "epoch": 10.303448275862069,
642
+ "grad_norm": 2.327833414077759,
643
+ "learning_rate": 1.0781440781440782e-05,
644
+ "loss": 0.1609,
645
+ "step": 1875
646
+ },
647
+ {
648
+ "epoch": 10.441379310344828,
649
+ "grad_norm": 8.823471069335938,
650
+ "learning_rate": 1.062881562881563e-05,
651
+ "loss": 0.2276,
652
+ "step": 1900
653
+ },
654
+ {
655
+ "epoch": 10.579310344827586,
656
+ "grad_norm": 43.53721237182617,
657
+ "learning_rate": 1.0476190476190477e-05,
658
+ "loss": 0.1655,
659
+ "step": 1925
660
+ },
661
+ {
662
+ "epoch": 10.717241379310344,
663
+ "grad_norm": 13.283941268920898,
664
+ "learning_rate": 1.0323565323565325e-05,
665
+ "loss": 0.1537,
666
+ "step": 1950
667
+ },
668
+ {
669
+ "epoch": 10.855172413793104,
670
+ "grad_norm": 94.66608428955078,
671
+ "learning_rate": 1.0170940170940172e-05,
672
+ "loss": 0.2133,
673
+ "step": 1975
674
+ },
675
+ {
676
+ "epoch": 10.993103448275862,
677
+ "grad_norm": 0.07509344071149826,
678
+ "learning_rate": 1.0018315018315019e-05,
679
+ "loss": 0.1114,
680
+ "step": 2000
681
+ },
682
+ {
683
+ "epoch": 11.0,
684
+ "eval_accuracy": 0.5614035087719298,
685
+ "eval_loss": 3.749659776687622,
686
+ "eval_macro_f1": 0.5621867110340407,
687
+ "eval_runtime": 1.2748,
688
+ "eval_samples_per_second": 491.848,
689
+ "eval_steps_per_second": 31.378,
690
+ "eval_weighted_f1": 0.5609011753866963,
691
+ "step": 2002
692
+ },
693
+ {
694
+ "epoch": 11.126896551724139,
695
+ "grad_norm": 15.820819854736328,
696
+ "learning_rate": 9.865689865689867e-06,
697
+ "loss": 0.077,
698
+ "step": 2025
699
+ },
700
+ {
701
+ "epoch": 11.264827586206897,
702
+ "grad_norm": 107.31636047363281,
703
+ "learning_rate": 9.713064713064714e-06,
704
+ "loss": 0.1285,
705
+ "step": 2050
706
+ },
707
+ {
708
+ "epoch": 11.402758620689655,
709
+ "grad_norm": 0.5315859913825989,
710
+ "learning_rate": 9.560439560439562e-06,
711
+ "loss": 0.157,
712
+ "step": 2075
713
+ },
714
+ {
715
+ "epoch": 11.540689655172415,
716
+ "grad_norm": 0.2875083088874817,
717
+ "learning_rate": 9.407814407814409e-06,
718
+ "loss": 0.041,
719
+ "step": 2100
720
+ },
721
+ {
722
+ "epoch": 11.678620689655173,
723
+ "grad_norm": 97.9702377319336,
724
+ "learning_rate": 9.255189255189256e-06,
725
+ "loss": 0.1475,
726
+ "step": 2125
727
+ },
728
+ {
729
+ "epoch": 11.81655172413793,
730
+ "grad_norm": 164.09478759765625,
731
+ "learning_rate": 9.102564102564104e-06,
732
+ "loss": 0.1397,
733
+ "step": 2150
734
+ },
735
+ {
736
+ "epoch": 11.95448275862069,
737
+ "grad_norm": 41.94971466064453,
738
+ "learning_rate": 8.94993894993895e-06,
739
+ "loss": 0.1254,
740
+ "step": 2175
741
+ },
742
+ {
743
+ "epoch": 12.0,
744
+ "eval_accuracy": 0.5518341307814992,
745
+ "eval_loss": 3.9953196048736572,
746
+ "eval_macro_f1": 0.5506048991912492,
747
+ "eval_runtime": 1.3348,
748
+ "eval_samples_per_second": 469.744,
749
+ "eval_steps_per_second": 29.968,
750
+ "eval_weighted_f1": 0.5515593631395048,
751
+ "step": 2184
752
+ }
753
+ ],
754
+ "logging_steps": 25,
755
+ "max_steps": 3640,
756
+ "num_input_tokens_seen": 0,
757
+ "num_train_epochs": 20,
758
+ "save_steps": 500,
759
+ "stateful_callbacks": {
760
+ "EarlyStoppingCallback": {
761
+ "args": {
762
+ "early_stopping_patience": 3,
763
+ "early_stopping_threshold": 0.0
764
+ },
765
+ "attributes": {
766
+ "early_stopping_patience_counter": 3
767
+ }
768
+ },
769
+ "TrainerControl": {
770
+ "args": {
771
+ "should_epoch_stop": false,
772
+ "should_evaluate": false,
773
+ "should_log": false,
774
+ "should_save": true,
775
+ "should_training_stop": true
776
+ },
777
+ "attributes": {}
778
+ }
779
+ },
780
+ "total_flos": 0.0,
781
+ "train_batch_size": 8,
782
+ "trial_name": null,
783
+ "trial_params": null
784
+ }
ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1026/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-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1026/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1026/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-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1026/trainer_state.json ADDED
@@ -0,0 +1,396 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1026,
3
+ "best_metric": 0.5893419847764664,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1026",
5
+ "epoch": 6.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1026,
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": 25.180173873901367,
15
+ "learning_rate": 1.4035087719298246e-06,
16
+ "loss": 4.6414,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.2936857562408223,
21
+ "grad_norm": 25.767900466918945,
22
+ "learning_rate": 2.865497076023392e-06,
23
+ "loss": 4.5057,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.44052863436123346,
28
+ "grad_norm": 22.705293655395508,
29
+ "learning_rate": 4.3274853801169596e-06,
30
+ "loss": 4.4364,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5873715124816447,
35
+ "grad_norm": 21.249404907226562,
36
+ "learning_rate": 5.789473684210527e-06,
37
+ "loss": 4.4985,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.7342143906020558,
42
+ "grad_norm": 17.13970947265625,
43
+ "learning_rate": 7.251461988304094e-06,
44
+ "loss": 4.38,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8810572687224669,
49
+ "grad_norm": 25.133960723876953,
50
+ "learning_rate": 8.713450292397661e-06,
51
+ "loss": 4.3492,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 1.0,
56
+ "eval_accuracy": 0.3879003558718861,
57
+ "eval_loss": 1.0845022201538086,
58
+ "eval_macro_f1": 0.3415916490986263,
59
+ "eval_runtime": 1.2086,
60
+ "eval_samples_per_second": 465.006,
61
+ "eval_steps_per_second": 29.787,
62
+ "eval_weighted_f1": 0.3394971773692272,
63
+ "step": 171
64
+ },
65
+ {
66
+ "epoch": 1.0234948604992657,
67
+ "grad_norm": 36.66134262084961,
68
+ "learning_rate": 1.017543859649123e-05,
69
+ "loss": 4.2753,
70
+ "step": 175
71
+ },
72
+ {
73
+ "epoch": 1.170337738619677,
74
+ "grad_norm": 17.75954246520996,
75
+ "learning_rate": 1.1637426900584797e-05,
76
+ "loss": 4.2327,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.3171806167400881,
81
+ "grad_norm": 22.90311050415039,
82
+ "learning_rate": 1.3099415204678362e-05,
83
+ "loss": 4.2055,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.4640234948604993,
88
+ "grad_norm": 32.75823211669922,
89
+ "learning_rate": 1.4561403508771931e-05,
90
+ "loss": 4.331,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.6108663729809103,
95
+ "grad_norm": 25.388853073120117,
96
+ "learning_rate": 1.60233918128655e-05,
97
+ "loss": 4.3045,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.7577092511013217,
102
+ "grad_norm": 29.442508697509766,
103
+ "learning_rate": 1.7485380116959064e-05,
104
+ "loss": 4.3285,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.9045521292217327,
109
+ "grad_norm": 24.855016708374023,
110
+ "learning_rate": 1.894736842105263e-05,
111
+ "loss": 4.246,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 2.0,
116
+ "eval_accuracy": 0.40747330960854095,
117
+ "eval_loss": 1.1078580617904663,
118
+ "eval_macro_f1": 0.3537153700705074,
119
+ "eval_runtime": 1.1268,
120
+ "eval_samples_per_second": 498.777,
121
+ "eval_steps_per_second": 31.95,
122
+ "eval_weighted_f1": 0.3479225287782979,
123
+ "step": 342
124
+ },
125
+ {
126
+ "epoch": 2.0469897209985315,
127
+ "grad_norm": 22.83465003967285,
128
+ "learning_rate": 1.99545159194282e-05,
129
+ "loss": 3.9936,
130
+ "step": 350
131
+ },
132
+ {
133
+ "epoch": 2.193832599118943,
134
+ "grad_norm": 15.519044876098633,
135
+ "learning_rate": 1.9792072774528916e-05,
136
+ "loss": 3.9323,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.340675477239354,
141
+ "grad_norm": 26.67107391357422,
142
+ "learning_rate": 1.962962962962963e-05,
143
+ "loss": 3.7592,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.4875183553597653,
148
+ "grad_norm": 19.69229507446289,
149
+ "learning_rate": 1.9467186484730346e-05,
150
+ "loss": 3.783,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.6343612334801763,
155
+ "grad_norm": 24.198707580566406,
156
+ "learning_rate": 1.930474333983106e-05,
157
+ "loss": 3.8374,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.7812041116005872,
162
+ "grad_norm": 35.194175720214844,
163
+ "learning_rate": 1.9142300194931776e-05,
164
+ "loss": 3.7651,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.9280469897209986,
169
+ "grad_norm": 32.717124938964844,
170
+ "learning_rate": 1.897985705003249e-05,
171
+ "loss": 3.6372,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 3.0,
176
+ "eval_accuracy": 0.5498220640569395,
177
+ "eval_loss": 0.9409611821174622,
178
+ "eval_macro_f1": 0.5369368641707378,
179
+ "eval_runtime": 1.1652,
180
+ "eval_samples_per_second": 482.301,
181
+ "eval_steps_per_second": 30.895,
182
+ "eval_weighted_f1": 0.5375987962984153,
183
+ "step": 513
184
+ },
185
+ {
186
+ "epoch": 3.0704845814977975,
187
+ "grad_norm": 25.14092445373535,
188
+ "learning_rate": 1.8817413905133203e-05,
189
+ "loss": 3.1335,
190
+ "step": 525
191
+ },
192
+ {
193
+ "epoch": 3.2173274596182084,
194
+ "grad_norm": 33.47543716430664,
195
+ "learning_rate": 1.8654970760233918e-05,
196
+ "loss": 2.9605,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.36417033773862,
201
+ "grad_norm": 42.239418029785156,
202
+ "learning_rate": 1.8492527615334633e-05,
203
+ "loss": 3.0959,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.511013215859031,
208
+ "grad_norm": 27.37813377380371,
209
+ "learning_rate": 1.833008447043535e-05,
210
+ "loss": 2.9869,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.657856093979442,
215
+ "grad_norm": 30.336503982543945,
216
+ "learning_rate": 1.8167641325536064e-05,
217
+ "loss": 2.9882,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.804698972099853,
222
+ "grad_norm": 54.229408264160156,
223
+ "learning_rate": 1.800519818063678e-05,
224
+ "loss": 2.9932,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.951541850220264,
229
+ "grad_norm": 41.398597717285156,
230
+ "learning_rate": 1.7842755035737494e-05,
231
+ "loss": 2.7499,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 4.0,
236
+ "eval_accuracy": 0.5765124555160143,
237
+ "eval_loss": 0.9548276662826538,
238
+ "eval_macro_f1": 0.5745254022822105,
239
+ "eval_runtime": 1.2157,
240
+ "eval_samples_per_second": 462.274,
241
+ "eval_steps_per_second": 29.612,
242
+ "eval_weighted_f1": 0.5748016121055329,
243
+ "step": 684
244
+ },
245
+ {
246
+ "epoch": 4.093979441997063,
247
+ "grad_norm": 51.15410614013672,
248
+ "learning_rate": 1.768031189083821e-05,
249
+ "loss": 2.2844,
250
+ "step": 700
251
+ },
252
+ {
253
+ "epoch": 4.240822320117474,
254
+ "grad_norm": 47.037986755371094,
255
+ "learning_rate": 1.751786874593892e-05,
256
+ "loss": 1.936,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.387665198237886,
261
+ "grad_norm": 53.382972717285156,
262
+ "learning_rate": 1.7355425601039636e-05,
263
+ "loss": 1.827,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.534508076358296,
268
+ "grad_norm": 58.742332458496094,
269
+ "learning_rate": 1.719298245614035e-05,
270
+ "loss": 2.2023,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.681350954478708,
275
+ "grad_norm": 50.0244026184082,
276
+ "learning_rate": 1.7030539311241066e-05,
277
+ "loss": 1.8348,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.828193832599119,
282
+ "grad_norm": 73.3564682006836,
283
+ "learning_rate": 1.686809616634178e-05,
284
+ "loss": 1.8355,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.975036710719531,
289
+ "grad_norm": 64.41809844970703,
290
+ "learning_rate": 1.6705653021442497e-05,
291
+ "loss": 1.8025,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 5.0,
296
+ "eval_accuracy": 0.5533807829181495,
297
+ "eval_loss": 1.130232572555542,
298
+ "eval_macro_f1": 0.5544698179166704,
299
+ "eval_runtime": 1.1418,
300
+ "eval_samples_per_second": 492.203,
301
+ "eval_steps_per_second": 31.529,
302
+ "eval_weighted_f1": 0.5533888776357623,
303
+ "step": 855
304
+ },
305
+ {
306
+ "epoch": 5.117474302496329,
307
+ "grad_norm": 33.71281051635742,
308
+ "learning_rate": 1.654320987654321e-05,
309
+ "loss": 1.1478,
310
+ "step": 875
311
+ },
312
+ {
313
+ "epoch": 5.26431718061674,
314
+ "grad_norm": 60.219451904296875,
315
+ "learning_rate": 1.6380766731643927e-05,
316
+ "loss": 1.0621,
317
+ "step": 900
318
+ },
319
+ {
320
+ "epoch": 5.411160058737151,
321
+ "grad_norm": 69.2081069946289,
322
+ "learning_rate": 1.621832358674464e-05,
323
+ "loss": 1.1548,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.558002936857562,
328
+ "grad_norm": 43.02848815917969,
329
+ "learning_rate": 1.6055880441845354e-05,
330
+ "loss": 0.985,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.704845814977974,
335
+ "grad_norm": 24.841623306274414,
336
+ "learning_rate": 1.589343729694607e-05,
337
+ "loss": 1.079,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.851688693098385,
342
+ "grad_norm": 41.482872009277344,
343
+ "learning_rate": 1.5730994152046787e-05,
344
+ "loss": 1.1817,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.998531571218796,
349
+ "grad_norm": 25.20818328857422,
350
+ "learning_rate": 1.55685510071475e-05,
351
+ "loss": 1.047,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 6.0,
356
+ "eval_accuracy": 0.599644128113879,
357
+ "eval_loss": 1.3713058233261108,
358
+ "eval_macro_f1": 0.5893419847764664,
359
+ "eval_runtime": 1.2066,
360
+ "eval_samples_per_second": 465.773,
361
+ "eval_steps_per_second": 29.836,
362
+ "eval_weighted_f1": 0.590349818049014,
363
+ "step": 1026
364
+ }
365
+ ],
366
+ "logging_steps": 25,
367
+ "max_steps": 3420,
368
+ "num_input_tokens_seen": 0,
369
+ "num_train_epochs": 20,
370
+ "save_steps": 500,
371
+ "stateful_callbacks": {
372
+ "EarlyStoppingCallback": {
373
+ "args": {
374
+ "early_stopping_patience": 3,
375
+ "early_stopping_threshold": 0.0
376
+ },
377
+ "attributes": {
378
+ "early_stopping_patience_counter": 0
379
+ }
380
+ },
381
+ "TrainerControl": {
382
+ "args": {
383
+ "should_epoch_stop": false,
384
+ "should_evaluate": false,
385
+ "should_log": false,
386
+ "should_save": true,
387
+ "should_training_stop": false
388
+ },
389
+ "attributes": {}
390
+ }
391
+ },
392
+ "total_flos": 0.0,
393
+ "train_batch_size": 8,
394
+ "trial_name": null,
395
+ "trial_params": null
396
+ }
ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1539/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-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1539/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1539/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-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1539/trainer_state.json ADDED
@@ -0,0 +1,569 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1026,
3
+ "best_metric": 0.5893419847764664,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-01-s42-20260830T234048Z-f21e2f7e/trainer/checkpoint-1026",
5
+ "epoch": 9.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1539,
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": 25.180173873901367,
15
+ "learning_rate": 1.4035087719298246e-06,
16
+ "loss": 4.6414,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.2936857562408223,
21
+ "grad_norm": 25.767900466918945,
22
+ "learning_rate": 2.865497076023392e-06,
23
+ "loss": 4.5057,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.44052863436123346,
28
+ "grad_norm": 22.705293655395508,
29
+ "learning_rate": 4.3274853801169596e-06,
30
+ "loss": 4.4364,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5873715124816447,
35
+ "grad_norm": 21.249404907226562,
36
+ "learning_rate": 5.789473684210527e-06,
37
+ "loss": 4.4985,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.7342143906020558,
42
+ "grad_norm": 17.13970947265625,
43
+ "learning_rate": 7.251461988304094e-06,
44
+ "loss": 4.38,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8810572687224669,
49
+ "grad_norm": 25.133960723876953,
50
+ "learning_rate": 8.713450292397661e-06,
51
+ "loss": 4.3492,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 1.0,
56
+ "eval_accuracy": 0.3879003558718861,
57
+ "eval_loss": 1.0845022201538086,
58
+ "eval_macro_f1": 0.3415916490986263,
59
+ "eval_runtime": 1.2086,
60
+ "eval_samples_per_second": 465.006,
61
+ "eval_steps_per_second": 29.787,
62
+ "eval_weighted_f1": 0.3394971773692272,
63
+ "step": 171
64
+ },
65
+ {
66
+ "epoch": 1.0234948604992657,
67
+ "grad_norm": 36.66134262084961,
68
+ "learning_rate": 1.017543859649123e-05,
69
+ "loss": 4.2753,
70
+ "step": 175
71
+ },
72
+ {
73
+ "epoch": 1.170337738619677,
74
+ "grad_norm": 17.75954246520996,
75
+ "learning_rate": 1.1637426900584797e-05,
76
+ "loss": 4.2327,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.3171806167400881,
81
+ "grad_norm": 22.90311050415039,
82
+ "learning_rate": 1.3099415204678362e-05,
83
+ "loss": 4.2055,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.4640234948604993,
88
+ "grad_norm": 32.75823211669922,
89
+ "learning_rate": 1.4561403508771931e-05,
90
+ "loss": 4.331,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.6108663729809103,
95
+ "grad_norm": 25.388853073120117,
96
+ "learning_rate": 1.60233918128655e-05,
97
+ "loss": 4.3045,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.7577092511013217,
102
+ "grad_norm": 29.442508697509766,
103
+ "learning_rate": 1.7485380116959064e-05,
104
+ "loss": 4.3285,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.9045521292217327,
109
+ "grad_norm": 24.855016708374023,
110
+ "learning_rate": 1.894736842105263e-05,
111
+ "loss": 4.246,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 2.0,
116
+ "eval_accuracy": 0.40747330960854095,
117
+ "eval_loss": 1.1078580617904663,
118
+ "eval_macro_f1": 0.3537153700705074,
119
+ "eval_runtime": 1.1268,
120
+ "eval_samples_per_second": 498.777,
121
+ "eval_steps_per_second": 31.95,
122
+ "eval_weighted_f1": 0.3479225287782979,
123
+ "step": 342
124
+ },
125
+ {
126
+ "epoch": 2.0469897209985315,
127
+ "grad_norm": 22.83465003967285,
128
+ "learning_rate": 1.99545159194282e-05,
129
+ "loss": 3.9936,
130
+ "step": 350
131
+ },
132
+ {
133
+ "epoch": 2.193832599118943,
134
+ "grad_norm": 15.519044876098633,
135
+ "learning_rate": 1.9792072774528916e-05,
136
+ "loss": 3.9323,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.340675477239354,
141
+ "grad_norm": 26.67107391357422,
142
+ "learning_rate": 1.962962962962963e-05,
143
+ "loss": 3.7592,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.4875183553597653,
148
+ "grad_norm": 19.69229507446289,
149
+ "learning_rate": 1.9467186484730346e-05,
150
+ "loss": 3.783,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.6343612334801763,
155
+ "grad_norm": 24.198707580566406,
156
+ "learning_rate": 1.930474333983106e-05,
157
+ "loss": 3.8374,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.7812041116005872,
162
+ "grad_norm": 35.194175720214844,
163
+ "learning_rate": 1.9142300194931776e-05,
164
+ "loss": 3.7651,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.9280469897209986,
169
+ "grad_norm": 32.717124938964844,
170
+ "learning_rate": 1.897985705003249e-05,
171
+ "loss": 3.6372,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 3.0,
176
+ "eval_accuracy": 0.5498220640569395,
177
+ "eval_loss": 0.9409611821174622,
178
+ "eval_macro_f1": 0.5369368641707378,
179
+ "eval_runtime": 1.1652,
180
+ "eval_samples_per_second": 482.301,
181
+ "eval_steps_per_second": 30.895,
182
+ "eval_weighted_f1": 0.5375987962984153,
183
+ "step": 513
184
+ },
185
+ {
186
+ "epoch": 3.0704845814977975,
187
+ "grad_norm": 25.14092445373535,
188
+ "learning_rate": 1.8817413905133203e-05,
189
+ "loss": 3.1335,
190
+ "step": 525
191
+ },
192
+ {
193
+ "epoch": 3.2173274596182084,
194
+ "grad_norm": 33.47543716430664,
195
+ "learning_rate": 1.8654970760233918e-05,
196
+ "loss": 2.9605,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.36417033773862,
201
+ "grad_norm": 42.239418029785156,
202
+ "learning_rate": 1.8492527615334633e-05,
203
+ "loss": 3.0959,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.511013215859031,
208
+ "grad_norm": 27.37813377380371,
209
+ "learning_rate": 1.833008447043535e-05,
210
+ "loss": 2.9869,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.657856093979442,
215
+ "grad_norm": 30.336503982543945,
216
+ "learning_rate": 1.8167641325536064e-05,
217
+ "loss": 2.9882,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.804698972099853,
222
+ "grad_norm": 54.229408264160156,
223
+ "learning_rate": 1.800519818063678e-05,
224
+ "loss": 2.9932,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.951541850220264,
229
+ "grad_norm": 41.398597717285156,
230
+ "learning_rate": 1.7842755035737494e-05,
231
+ "loss": 2.7499,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 4.0,
236
+ "eval_accuracy": 0.5765124555160143,
237
+ "eval_loss": 0.9548276662826538,
238
+ "eval_macro_f1": 0.5745254022822105,
239
+ "eval_runtime": 1.2157,
240
+ "eval_samples_per_second": 462.274,
241
+ "eval_steps_per_second": 29.612,
242
+ "eval_weighted_f1": 0.5748016121055329,
243
+ "step": 684
244
+ },
245
+ {
246
+ "epoch": 4.093979441997063,
247
+ "grad_norm": 51.15410614013672,
248
+ "learning_rate": 1.768031189083821e-05,
249
+ "loss": 2.2844,
250
+ "step": 700
251
+ },
252
+ {
253
+ "epoch": 4.240822320117474,
254
+ "grad_norm": 47.037986755371094,
255
+ "learning_rate": 1.751786874593892e-05,
256
+ "loss": 1.936,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.387665198237886,
261
+ "grad_norm": 53.382972717285156,
262
+ "learning_rate": 1.7355425601039636e-05,
263
+ "loss": 1.827,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.534508076358296,
268
+ "grad_norm": 58.742332458496094,
269
+ "learning_rate": 1.719298245614035e-05,
270
+ "loss": 2.2023,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.681350954478708,
275
+ "grad_norm": 50.0244026184082,
276
+ "learning_rate": 1.7030539311241066e-05,
277
+ "loss": 1.8348,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.828193832599119,
282
+ "grad_norm": 73.3564682006836,
283
+ "learning_rate": 1.686809616634178e-05,
284
+ "loss": 1.8355,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.975036710719531,
289
+ "grad_norm": 64.41809844970703,
290
+ "learning_rate": 1.6705653021442497e-05,
291
+ "loss": 1.8025,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 5.0,
296
+ "eval_accuracy": 0.5533807829181495,
297
+ "eval_loss": 1.130232572555542,
298
+ "eval_macro_f1": 0.5544698179166704,
299
+ "eval_runtime": 1.1418,
300
+ "eval_samples_per_second": 492.203,
301
+ "eval_steps_per_second": 31.529,
302
+ "eval_weighted_f1": 0.5533888776357623,
303
+ "step": 855
304
+ },
305
+ {
306
+ "epoch": 5.117474302496329,
307
+ "grad_norm": 33.71281051635742,
308
+ "learning_rate": 1.654320987654321e-05,
309
+ "loss": 1.1478,
310
+ "step": 875
311
+ },
312
+ {
313
+ "epoch": 5.26431718061674,
314
+ "grad_norm": 60.219451904296875,
315
+ "learning_rate": 1.6380766731643927e-05,
316
+ "loss": 1.0621,
317
+ "step": 900
318
+ },
319
+ {
320
+ "epoch": 5.411160058737151,
321
+ "grad_norm": 69.2081069946289,
322
+ "learning_rate": 1.621832358674464e-05,
323
+ "loss": 1.1548,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.558002936857562,
328
+ "grad_norm": 43.02848815917969,
329
+ "learning_rate": 1.6055880441845354e-05,
330
+ "loss": 0.985,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.704845814977974,
335
+ "grad_norm": 24.841623306274414,
336
+ "learning_rate": 1.589343729694607e-05,
337
+ "loss": 1.079,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.851688693098385,
342
+ "grad_norm": 41.482872009277344,
343
+ "learning_rate": 1.5730994152046787e-05,
344
+ "loss": 1.1817,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.998531571218796,
349
+ "grad_norm": 25.20818328857422,
350
+ "learning_rate": 1.55685510071475e-05,
351
+ "loss": 1.047,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 6.0,
356
+ "eval_accuracy": 0.599644128113879,
357
+ "eval_loss": 1.3713058233261108,
358
+ "eval_macro_f1": 0.5893419847764664,
359
+ "eval_runtime": 1.2066,
360
+ "eval_samples_per_second": 465.773,
361
+ "eval_steps_per_second": 29.836,
362
+ "eval_weighted_f1": 0.590349818049014,
363
+ "step": 1026
364
+ },
365
+ {
366
+ "epoch": 6.140969162995595,
367
+ "grad_norm": 16.632728576660156,
368
+ "learning_rate": 1.5406107862248214e-05,
369
+ "loss": 0.3892,
370
+ "step": 1050
371
+ },
372
+ {
373
+ "epoch": 6.2878120411160054,
374
+ "grad_norm": 39.91593933105469,
375
+ "learning_rate": 1.524366471734893e-05,
376
+ "loss": 0.4401,
377
+ "step": 1075
378
+ },
379
+ {
380
+ "epoch": 6.434654919236417,
381
+ "grad_norm": 91.06832122802734,
382
+ "learning_rate": 1.5081221572449645e-05,
383
+ "loss": 0.7126,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.581497797356828,
388
+ "grad_norm": 73.668701171875,
389
+ "learning_rate": 1.4918778427550358e-05,
390
+ "loss": 0.5194,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.72834067547724,
395
+ "grad_norm": 34.290042877197266,
396
+ "learning_rate": 1.4756335282651073e-05,
397
+ "loss": 0.5552,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.87518355359765,
402
+ "grad_norm": 60.50444412231445,
403
+ "learning_rate": 1.4593892137751788e-05,
404
+ "loss": 0.5899,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 7.0,
409
+ "eval_accuracy": 0.5693950177935944,
410
+ "eval_loss": 1.885563611984253,
411
+ "eval_macro_f1": 0.5567535262818194,
412
+ "eval_runtime": 1.227,
413
+ "eval_samples_per_second": 458.015,
414
+ "eval_steps_per_second": 29.339,
415
+ "eval_weighted_f1": 0.5574885386951494,
416
+ "step": 1197
417
+ },
418
+ {
419
+ "epoch": 7.0176211453744495,
420
+ "grad_norm": 62.00462341308594,
421
+ "learning_rate": 1.4431448992852503e-05,
422
+ "loss": 0.6367,
423
+ "step": 1200
424
+ },
425
+ {
426
+ "epoch": 7.164464023494861,
427
+ "grad_norm": 22.830387115478516,
428
+ "learning_rate": 1.4269005847953217e-05,
429
+ "loss": 0.2574,
430
+ "step": 1225
431
+ },
432
+ {
433
+ "epoch": 7.311306901615271,
434
+ "grad_norm": 114.9410400390625,
435
+ "learning_rate": 1.4106562703053932e-05,
436
+ "loss": 0.3086,
437
+ "step": 1250
438
+ },
439
+ {
440
+ "epoch": 7.458149779735683,
441
+ "grad_norm": 49.340572357177734,
442
+ "learning_rate": 1.3944119558154647e-05,
443
+ "loss": 0.4897,
444
+ "step": 1275
445
+ },
446
+ {
447
+ "epoch": 7.604992657856094,
448
+ "grad_norm": 12.92598819732666,
449
+ "learning_rate": 1.3781676413255362e-05,
450
+ "loss": 0.3984,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.751835535976505,
455
+ "grad_norm": 65.94971466064453,
456
+ "learning_rate": 1.3619233268356076e-05,
457
+ "loss": 0.2408,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.898678414096916,
462
+ "grad_norm": 1.6258292198181152,
463
+ "learning_rate": 1.345679012345679e-05,
464
+ "loss": 0.2501,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 8.0,
469
+ "eval_accuracy": 0.5765124555160143,
470
+ "eval_loss": 2.364772081375122,
471
+ "eval_macro_f1": 0.5777111108678011,
472
+ "eval_runtime": 1.1688,
473
+ "eval_samples_per_second": 480.816,
474
+ "eval_steps_per_second": 30.8,
475
+ "eval_weighted_f1": 0.5776290297134498,
476
+ "step": 1368
477
+ },
478
+ {
479
+ "epoch": 8.041116005873715,
480
+ "grad_norm": 17.149738311767578,
481
+ "learning_rate": 1.3294346978557506e-05,
482
+ "loss": 0.2897,
483
+ "step": 1375
484
+ },
485
+ {
486
+ "epoch": 8.187958883994126,
487
+ "grad_norm": 7.26170539855957,
488
+ "learning_rate": 1.3131903833658221e-05,
489
+ "loss": 0.2377,
490
+ "step": 1400
491
+ },
492
+ {
493
+ "epoch": 8.334801762114537,
494
+ "grad_norm": 24.6292781829834,
495
+ "learning_rate": 1.2969460688758935e-05,
496
+ "loss": 0.3229,
497
+ "step": 1425
498
+ },
499
+ {
500
+ "epoch": 8.481644640234949,
501
+ "grad_norm": 64.48216247558594,
502
+ "learning_rate": 1.280701754385965e-05,
503
+ "loss": 0.1937,
504
+ "step": 1450
505
+ },
506
+ {
507
+ "epoch": 8.62848751835536,
508
+ "grad_norm": 9.738794326782227,
509
+ "learning_rate": 1.2644574398960365e-05,
510
+ "loss": 0.3145,
511
+ "step": 1475
512
+ },
513
+ {
514
+ "epoch": 8.775330396475772,
515
+ "grad_norm": 38.02442169189453,
516
+ "learning_rate": 1.248213125406108e-05,
517
+ "loss": 0.3057,
518
+ "step": 1500
519
+ },
520
+ {
521
+ "epoch": 8.922173274596181,
522
+ "grad_norm": 20.25743865966797,
523
+ "learning_rate": 1.2319688109161793e-05,
524
+ "loss": 0.2359,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 9.0,
529
+ "eval_accuracy": 0.5551601423487544,
530
+ "eval_loss": 2.9617931842803955,
531
+ "eval_macro_f1": 0.5296641282034465,
532
+ "eval_runtime": 1.153,
533
+ "eval_samples_per_second": 487.442,
534
+ "eval_steps_per_second": 31.224,
535
+ "eval_weighted_f1": 0.5315644001451645,
536
+ "step": 1539
537
+ }
538
+ ],
539
+ "logging_steps": 25,
540
+ "max_steps": 3420,
541
+ "num_input_tokens_seen": 0,
542
+ "num_train_epochs": 20,
543
+ "save_steps": 500,
544
+ "stateful_callbacks": {
545
+ "EarlyStoppingCallback": {
546
+ "args": {
547
+ "early_stopping_patience": 3,
548
+ "early_stopping_threshold": 0.0
549
+ },
550
+ "attributes": {
551
+ "early_stopping_patience_counter": 3
552
+ }
553
+ },
554
+ "TrainerControl": {
555
+ "args": {
556
+ "should_epoch_stop": false,
557
+ "should_evaluate": false,
558
+ "should_log": false,
559
+ "should_save": true,
560
+ "should_training_stop": true
561
+ },
562
+ "attributes": {}
563
+ }
564
+ },
565
+ "total_flos": 0.0,
566
+ "train_batch_size": 8,
567
+ "trial_name": null,
568
+ "trial_params": null
569
+ }
ood-vifactcheck-visobert-source-source-02-s42-20260830T234935Z-f21e2f7e/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-s42-20260830T234935Z-f21e2f7e/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-s42-20260830T234935Z-f21e2f7e/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-s42-20260830T234935Z-f21e2f7e/trainer/checkpoint-1267/trainer_state.json ADDED
@@ -0,0 +1,470 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 724,
3
+ "best_metric": 0.5654145108458176,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-02-s42-20260830T234935Z-f21e2f7e/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": 30.525136947631836,
15
+ "learning_rate": 1.3259668508287293e-06,
16
+ "loss": 4.7225,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27624309392265195,
21
+ "grad_norm": 25.079261779785156,
22
+ "learning_rate": 2.707182320441989e-06,
23
+ "loss": 4.5068,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.4143646408839779,
28
+ "grad_norm": 31.723726272583008,
29
+ "learning_rate": 4.088397790055249e-06,
30
+ "loss": 4.4644,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5524861878453039,
35
+ "grad_norm": 22.183345794677734,
36
+ "learning_rate": 5.469613259668509e-06,
37
+ "loss": 4.4343,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6906077348066298,
42
+ "grad_norm": 18.998119354248047,
43
+ "learning_rate": 6.850828729281769e-06,
44
+ "loss": 4.3637,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8287292817679558,
49
+ "grad_norm": 25.2176456451416,
50
+ "learning_rate": 8.232044198895029e-06,
51
+ "loss": 4.4254,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9668508287292817,
56
+ "grad_norm": 26.90962791442871,
57
+ "learning_rate": 9.613259668508288e-06,
58
+ "loss": 4.3531,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.43127147766323026,
64
+ "eval_loss": 1.067191481590271,
65
+ "eval_macro_f1": 0.4109970044083064,
66
+ "eval_runtime": 1.2669,
67
+ "eval_samples_per_second": 459.379,
68
+ "eval_steps_per_second": 29.204,
69
+ "eval_weighted_f1": 0.4129703285872264,
70
+ "step": 181
71
+ },
72
+ {
73
+ "epoch": 1.1049723756906078,
74
+ "grad_norm": 19.38654136657715,
75
+ "learning_rate": 1.0994475138121547e-05,
76
+ "loss": 4.3158,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2430939226519337,
81
+ "grad_norm": 55.797576904296875,
82
+ "learning_rate": 1.2375690607734806e-05,
83
+ "loss": 4.1931,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3812154696132597,
88
+ "grad_norm": 21.301435470581055,
89
+ "learning_rate": 1.3756906077348067e-05,
90
+ "loss": 4.2454,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.5193370165745856,
95
+ "grad_norm": 37.112369537353516,
96
+ "learning_rate": 1.5138121546961326e-05,
97
+ "loss": 4.2336,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6574585635359116,
102
+ "grad_norm": 25.38357162475586,
103
+ "learning_rate": 1.6519337016574586e-05,
104
+ "loss": 4.2979,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7955801104972375,
109
+ "grad_norm": 31.941804885864258,
110
+ "learning_rate": 1.7900552486187846e-05,
111
+ "loss": 4.2452,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.9337016574585635,
116
+ "grad_norm": 24.728038787841797,
117
+ "learning_rate": 1.9281767955801107e-05,
118
+ "loss": 4.1803,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.48625429553264604,
124
+ "eval_loss": 1.0152088403701782,
125
+ "eval_macro_f1": 0.4541236579526073,
126
+ "eval_runtime": 1.1953,
127
+ "eval_samples_per_second": 486.906,
128
+ "eval_steps_per_second": 30.954,
129
+ "eval_weighted_f1": 0.4591091168138351,
130
+ "step": 362
131
+ },
132
+ {
133
+ "epoch": 2.0718232044198897,
134
+ "grad_norm": 22.7708683013916,
135
+ "learning_rate": 1.9926335174953962e-05,
136
+ "loss": 3.9724,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.2099447513812156,
141
+ "grad_norm": 30.681182861328125,
142
+ "learning_rate": 1.9772866789441377e-05,
143
+ "loss": 3.6938,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.3480662983425415,
148
+ "grad_norm": 44.32679748535156,
149
+ "learning_rate": 1.9619398403928794e-05,
150
+ "loss": 3.7447,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.4861878453038675,
155
+ "grad_norm": 40.52362060546875,
156
+ "learning_rate": 1.946593001841621e-05,
157
+ "loss": 3.6983,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.6243093922651934,
162
+ "grad_norm": 32.07386016845703,
163
+ "learning_rate": 1.9312461632903623e-05,
164
+ "loss": 3.6036,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.7624309392265194,
169
+ "grad_norm": 30.888090133666992,
170
+ "learning_rate": 1.915899324739104e-05,
171
+ "loss": 3.7066,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.9005524861878453,
176
+ "grad_norm": 38.65287780761719,
177
+ "learning_rate": 1.9005524861878455e-05,
178
+ "loss": 3.7704,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5412371134020618,
184
+ "eval_loss": 1.0115411281585693,
185
+ "eval_macro_f1": 0.5091485188725755,
186
+ "eval_runtime": 1.2688,
187
+ "eval_samples_per_second": 458.708,
188
+ "eval_steps_per_second": 29.162,
189
+ "eval_weighted_f1": 0.511444137631614,
190
+ "step": 543
191
+ },
192
+ {
193
+ "epoch": 3.0386740331491713,
194
+ "grad_norm": 19.43329429626465,
195
+ "learning_rate": 1.885205647636587e-05,
196
+ "loss": 3.4979,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.1767955801104972,
201
+ "grad_norm": 60.57960891723633,
202
+ "learning_rate": 1.8698588090853287e-05,
203
+ "loss": 2.9065,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.314917127071823,
208
+ "grad_norm": 27.77372169494629,
209
+ "learning_rate": 1.85451197053407e-05,
210
+ "loss": 3.0349,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.453038674033149,
215
+ "grad_norm": 33.49660110473633,
216
+ "learning_rate": 1.8391651319828116e-05,
217
+ "loss": 3.1085,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.591160220994475,
222
+ "grad_norm": 48.54967498779297,
223
+ "learning_rate": 1.823818293431553e-05,
224
+ "loss": 2.9585,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.729281767955801,
229
+ "grad_norm": 35.72766876220703,
230
+ "learning_rate": 1.8084714548802948e-05,
231
+ "loss": 2.6746,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.867403314917127,
236
+ "grad_norm": 36.62098693847656,
237
+ "learning_rate": 1.7931246163290362e-05,
238
+ "loss": 2.7954,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 4.0,
243
+ "eval_accuracy": 0.5618556701030928,
244
+ "eval_loss": 0.9885299801826477,
245
+ "eval_macro_f1": 0.5654145108458176,
246
+ "eval_runtime": 1.2054,
247
+ "eval_samples_per_second": 482.813,
248
+ "eval_steps_per_second": 30.694,
249
+ "eval_weighted_f1": 0.5656543717421049,
250
+ "step": 724
251
+ },
252
+ {
253
+ "epoch": 4.005524861878453,
254
+ "grad_norm": 37.63091278076172,
255
+ "learning_rate": 1.7777777777777777e-05,
256
+ "loss": 2.7135,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.143646408839779,
261
+ "grad_norm": 50.06581497192383,
262
+ "learning_rate": 1.7624309392265194e-05,
263
+ "loss": 1.8905,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.281767955801105,
268
+ "grad_norm": 47.583831787109375,
269
+ "learning_rate": 1.747084100675261e-05,
270
+ "loss": 2.0541,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.419889502762431,
275
+ "grad_norm": 50.20119094848633,
276
+ "learning_rate": 1.7317372621240026e-05,
277
+ "loss": 1.9488,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.558011049723757,
282
+ "grad_norm": 33.835018157958984,
283
+ "learning_rate": 1.716390423572744e-05,
284
+ "loss": 1.9005,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.696132596685083,
289
+ "grad_norm": 79.6776123046875,
290
+ "learning_rate": 1.701043585021486e-05,
291
+ "loss": 2.0593,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.834254143646409,
296
+ "grad_norm": 62.302852630615234,
297
+ "learning_rate": 1.6856967464702273e-05,
298
+ "loss": 1.9157,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.972375690607735,
303
+ "grad_norm": 48.55406188964844,
304
+ "learning_rate": 1.670349907918969e-05,
305
+ "loss": 2.132,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 5.0,
310
+ "eval_accuracy": 0.5687285223367697,
311
+ "eval_loss": 1.1618019342422485,
312
+ "eval_macro_f1": 0.5545546337187448,
313
+ "eval_runtime": 1.2231,
314
+ "eval_samples_per_second": 475.848,
315
+ "eval_steps_per_second": 30.252,
316
+ "eval_weighted_f1": 0.5550869330941071,
317
+ "step": 905
318
+ },
319
+ {
320
+ "epoch": 5.110497237569061,
321
+ "grad_norm": 67.3942642211914,
322
+ "learning_rate": 1.6550030693677105e-05,
323
+ "loss": 1.4671,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.248618784530387,
328
+ "grad_norm": 67.89849090576172,
329
+ "learning_rate": 1.639656230816452e-05,
330
+ "loss": 1.1435,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.386740331491713,
335
+ "grad_norm": 73.80731201171875,
336
+ "learning_rate": 1.6243093922651937e-05,
337
+ "loss": 1.1941,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.524861878453039,
342
+ "grad_norm": 91.21694946289062,
343
+ "learning_rate": 1.608962553713935e-05,
344
+ "loss": 1.0504,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.662983425414365,
349
+ "grad_norm": 93.40288543701172,
350
+ "learning_rate": 1.5936157151626766e-05,
351
+ "loss": 1.2649,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.801104972375691,
356
+ "grad_norm": 54.983489990234375,
357
+ "learning_rate": 1.5782688766114183e-05,
358
+ "loss": 1.2724,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.939226519337017,
363
+ "grad_norm": 61.70564651489258,
364
+ "learning_rate": 1.5629220380601598e-05,
365
+ "loss": 1.2303,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 6.0,
370
+ "eval_accuracy": 0.5343642611683849,
371
+ "eval_loss": 1.7139558792114258,
372
+ "eval_macro_f1": 0.5136235063395753,
373
+ "eval_runtime": 1.1899,
374
+ "eval_samples_per_second": 489.116,
375
+ "eval_steps_per_second": 31.095,
376
+ "eval_weighted_f1": 0.5106939315078742,
377
+ "step": 1086
378
+ },
379
+ {
380
+ "epoch": 6.077348066298343,
381
+ "grad_norm": 60.249366760253906,
382
+ "learning_rate": 1.5475751995089012e-05,
383
+ "loss": 0.9093,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.2154696132596685,
388
+ "grad_norm": 61.90557861328125,
389
+ "learning_rate": 1.532228360957643e-05,
390
+ "loss": 0.5771,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.3535911602209945,
395
+ "grad_norm": 61.39337158203125,
396
+ "learning_rate": 1.5168815224063844e-05,
397
+ "loss": 0.5626,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.49171270718232,
402
+ "grad_norm": 165.74009704589844,
403
+ "learning_rate": 1.5015346838551258e-05,
404
+ "loss": 0.5844,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.629834254143646,
409
+ "grad_norm": 43.92707824707031,
410
+ "learning_rate": 1.4861878453038676e-05,
411
+ "loss": 0.7747,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.767955801104972,
416
+ "grad_norm": 28.712112426757812,
417
+ "learning_rate": 1.470841006752609e-05,
418
+ "loss": 0.5239,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.906077348066298,
423
+ "grad_norm": 58.43913650512695,
424
+ "learning_rate": 1.4554941682013506e-05,
425
+ "loss": 0.5811,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 7.0,
430
+ "eval_accuracy": 0.5378006872852233,
431
+ "eval_loss": 2.1551144123077393,
432
+ "eval_macro_f1": 0.5242238338773132,
433
+ "eval_runtime": 1.2194,
434
+ "eval_samples_per_second": 477.266,
435
+ "eval_steps_per_second": 30.342,
436
+ "eval_weighted_f1": 0.5231536776123542,
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-s42-20260830T234935Z-f21e2f7e/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-s42-20260830T234935Z-f21e2f7e/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-s42-20260830T234935Z-f21e2f7e/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-s42-20260830T234935Z-f21e2f7e/trainer/checkpoint-724/trainer_state.json ADDED
@@ -0,0 +1,283 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 724,
3
+ "best_metric": 0.5654145108458176,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-02-s42-20260830T234935Z-f21e2f7e/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": 30.525136947631836,
15
+ "learning_rate": 1.3259668508287293e-06,
16
+ "loss": 4.7225,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27624309392265195,
21
+ "grad_norm": 25.079261779785156,
22
+ "learning_rate": 2.707182320441989e-06,
23
+ "loss": 4.5068,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.4143646408839779,
28
+ "grad_norm": 31.723726272583008,
29
+ "learning_rate": 4.088397790055249e-06,
30
+ "loss": 4.4644,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5524861878453039,
35
+ "grad_norm": 22.183345794677734,
36
+ "learning_rate": 5.469613259668509e-06,
37
+ "loss": 4.4343,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6906077348066298,
42
+ "grad_norm": 18.998119354248047,
43
+ "learning_rate": 6.850828729281769e-06,
44
+ "loss": 4.3637,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8287292817679558,
49
+ "grad_norm": 25.2176456451416,
50
+ "learning_rate": 8.232044198895029e-06,
51
+ "loss": 4.4254,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9668508287292817,
56
+ "grad_norm": 26.90962791442871,
57
+ "learning_rate": 9.613259668508288e-06,
58
+ "loss": 4.3531,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.43127147766323026,
64
+ "eval_loss": 1.067191481590271,
65
+ "eval_macro_f1": 0.4109970044083064,
66
+ "eval_runtime": 1.2669,
67
+ "eval_samples_per_second": 459.379,
68
+ "eval_steps_per_second": 29.204,
69
+ "eval_weighted_f1": 0.4129703285872264,
70
+ "step": 181
71
+ },
72
+ {
73
+ "epoch": 1.1049723756906078,
74
+ "grad_norm": 19.38654136657715,
75
+ "learning_rate": 1.0994475138121547e-05,
76
+ "loss": 4.3158,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2430939226519337,
81
+ "grad_norm": 55.797576904296875,
82
+ "learning_rate": 1.2375690607734806e-05,
83
+ "loss": 4.1931,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3812154696132597,
88
+ "grad_norm": 21.301435470581055,
89
+ "learning_rate": 1.3756906077348067e-05,
90
+ "loss": 4.2454,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.5193370165745856,
95
+ "grad_norm": 37.112369537353516,
96
+ "learning_rate": 1.5138121546961326e-05,
97
+ "loss": 4.2336,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6574585635359116,
102
+ "grad_norm": 25.38357162475586,
103
+ "learning_rate": 1.6519337016574586e-05,
104
+ "loss": 4.2979,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7955801104972375,
109
+ "grad_norm": 31.941804885864258,
110
+ "learning_rate": 1.7900552486187846e-05,
111
+ "loss": 4.2452,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.9337016574585635,
116
+ "grad_norm": 24.728038787841797,
117
+ "learning_rate": 1.9281767955801107e-05,
118
+ "loss": 4.1803,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.48625429553264604,
124
+ "eval_loss": 1.0152088403701782,
125
+ "eval_macro_f1": 0.4541236579526073,
126
+ "eval_runtime": 1.1953,
127
+ "eval_samples_per_second": 486.906,
128
+ "eval_steps_per_second": 30.954,
129
+ "eval_weighted_f1": 0.4591091168138351,
130
+ "step": 362
131
+ },
132
+ {
133
+ "epoch": 2.0718232044198897,
134
+ "grad_norm": 22.7708683013916,
135
+ "learning_rate": 1.9926335174953962e-05,
136
+ "loss": 3.9724,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.2099447513812156,
141
+ "grad_norm": 30.681182861328125,
142
+ "learning_rate": 1.9772866789441377e-05,
143
+ "loss": 3.6938,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.3480662983425415,
148
+ "grad_norm": 44.32679748535156,
149
+ "learning_rate": 1.9619398403928794e-05,
150
+ "loss": 3.7447,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.4861878453038675,
155
+ "grad_norm": 40.52362060546875,
156
+ "learning_rate": 1.946593001841621e-05,
157
+ "loss": 3.6983,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.6243093922651934,
162
+ "grad_norm": 32.07386016845703,
163
+ "learning_rate": 1.9312461632903623e-05,
164
+ "loss": 3.6036,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.7624309392265194,
169
+ "grad_norm": 30.888090133666992,
170
+ "learning_rate": 1.915899324739104e-05,
171
+ "loss": 3.7066,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.9005524861878453,
176
+ "grad_norm": 38.65287780761719,
177
+ "learning_rate": 1.9005524861878455e-05,
178
+ "loss": 3.7704,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5412371134020618,
184
+ "eval_loss": 1.0115411281585693,
185
+ "eval_macro_f1": 0.5091485188725755,
186
+ "eval_runtime": 1.2688,
187
+ "eval_samples_per_second": 458.708,
188
+ "eval_steps_per_second": 29.162,
189
+ "eval_weighted_f1": 0.511444137631614,
190
+ "step": 543
191
+ },
192
+ {
193
+ "epoch": 3.0386740331491713,
194
+ "grad_norm": 19.43329429626465,
195
+ "learning_rate": 1.885205647636587e-05,
196
+ "loss": 3.4979,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.1767955801104972,
201
+ "grad_norm": 60.57960891723633,
202
+ "learning_rate": 1.8698588090853287e-05,
203
+ "loss": 2.9065,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.314917127071823,
208
+ "grad_norm": 27.77372169494629,
209
+ "learning_rate": 1.85451197053407e-05,
210
+ "loss": 3.0349,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.453038674033149,
215
+ "grad_norm": 33.49660110473633,
216
+ "learning_rate": 1.8391651319828116e-05,
217
+ "loss": 3.1085,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.591160220994475,
222
+ "grad_norm": 48.54967498779297,
223
+ "learning_rate": 1.823818293431553e-05,
224
+ "loss": 2.9585,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.729281767955801,
229
+ "grad_norm": 35.72766876220703,
230
+ "learning_rate": 1.8084714548802948e-05,
231
+ "loss": 2.6746,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.867403314917127,
236
+ "grad_norm": 36.62098693847656,
237
+ "learning_rate": 1.7931246163290362e-05,
238
+ "loss": 2.7954,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 4.0,
243
+ "eval_accuracy": 0.5618556701030928,
244
+ "eval_loss": 0.9885299801826477,
245
+ "eval_macro_f1": 0.5654145108458176,
246
+ "eval_runtime": 1.2054,
247
+ "eval_samples_per_second": 482.813,
248
+ "eval_steps_per_second": 30.694,
249
+ "eval_weighted_f1": 0.5656543717421049,
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-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2013/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-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2013/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2013/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-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2013/trainer_state.json ADDED
@@ -0,0 +1,724 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2013,
3
+ "best_metric": 0.5850014748355203,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2013",
5
+ "epoch": 11.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2013,
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": 35.02544403076172,
15
+ "learning_rate": 1.3114754098360657e-06,
16
+ "loss": 4.6872,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27434842249657065,
21
+ "grad_norm": 21.24454689025879,
22
+ "learning_rate": 2.677595628415301e-06,
23
+ "loss": 4.5419,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.411522633744856,
28
+ "grad_norm": 20.140762329101562,
29
+ "learning_rate": 4.043715846994536e-06,
30
+ "loss": 4.5041,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5486968449931413,
35
+ "grad_norm": 36.08255386352539,
36
+ "learning_rate": 5.409836065573772e-06,
37
+ "loss": 4.4129,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6858710562414266,
42
+ "grad_norm": 17.723480224609375,
43
+ "learning_rate": 6.775956284153006e-06,
44
+ "loss": 4.4218,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.823045267489712,
49
+ "grad_norm": 30.218822479248047,
50
+ "learning_rate": 8.14207650273224e-06,
51
+ "loss": 4.394,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9602194787379973,
56
+ "grad_norm": 36.463706970214844,
57
+ "learning_rate": 9.508196721311476e-06,
58
+ "loss": 4.3698,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.41098169717138106,
64
+ "eval_loss": 1.0792137384414673,
65
+ "eval_macro_f1": 0.4061852484724344,
66
+ "eval_runtime": 1.1904,
67
+ "eval_samples_per_second": 504.879,
68
+ "eval_steps_per_second": 31.922,
69
+ "eval_weighted_f1": 0.40612451900351487,
70
+ "step": 183
71
+ },
72
+ {
73
+ "epoch": 1.093278463648834,
74
+ "grad_norm": 28.814289093017578,
75
+ "learning_rate": 1.0874316939890712e-05,
76
+ "loss": 4.1889,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2304526748971194,
81
+ "grad_norm": 16.32625961303711,
82
+ "learning_rate": 1.2240437158469946e-05,
83
+ "loss": 4.2271,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3676268861454046,
88
+ "grad_norm": 25.90184783935547,
89
+ "learning_rate": 1.3606557377049181e-05,
90
+ "loss": 4.2057,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.50480109739369,
95
+ "grad_norm": 22.540451049804688,
96
+ "learning_rate": 1.4972677595628417e-05,
97
+ "loss": 4.251,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6419753086419753,
102
+ "grad_norm": 24.28812026977539,
103
+ "learning_rate": 1.633879781420765e-05,
104
+ "loss": 4.3731,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7791495198902605,
109
+ "grad_norm": 39.98865509033203,
110
+ "learning_rate": 1.7704918032786887e-05,
111
+ "loss": 4.2678,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.916323731138546,
116
+ "grad_norm": 29.907140731811523,
117
+ "learning_rate": 1.9071038251366122e-05,
118
+ "loss": 4.1656,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.49916805324459235,
124
+ "eval_loss": 1.0238842964172363,
125
+ "eval_macro_f1": 0.484835268639423,
126
+ "eval_runtime": 1.1675,
127
+ "eval_samples_per_second": 514.778,
128
+ "eval_steps_per_second": 32.548,
129
+ "eval_weighted_f1": 0.486985131271952,
130
+ "step": 366
131
+ },
132
+ {
133
+ "epoch": 2.049382716049383,
134
+ "grad_norm": 23.380916595458984,
135
+ "learning_rate": 1.995142683667274e-05,
136
+ "loss": 3.9306,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.186556927297668,
141
+ "grad_norm": 39.58391571044922,
142
+ "learning_rate": 1.9799635701275045e-05,
143
+ "loss": 3.6788,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.3237311385459534,
148
+ "grad_norm": 20.238113403320312,
149
+ "learning_rate": 1.9647844565877354e-05,
150
+ "loss": 3.7912,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.460905349794239,
155
+ "grad_norm": 29.342870712280273,
156
+ "learning_rate": 1.9496053430479663e-05,
157
+ "loss": 3.8442,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.598079561042524,
162
+ "grad_norm": 55.8082389831543,
163
+ "learning_rate": 1.934426229508197e-05,
164
+ "loss": 3.711,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.7352537722908092,
169
+ "grad_norm": 40.55704879760742,
170
+ "learning_rate": 1.9192471159684274e-05,
171
+ "loss": 3.7917,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.8724279835390947,
176
+ "grad_norm": 34.73087692260742,
177
+ "learning_rate": 1.9040680024286583e-05,
178
+ "loss": 3.6319,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5391014975041597,
184
+ "eval_loss": 0.9485401511192322,
185
+ "eval_macro_f1": 0.5393116064326674,
186
+ "eval_runtime": 1.307,
187
+ "eval_samples_per_second": 459.841,
188
+ "eval_steps_per_second": 29.075,
189
+ "eval_weighted_f1": 0.538541502679688,
190
+ "step": 549
191
+ },
192
+ {
193
+ "epoch": 3.0054869684499312,
194
+ "grad_norm": 24.43846893310547,
195
+ "learning_rate": 1.888888888888889e-05,
196
+ "loss": 3.5135,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.1426611796982167,
201
+ "grad_norm": 56.53961181640625,
202
+ "learning_rate": 1.8737097753491198e-05,
203
+ "loss": 2.9245,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.279835390946502,
208
+ "grad_norm": 37.9984016418457,
209
+ "learning_rate": 1.8585306618093504e-05,
210
+ "loss": 2.8603,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.4170096021947876,
215
+ "grad_norm": 26.55724334716797,
216
+ "learning_rate": 1.8433515482695813e-05,
217
+ "loss": 2.9073,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.5541838134430725,
222
+ "grad_norm": 48.19314193725586,
223
+ "learning_rate": 1.828172434729812e-05,
224
+ "loss": 2.7598,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.691358024691358,
229
+ "grad_norm": 35.43125915527344,
230
+ "learning_rate": 1.8129933211900424e-05,
231
+ "loss": 2.8162,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.8285322359396434,
236
+ "grad_norm": 41.655555725097656,
237
+ "learning_rate": 1.7978142076502733e-05,
238
+ "loss": 2.8093,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.9657064471879284,
243
+ "grad_norm": 54.582244873046875,
244
+ "learning_rate": 1.7826350941105042e-05,
245
+ "loss": 2.9492,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 4.0,
250
+ "eval_accuracy": 0.5424292845257903,
251
+ "eval_loss": 1.0065358877182007,
252
+ "eval_macro_f1": 0.5468183232889116,
253
+ "eval_runtime": 1.2725,
254
+ "eval_samples_per_second": 472.305,
255
+ "eval_steps_per_second": 29.863,
256
+ "eval_weighted_f1": 0.5465966591345861,
257
+ "step": 732
258
+ },
259
+ {
260
+ "epoch": 4.098765432098766,
261
+ "grad_norm": 27.443614959716797,
262
+ "learning_rate": 1.7674559805707348e-05,
263
+ "loss": 2.2806,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.23593964334705,
268
+ "grad_norm": 35.80004119873047,
269
+ "learning_rate": 1.7522768670309657e-05,
270
+ "loss": 2.0458,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.373113854595336,
275
+ "grad_norm": 33.068687438964844,
276
+ "learning_rate": 1.7370977534911962e-05,
277
+ "loss": 1.9547,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.510288065843621,
282
+ "grad_norm": 52.475746154785156,
283
+ "learning_rate": 1.7219186399514268e-05,
284
+ "loss": 2.0498,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.647462277091907,
289
+ "grad_norm": 36.4451789855957,
290
+ "learning_rate": 1.7067395264116577e-05,
291
+ "loss": 1.9751,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.784636488340192,
296
+ "grad_norm": 29.826025009155273,
297
+ "learning_rate": 1.6915604128718883e-05,
298
+ "loss": 1.7888,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.921810699588478,
303
+ "grad_norm": 36.69779586791992,
304
+ "learning_rate": 1.6763812993321192e-05,
305
+ "loss": 1.8333,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 5.0,
310
+ "eval_accuracy": 0.5490848585690515,
311
+ "eval_loss": 1.2077672481536865,
312
+ "eval_macro_f1": 0.5414092808818651,
313
+ "eval_runtime": 1.3947,
314
+ "eval_samples_per_second": 430.931,
315
+ "eval_steps_per_second": 27.247,
316
+ "eval_weighted_f1": 0.5383178751053753,
317
+ "step": 915
318
+ },
319
+ {
320
+ "epoch": 5.054869684499314,
321
+ "grad_norm": 68.85910034179688,
322
+ "learning_rate": 1.66120218579235e-05,
323
+ "loss": 1.4793,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.1920438957476,
328
+ "grad_norm": 85.61286163330078,
329
+ "learning_rate": 1.6460230722525807e-05,
330
+ "loss": 1.252,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.329218106995885,
335
+ "grad_norm": 56.671077728271484,
336
+ "learning_rate": 1.6308439587128112e-05,
337
+ "loss": 1.35,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.4663923182441705,
342
+ "grad_norm": 62.90438461303711,
343
+ "learning_rate": 1.615664845173042e-05,
344
+ "loss": 1.1463,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.603566529492455,
349
+ "grad_norm": 93.8409423828125,
350
+ "learning_rate": 1.6004857316332727e-05,
351
+ "loss": 1.003,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.7407407407407405,
356
+ "grad_norm": 36.048099517822266,
357
+ "learning_rate": 1.5853066180935036e-05,
358
+ "loss": 1.2405,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.877914951989026,
363
+ "grad_norm": 84.30831909179688,
364
+ "learning_rate": 1.570127504553734e-05,
365
+ "loss": 1.0601,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 6.0,
370
+ "eval_accuracy": 0.5391014975041597,
371
+ "eval_loss": 1.6696314811706543,
372
+ "eval_macro_f1": 0.5426858378219005,
373
+ "eval_runtime": 1.23,
374
+ "eval_samples_per_second": 488.626,
375
+ "eval_steps_per_second": 30.895,
376
+ "eval_weighted_f1": 0.5424897563415262,
377
+ "step": 1098
378
+ },
379
+ {
380
+ "epoch": 6.0109739368998625,
381
+ "grad_norm": 26.06507110595703,
382
+ "learning_rate": 1.5549483910139647e-05,
383
+ "loss": 0.9767,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.148148148148148,
388
+ "grad_norm": 46.933868408203125,
389
+ "learning_rate": 1.5397692774741956e-05,
390
+ "loss": 0.7216,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.285322359396433,
395
+ "grad_norm": 51.907161712646484,
396
+ "learning_rate": 1.5245901639344264e-05,
397
+ "loss": 0.6712,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.422496570644719,
402
+ "grad_norm": 29.579395294189453,
403
+ "learning_rate": 1.5094110503946571e-05,
404
+ "loss": 0.7167,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.559670781893004,
409
+ "grad_norm": 19.197471618652344,
410
+ "learning_rate": 1.4942319368548878e-05,
411
+ "loss": 0.7279,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.69684499314129,
416
+ "grad_norm": 45.326900482177734,
417
+ "learning_rate": 1.4790528233151184e-05,
418
+ "loss": 0.7464,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.834019204389575,
423
+ "grad_norm": 59.55309295654297,
424
+ "learning_rate": 1.4638737097753493e-05,
425
+ "loss": 0.5537,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 6.97119341563786,
430
+ "grad_norm": 74.94731140136719,
431
+ "learning_rate": 1.44869459623558e-05,
432
+ "loss": 0.6839,
433
+ "step": 1275
434
+ },
435
+ {
436
+ "epoch": 7.0,
437
+ "eval_accuracy": 0.5657237936772047,
438
+ "eval_loss": 2.0491621494293213,
439
+ "eval_macro_f1": 0.5564754364524109,
440
+ "eval_runtime": 1.1854,
441
+ "eval_samples_per_second": 507.023,
442
+ "eval_steps_per_second": 32.058,
443
+ "eval_weighted_f1": 0.5552805344381514,
444
+ "step": 1281
445
+ },
446
+ {
447
+ "epoch": 7.104252400548697,
448
+ "grad_norm": 24.85959815979004,
449
+ "learning_rate": 1.4335154826958106e-05,
450
+ "loss": 0.4821,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.2414266117969825,
455
+ "grad_norm": 76.40132141113281,
456
+ "learning_rate": 1.4183363691560415e-05,
457
+ "loss": 0.4214,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.378600823045267,
462
+ "grad_norm": 16.172136306762695,
463
+ "learning_rate": 1.403157255616272e-05,
464
+ "loss": 0.3566,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.5157750342935525,
469
+ "grad_norm": 29.87400245666504,
470
+ "learning_rate": 1.3879781420765028e-05,
471
+ "loss": 0.4715,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.652949245541838,
476
+ "grad_norm": 26.169836044311523,
477
+ "learning_rate": 1.3727990285367337e-05,
478
+ "loss": 0.2751,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 7.790123456790123,
483
+ "grad_norm": 37.679039001464844,
484
+ "learning_rate": 1.3576199149969643e-05,
485
+ "loss": 0.3539,
486
+ "step": 1425
487
+ },
488
+ {
489
+ "epoch": 7.927297668038409,
490
+ "grad_norm": 40.94674301147461,
491
+ "learning_rate": 1.342440801457195e-05,
492
+ "loss": 0.4151,
493
+ "step": 1450
494
+ },
495
+ {
496
+ "epoch": 8.0,
497
+ "eval_accuracy": 0.5707154742096506,
498
+ "eval_loss": 2.1250150203704834,
499
+ "eval_macro_f1": 0.5681455854085589,
500
+ "eval_runtime": 1.4121,
501
+ "eval_samples_per_second": 425.6,
502
+ "eval_steps_per_second": 26.91,
503
+ "eval_weighted_f1": 0.5670316793297748,
504
+ "step": 1464
505
+ },
506
+ {
507
+ "epoch": 8.060356652949245,
508
+ "grad_norm": 32.447471618652344,
509
+ "learning_rate": 1.3272616879174259e-05,
510
+ "loss": 0.5599,
511
+ "step": 1475
512
+ },
513
+ {
514
+ "epoch": 8.197530864197532,
515
+ "grad_norm": 55.455440521240234,
516
+ "learning_rate": 1.3120825743776565e-05,
517
+ "loss": 0.1749,
518
+ "step": 1500
519
+ },
520
+ {
521
+ "epoch": 8.334705075445816,
522
+ "grad_norm": 1.4466408491134644,
523
+ "learning_rate": 1.2969034608378872e-05,
524
+ "loss": 0.2614,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 8.4718792866941,
529
+ "grad_norm": 20.379159927368164,
530
+ "learning_rate": 1.2817243472981178e-05,
531
+ "loss": 0.2438,
532
+ "step": 1550
533
+ },
534
+ {
535
+ "epoch": 8.609053497942387,
536
+ "grad_norm": 16.515714645385742,
537
+ "learning_rate": 1.2665452337583487e-05,
538
+ "loss": 0.3285,
539
+ "step": 1575
540
+ },
541
+ {
542
+ "epoch": 8.746227709190672,
543
+ "grad_norm": 84.41275024414062,
544
+ "learning_rate": 1.2513661202185794e-05,
545
+ "loss": 0.3344,
546
+ "step": 1600
547
+ },
548
+ {
549
+ "epoch": 8.883401920438958,
550
+ "grad_norm": 35.10468673706055,
551
+ "learning_rate": 1.23618700667881e-05,
552
+ "loss": 0.3595,
553
+ "step": 1625
554
+ },
555
+ {
556
+ "epoch": 9.0,
557
+ "eval_accuracy": 0.5540765391014975,
558
+ "eval_loss": 2.615231990814209,
559
+ "eval_macro_f1": 0.5529810778295975,
560
+ "eval_runtime": 1.3103,
561
+ "eval_samples_per_second": 458.67,
562
+ "eval_steps_per_second": 29.001,
563
+ "eval_weighted_f1": 0.5524891248981554,
564
+ "step": 1647
565
+ },
566
+ {
567
+ "epoch": 9.016460905349794,
568
+ "grad_norm": 10.12568473815918,
569
+ "learning_rate": 1.2210078931390409e-05,
570
+ "loss": 0.1753,
571
+ "step": 1650
572
+ },
573
+ {
574
+ "epoch": 9.15363511659808,
575
+ "grad_norm": 5.8272929191589355,
576
+ "learning_rate": 1.2058287795992716e-05,
577
+ "loss": 0.2451,
578
+ "step": 1675
579
+ },
580
+ {
581
+ "epoch": 9.290809327846365,
582
+ "grad_norm": 3.4700303077697754,
583
+ "learning_rate": 1.1906496660595022e-05,
584
+ "loss": 0.2241,
585
+ "step": 1700
586
+ },
587
+ {
588
+ "epoch": 9.427983539094651,
589
+ "grad_norm": 69.6553955078125,
590
+ "learning_rate": 1.1754705525197329e-05,
591
+ "loss": 0.1828,
592
+ "step": 1725
593
+ },
594
+ {
595
+ "epoch": 9.565157750342935,
596
+ "grad_norm": 55.69205856323242,
597
+ "learning_rate": 1.1602914389799638e-05,
598
+ "loss": 0.2487,
599
+ "step": 1750
600
+ },
601
+ {
602
+ "epoch": 9.70233196159122,
603
+ "grad_norm": 40.14244079589844,
604
+ "learning_rate": 1.1451123254401944e-05,
605
+ "loss": 0.1401,
606
+ "step": 1775
607
+ },
608
+ {
609
+ "epoch": 9.839506172839506,
610
+ "grad_norm": 5.400284767150879,
611
+ "learning_rate": 1.1299332119004251e-05,
612
+ "loss": 0.2419,
613
+ "step": 1800
614
+ },
615
+ {
616
+ "epoch": 9.976680384087791,
617
+ "grad_norm": 47.907840728759766,
618
+ "learning_rate": 1.1147540983606557e-05,
619
+ "loss": 0.0899,
620
+ "step": 1825
621
+ },
622
+ {
623
+ "epoch": 10.0,
624
+ "eval_accuracy": 0.5773710482529119,
625
+ "eval_loss": 3.038083553314209,
626
+ "eval_macro_f1": 0.5722266350495325,
627
+ "eval_runtime": 1.2585,
628
+ "eval_samples_per_second": 477.538,
629
+ "eval_steps_per_second": 30.194,
630
+ "eval_weighted_f1": 0.5704822977346388,
631
+ "step": 1830
632
+ },
633
+ {
634
+ "epoch": 10.109739368998628,
635
+ "grad_norm": 10.559795379638672,
636
+ "learning_rate": 1.0995749848208866e-05,
637
+ "loss": 0.0393,
638
+ "step": 1850
639
+ },
640
+ {
641
+ "epoch": 10.246913580246913,
642
+ "grad_norm": 141.248046875,
643
+ "learning_rate": 1.0843958712811173e-05,
644
+ "loss": 0.1879,
645
+ "step": 1875
646
+ },
647
+ {
648
+ "epoch": 10.3840877914952,
649
+ "grad_norm": 98.71314239501953,
650
+ "learning_rate": 1.0692167577413479e-05,
651
+ "loss": 0.1018,
652
+ "step": 1900
653
+ },
654
+ {
655
+ "epoch": 10.521262002743484,
656
+ "grad_norm": 0.7227563858032227,
657
+ "learning_rate": 1.0540376442015788e-05,
658
+ "loss": 0.0758,
659
+ "step": 1925
660
+ },
661
+ {
662
+ "epoch": 10.65843621399177,
663
+ "grad_norm": 30.410173416137695,
664
+ "learning_rate": 1.0388585306618095e-05,
665
+ "loss": 0.0888,
666
+ "step": 1950
667
+ },
668
+ {
669
+ "epoch": 10.795610425240055,
670
+ "grad_norm": 6.624603748321533,
671
+ "learning_rate": 1.02367941712204e-05,
672
+ "loss": 0.11,
673
+ "step": 1975
674
+ },
675
+ {
676
+ "epoch": 10.932784636488341,
677
+ "grad_norm": 1.9200135469436646,
678
+ "learning_rate": 1.008500303582271e-05,
679
+ "loss": 0.1223,
680
+ "step": 2000
681
+ },
682
+ {
683
+ "epoch": 11.0,
684
+ "eval_accuracy": 0.589018302828619,
685
+ "eval_loss": 3.6963894367218018,
686
+ "eval_macro_f1": 0.5850014748355203,
687
+ "eval_runtime": 1.1961,
688
+ "eval_samples_per_second": 502.487,
689
+ "eval_steps_per_second": 31.771,
690
+ "eval_weighted_f1": 0.5840226618260116,
691
+ "step": 2013
692
+ }
693
+ ],
694
+ "logging_steps": 25,
695
+ "max_steps": 3660,
696
+ "num_input_tokens_seen": 0,
697
+ "num_train_epochs": 20,
698
+ "save_steps": 500,
699
+ "stateful_callbacks": {
700
+ "EarlyStoppingCallback": {
701
+ "args": {
702
+ "early_stopping_patience": 3,
703
+ "early_stopping_threshold": 0.0
704
+ },
705
+ "attributes": {
706
+ "early_stopping_patience_counter": 0
707
+ }
708
+ },
709
+ "TrainerControl": {
710
+ "args": {
711
+ "should_epoch_stop": false,
712
+ "should_evaluate": false,
713
+ "should_log": false,
714
+ "should_save": true,
715
+ "should_training_stop": false
716
+ },
717
+ "attributes": {}
718
+ }
719
+ },
720
+ "total_flos": 0.0,
721
+ "train_batch_size": 8,
722
+ "trial_name": null,
723
+ "trial_params": null
724
+ }
ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2562/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-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2562/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2562/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-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2562/trainer_state.json ADDED
@@ -0,0 +1,911 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2013,
3
+ "best_metric": 0.5850014748355203,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-03-s42-20260830T235703Z-f21e2f7e/trainer/checkpoint-2013",
5
+ "epoch": 14.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2562,
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": 35.02544403076172,
15
+ "learning_rate": 1.3114754098360657e-06,
16
+ "loss": 4.6872,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27434842249657065,
21
+ "grad_norm": 21.24454689025879,
22
+ "learning_rate": 2.677595628415301e-06,
23
+ "loss": 4.5419,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.411522633744856,
28
+ "grad_norm": 20.140762329101562,
29
+ "learning_rate": 4.043715846994536e-06,
30
+ "loss": 4.5041,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5486968449931413,
35
+ "grad_norm": 36.08255386352539,
36
+ "learning_rate": 5.409836065573772e-06,
37
+ "loss": 4.4129,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6858710562414266,
42
+ "grad_norm": 17.723480224609375,
43
+ "learning_rate": 6.775956284153006e-06,
44
+ "loss": 4.4218,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.823045267489712,
49
+ "grad_norm": 30.218822479248047,
50
+ "learning_rate": 8.14207650273224e-06,
51
+ "loss": 4.394,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9602194787379973,
56
+ "grad_norm": 36.463706970214844,
57
+ "learning_rate": 9.508196721311476e-06,
58
+ "loss": 4.3698,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.41098169717138106,
64
+ "eval_loss": 1.0792137384414673,
65
+ "eval_macro_f1": 0.4061852484724344,
66
+ "eval_runtime": 1.1904,
67
+ "eval_samples_per_second": 504.879,
68
+ "eval_steps_per_second": 31.922,
69
+ "eval_weighted_f1": 0.40612451900351487,
70
+ "step": 183
71
+ },
72
+ {
73
+ "epoch": 1.093278463648834,
74
+ "grad_norm": 28.814289093017578,
75
+ "learning_rate": 1.0874316939890712e-05,
76
+ "loss": 4.1889,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2304526748971194,
81
+ "grad_norm": 16.32625961303711,
82
+ "learning_rate": 1.2240437158469946e-05,
83
+ "loss": 4.2271,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3676268861454046,
88
+ "grad_norm": 25.90184783935547,
89
+ "learning_rate": 1.3606557377049181e-05,
90
+ "loss": 4.2057,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.50480109739369,
95
+ "grad_norm": 22.540451049804688,
96
+ "learning_rate": 1.4972677595628417e-05,
97
+ "loss": 4.251,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6419753086419753,
102
+ "grad_norm": 24.28812026977539,
103
+ "learning_rate": 1.633879781420765e-05,
104
+ "loss": 4.3731,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7791495198902605,
109
+ "grad_norm": 39.98865509033203,
110
+ "learning_rate": 1.7704918032786887e-05,
111
+ "loss": 4.2678,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.916323731138546,
116
+ "grad_norm": 29.907140731811523,
117
+ "learning_rate": 1.9071038251366122e-05,
118
+ "loss": 4.1656,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.49916805324459235,
124
+ "eval_loss": 1.0238842964172363,
125
+ "eval_macro_f1": 0.484835268639423,
126
+ "eval_runtime": 1.1675,
127
+ "eval_samples_per_second": 514.778,
128
+ "eval_steps_per_second": 32.548,
129
+ "eval_weighted_f1": 0.486985131271952,
130
+ "step": 366
131
+ },
132
+ {
133
+ "epoch": 2.049382716049383,
134
+ "grad_norm": 23.380916595458984,
135
+ "learning_rate": 1.995142683667274e-05,
136
+ "loss": 3.9306,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.186556927297668,
141
+ "grad_norm": 39.58391571044922,
142
+ "learning_rate": 1.9799635701275045e-05,
143
+ "loss": 3.6788,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.3237311385459534,
148
+ "grad_norm": 20.238113403320312,
149
+ "learning_rate": 1.9647844565877354e-05,
150
+ "loss": 3.7912,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.460905349794239,
155
+ "grad_norm": 29.342870712280273,
156
+ "learning_rate": 1.9496053430479663e-05,
157
+ "loss": 3.8442,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.598079561042524,
162
+ "grad_norm": 55.8082389831543,
163
+ "learning_rate": 1.934426229508197e-05,
164
+ "loss": 3.711,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.7352537722908092,
169
+ "grad_norm": 40.55704879760742,
170
+ "learning_rate": 1.9192471159684274e-05,
171
+ "loss": 3.7917,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.8724279835390947,
176
+ "grad_norm": 34.73087692260742,
177
+ "learning_rate": 1.9040680024286583e-05,
178
+ "loss": 3.6319,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5391014975041597,
184
+ "eval_loss": 0.9485401511192322,
185
+ "eval_macro_f1": 0.5393116064326674,
186
+ "eval_runtime": 1.307,
187
+ "eval_samples_per_second": 459.841,
188
+ "eval_steps_per_second": 29.075,
189
+ "eval_weighted_f1": 0.538541502679688,
190
+ "step": 549
191
+ },
192
+ {
193
+ "epoch": 3.0054869684499312,
194
+ "grad_norm": 24.43846893310547,
195
+ "learning_rate": 1.888888888888889e-05,
196
+ "loss": 3.5135,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.1426611796982167,
201
+ "grad_norm": 56.53961181640625,
202
+ "learning_rate": 1.8737097753491198e-05,
203
+ "loss": 2.9245,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.279835390946502,
208
+ "grad_norm": 37.9984016418457,
209
+ "learning_rate": 1.8585306618093504e-05,
210
+ "loss": 2.8603,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.4170096021947876,
215
+ "grad_norm": 26.55724334716797,
216
+ "learning_rate": 1.8433515482695813e-05,
217
+ "loss": 2.9073,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.5541838134430725,
222
+ "grad_norm": 48.19314193725586,
223
+ "learning_rate": 1.828172434729812e-05,
224
+ "loss": 2.7598,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.691358024691358,
229
+ "grad_norm": 35.43125915527344,
230
+ "learning_rate": 1.8129933211900424e-05,
231
+ "loss": 2.8162,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.8285322359396434,
236
+ "grad_norm": 41.655555725097656,
237
+ "learning_rate": 1.7978142076502733e-05,
238
+ "loss": 2.8093,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.9657064471879284,
243
+ "grad_norm": 54.582244873046875,
244
+ "learning_rate": 1.7826350941105042e-05,
245
+ "loss": 2.9492,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 4.0,
250
+ "eval_accuracy": 0.5424292845257903,
251
+ "eval_loss": 1.0065358877182007,
252
+ "eval_macro_f1": 0.5468183232889116,
253
+ "eval_runtime": 1.2725,
254
+ "eval_samples_per_second": 472.305,
255
+ "eval_steps_per_second": 29.863,
256
+ "eval_weighted_f1": 0.5465966591345861,
257
+ "step": 732
258
+ },
259
+ {
260
+ "epoch": 4.098765432098766,
261
+ "grad_norm": 27.443614959716797,
262
+ "learning_rate": 1.7674559805707348e-05,
263
+ "loss": 2.2806,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.23593964334705,
268
+ "grad_norm": 35.80004119873047,
269
+ "learning_rate": 1.7522768670309657e-05,
270
+ "loss": 2.0458,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.373113854595336,
275
+ "grad_norm": 33.068687438964844,
276
+ "learning_rate": 1.7370977534911962e-05,
277
+ "loss": 1.9547,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.510288065843621,
282
+ "grad_norm": 52.475746154785156,
283
+ "learning_rate": 1.7219186399514268e-05,
284
+ "loss": 2.0498,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.647462277091907,
289
+ "grad_norm": 36.4451789855957,
290
+ "learning_rate": 1.7067395264116577e-05,
291
+ "loss": 1.9751,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.784636488340192,
296
+ "grad_norm": 29.826025009155273,
297
+ "learning_rate": 1.6915604128718883e-05,
298
+ "loss": 1.7888,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.921810699588478,
303
+ "grad_norm": 36.69779586791992,
304
+ "learning_rate": 1.6763812993321192e-05,
305
+ "loss": 1.8333,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 5.0,
310
+ "eval_accuracy": 0.5490848585690515,
311
+ "eval_loss": 1.2077672481536865,
312
+ "eval_macro_f1": 0.5414092808818651,
313
+ "eval_runtime": 1.3947,
314
+ "eval_samples_per_second": 430.931,
315
+ "eval_steps_per_second": 27.247,
316
+ "eval_weighted_f1": 0.5383178751053753,
317
+ "step": 915
318
+ },
319
+ {
320
+ "epoch": 5.054869684499314,
321
+ "grad_norm": 68.85910034179688,
322
+ "learning_rate": 1.66120218579235e-05,
323
+ "loss": 1.4793,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.1920438957476,
328
+ "grad_norm": 85.61286163330078,
329
+ "learning_rate": 1.6460230722525807e-05,
330
+ "loss": 1.252,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.329218106995885,
335
+ "grad_norm": 56.671077728271484,
336
+ "learning_rate": 1.6308439587128112e-05,
337
+ "loss": 1.35,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.4663923182441705,
342
+ "grad_norm": 62.90438461303711,
343
+ "learning_rate": 1.615664845173042e-05,
344
+ "loss": 1.1463,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.603566529492455,
349
+ "grad_norm": 93.8409423828125,
350
+ "learning_rate": 1.6004857316332727e-05,
351
+ "loss": 1.003,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.7407407407407405,
356
+ "grad_norm": 36.048099517822266,
357
+ "learning_rate": 1.5853066180935036e-05,
358
+ "loss": 1.2405,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.877914951989026,
363
+ "grad_norm": 84.30831909179688,
364
+ "learning_rate": 1.570127504553734e-05,
365
+ "loss": 1.0601,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 6.0,
370
+ "eval_accuracy": 0.5391014975041597,
371
+ "eval_loss": 1.6696314811706543,
372
+ "eval_macro_f1": 0.5426858378219005,
373
+ "eval_runtime": 1.23,
374
+ "eval_samples_per_second": 488.626,
375
+ "eval_steps_per_second": 30.895,
376
+ "eval_weighted_f1": 0.5424897563415262,
377
+ "step": 1098
378
+ },
379
+ {
380
+ "epoch": 6.0109739368998625,
381
+ "grad_norm": 26.06507110595703,
382
+ "learning_rate": 1.5549483910139647e-05,
383
+ "loss": 0.9767,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.148148148148148,
388
+ "grad_norm": 46.933868408203125,
389
+ "learning_rate": 1.5397692774741956e-05,
390
+ "loss": 0.7216,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.285322359396433,
395
+ "grad_norm": 51.907161712646484,
396
+ "learning_rate": 1.5245901639344264e-05,
397
+ "loss": 0.6712,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.422496570644719,
402
+ "grad_norm": 29.579395294189453,
403
+ "learning_rate": 1.5094110503946571e-05,
404
+ "loss": 0.7167,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.559670781893004,
409
+ "grad_norm": 19.197471618652344,
410
+ "learning_rate": 1.4942319368548878e-05,
411
+ "loss": 0.7279,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.69684499314129,
416
+ "grad_norm": 45.326900482177734,
417
+ "learning_rate": 1.4790528233151184e-05,
418
+ "loss": 0.7464,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.834019204389575,
423
+ "grad_norm": 59.55309295654297,
424
+ "learning_rate": 1.4638737097753493e-05,
425
+ "loss": 0.5537,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 6.97119341563786,
430
+ "grad_norm": 74.94731140136719,
431
+ "learning_rate": 1.44869459623558e-05,
432
+ "loss": 0.6839,
433
+ "step": 1275
434
+ },
435
+ {
436
+ "epoch": 7.0,
437
+ "eval_accuracy": 0.5657237936772047,
438
+ "eval_loss": 2.0491621494293213,
439
+ "eval_macro_f1": 0.5564754364524109,
440
+ "eval_runtime": 1.1854,
441
+ "eval_samples_per_second": 507.023,
442
+ "eval_steps_per_second": 32.058,
443
+ "eval_weighted_f1": 0.5552805344381514,
444
+ "step": 1281
445
+ },
446
+ {
447
+ "epoch": 7.104252400548697,
448
+ "grad_norm": 24.85959815979004,
449
+ "learning_rate": 1.4335154826958106e-05,
450
+ "loss": 0.4821,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.2414266117969825,
455
+ "grad_norm": 76.40132141113281,
456
+ "learning_rate": 1.4183363691560415e-05,
457
+ "loss": 0.4214,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.378600823045267,
462
+ "grad_norm": 16.172136306762695,
463
+ "learning_rate": 1.403157255616272e-05,
464
+ "loss": 0.3566,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.5157750342935525,
469
+ "grad_norm": 29.87400245666504,
470
+ "learning_rate": 1.3879781420765028e-05,
471
+ "loss": 0.4715,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.652949245541838,
476
+ "grad_norm": 26.169836044311523,
477
+ "learning_rate": 1.3727990285367337e-05,
478
+ "loss": 0.2751,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 7.790123456790123,
483
+ "grad_norm": 37.679039001464844,
484
+ "learning_rate": 1.3576199149969643e-05,
485
+ "loss": 0.3539,
486
+ "step": 1425
487
+ },
488
+ {
489
+ "epoch": 7.927297668038409,
490
+ "grad_norm": 40.94674301147461,
491
+ "learning_rate": 1.342440801457195e-05,
492
+ "loss": 0.4151,
493
+ "step": 1450
494
+ },
495
+ {
496
+ "epoch": 8.0,
497
+ "eval_accuracy": 0.5707154742096506,
498
+ "eval_loss": 2.1250150203704834,
499
+ "eval_macro_f1": 0.5681455854085589,
500
+ "eval_runtime": 1.4121,
501
+ "eval_samples_per_second": 425.6,
502
+ "eval_steps_per_second": 26.91,
503
+ "eval_weighted_f1": 0.5670316793297748,
504
+ "step": 1464
505
+ },
506
+ {
507
+ "epoch": 8.060356652949245,
508
+ "grad_norm": 32.447471618652344,
509
+ "learning_rate": 1.3272616879174259e-05,
510
+ "loss": 0.5599,
511
+ "step": 1475
512
+ },
513
+ {
514
+ "epoch": 8.197530864197532,
515
+ "grad_norm": 55.455440521240234,
516
+ "learning_rate": 1.3120825743776565e-05,
517
+ "loss": 0.1749,
518
+ "step": 1500
519
+ },
520
+ {
521
+ "epoch": 8.334705075445816,
522
+ "grad_norm": 1.4466408491134644,
523
+ "learning_rate": 1.2969034608378872e-05,
524
+ "loss": 0.2614,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 8.4718792866941,
529
+ "grad_norm": 20.379159927368164,
530
+ "learning_rate": 1.2817243472981178e-05,
531
+ "loss": 0.2438,
532
+ "step": 1550
533
+ },
534
+ {
535
+ "epoch": 8.609053497942387,
536
+ "grad_norm": 16.515714645385742,
537
+ "learning_rate": 1.2665452337583487e-05,
538
+ "loss": 0.3285,
539
+ "step": 1575
540
+ },
541
+ {
542
+ "epoch": 8.746227709190672,
543
+ "grad_norm": 84.41275024414062,
544
+ "learning_rate": 1.2513661202185794e-05,
545
+ "loss": 0.3344,
546
+ "step": 1600
547
+ },
548
+ {
549
+ "epoch": 8.883401920438958,
550
+ "grad_norm": 35.10468673706055,
551
+ "learning_rate": 1.23618700667881e-05,
552
+ "loss": 0.3595,
553
+ "step": 1625
554
+ },
555
+ {
556
+ "epoch": 9.0,
557
+ "eval_accuracy": 0.5540765391014975,
558
+ "eval_loss": 2.615231990814209,
559
+ "eval_macro_f1": 0.5529810778295975,
560
+ "eval_runtime": 1.3103,
561
+ "eval_samples_per_second": 458.67,
562
+ "eval_steps_per_second": 29.001,
563
+ "eval_weighted_f1": 0.5524891248981554,
564
+ "step": 1647
565
+ },
566
+ {
567
+ "epoch": 9.016460905349794,
568
+ "grad_norm": 10.12568473815918,
569
+ "learning_rate": 1.2210078931390409e-05,
570
+ "loss": 0.1753,
571
+ "step": 1650
572
+ },
573
+ {
574
+ "epoch": 9.15363511659808,
575
+ "grad_norm": 5.8272929191589355,
576
+ "learning_rate": 1.2058287795992716e-05,
577
+ "loss": 0.2451,
578
+ "step": 1675
579
+ },
580
+ {
581
+ "epoch": 9.290809327846365,
582
+ "grad_norm": 3.4700303077697754,
583
+ "learning_rate": 1.1906496660595022e-05,
584
+ "loss": 0.2241,
585
+ "step": 1700
586
+ },
587
+ {
588
+ "epoch": 9.427983539094651,
589
+ "grad_norm": 69.6553955078125,
590
+ "learning_rate": 1.1754705525197329e-05,
591
+ "loss": 0.1828,
592
+ "step": 1725
593
+ },
594
+ {
595
+ "epoch": 9.565157750342935,
596
+ "grad_norm": 55.69205856323242,
597
+ "learning_rate": 1.1602914389799638e-05,
598
+ "loss": 0.2487,
599
+ "step": 1750
600
+ },
601
+ {
602
+ "epoch": 9.70233196159122,
603
+ "grad_norm": 40.14244079589844,
604
+ "learning_rate": 1.1451123254401944e-05,
605
+ "loss": 0.1401,
606
+ "step": 1775
607
+ },
608
+ {
609
+ "epoch": 9.839506172839506,
610
+ "grad_norm": 5.400284767150879,
611
+ "learning_rate": 1.1299332119004251e-05,
612
+ "loss": 0.2419,
613
+ "step": 1800
614
+ },
615
+ {
616
+ "epoch": 9.976680384087791,
617
+ "grad_norm": 47.907840728759766,
618
+ "learning_rate": 1.1147540983606557e-05,
619
+ "loss": 0.0899,
620
+ "step": 1825
621
+ },
622
+ {
623
+ "epoch": 10.0,
624
+ "eval_accuracy": 0.5773710482529119,
625
+ "eval_loss": 3.038083553314209,
626
+ "eval_macro_f1": 0.5722266350495325,
627
+ "eval_runtime": 1.2585,
628
+ "eval_samples_per_second": 477.538,
629
+ "eval_steps_per_second": 30.194,
630
+ "eval_weighted_f1": 0.5704822977346388,
631
+ "step": 1830
632
+ },
633
+ {
634
+ "epoch": 10.109739368998628,
635
+ "grad_norm": 10.559795379638672,
636
+ "learning_rate": 1.0995749848208866e-05,
637
+ "loss": 0.0393,
638
+ "step": 1850
639
+ },
640
+ {
641
+ "epoch": 10.246913580246913,
642
+ "grad_norm": 141.248046875,
643
+ "learning_rate": 1.0843958712811173e-05,
644
+ "loss": 0.1879,
645
+ "step": 1875
646
+ },
647
+ {
648
+ "epoch": 10.3840877914952,
649
+ "grad_norm": 98.71314239501953,
650
+ "learning_rate": 1.0692167577413479e-05,
651
+ "loss": 0.1018,
652
+ "step": 1900
653
+ },
654
+ {
655
+ "epoch": 10.521262002743484,
656
+ "grad_norm": 0.7227563858032227,
657
+ "learning_rate": 1.0540376442015788e-05,
658
+ "loss": 0.0758,
659
+ "step": 1925
660
+ },
661
+ {
662
+ "epoch": 10.65843621399177,
663
+ "grad_norm": 30.410173416137695,
664
+ "learning_rate": 1.0388585306618095e-05,
665
+ "loss": 0.0888,
666
+ "step": 1950
667
+ },
668
+ {
669
+ "epoch": 10.795610425240055,
670
+ "grad_norm": 6.624603748321533,
671
+ "learning_rate": 1.02367941712204e-05,
672
+ "loss": 0.11,
673
+ "step": 1975
674
+ },
675
+ {
676
+ "epoch": 10.932784636488341,
677
+ "grad_norm": 1.9200135469436646,
678
+ "learning_rate": 1.008500303582271e-05,
679
+ "loss": 0.1223,
680
+ "step": 2000
681
+ },
682
+ {
683
+ "epoch": 11.0,
684
+ "eval_accuracy": 0.589018302828619,
685
+ "eval_loss": 3.6963894367218018,
686
+ "eval_macro_f1": 0.5850014748355203,
687
+ "eval_runtime": 1.1961,
688
+ "eval_samples_per_second": 502.487,
689
+ "eval_steps_per_second": 31.771,
690
+ "eval_weighted_f1": 0.5840226618260116,
691
+ "step": 2013
692
+ },
693
+ {
694
+ "epoch": 11.065843621399177,
695
+ "grad_norm": 1.25968599319458,
696
+ "learning_rate": 9.933211900425015e-06,
697
+ "loss": 0.1017,
698
+ "step": 2025
699
+ },
700
+ {
701
+ "epoch": 11.203017832647463,
702
+ "grad_norm": 0.8258217573165894,
703
+ "learning_rate": 9.781420765027323e-06,
704
+ "loss": 0.0625,
705
+ "step": 2050
706
+ },
707
+ {
708
+ "epoch": 11.340192043895748,
709
+ "grad_norm": 0.7899554371833801,
710
+ "learning_rate": 9.62962962962963e-06,
711
+ "loss": 0.0888,
712
+ "step": 2075
713
+ },
714
+ {
715
+ "epoch": 11.477366255144032,
716
+ "grad_norm": 0.4299376606941223,
717
+ "learning_rate": 9.477838494231937e-06,
718
+ "loss": 0.1112,
719
+ "step": 2100
720
+ },
721
+ {
722
+ "epoch": 11.614540466392318,
723
+ "grad_norm": 0.7064834237098694,
724
+ "learning_rate": 9.326047358834245e-06,
725
+ "loss": 0.2107,
726
+ "step": 2125
727
+ },
728
+ {
729
+ "epoch": 11.751714677640603,
730
+ "grad_norm": 90.66522216796875,
731
+ "learning_rate": 9.174256223436552e-06,
732
+ "loss": 0.1817,
733
+ "step": 2150
734
+ },
735
+ {
736
+ "epoch": 11.88888888888889,
737
+ "grad_norm": 0.8168855309486389,
738
+ "learning_rate": 9.02246508803886e-06,
739
+ "loss": 0.1311,
740
+ "step": 2175
741
+ },
742
+ {
743
+ "epoch": 12.0,
744
+ "eval_accuracy": 0.5773710482529119,
745
+ "eval_loss": 3.8245904445648193,
746
+ "eval_macro_f1": 0.5722194971658109,
747
+ "eval_runtime": 1.3792,
748
+ "eval_samples_per_second": 435.753,
749
+ "eval_steps_per_second": 27.552,
750
+ "eval_weighted_f1": 0.5694208396512743,
751
+ "step": 2196
752
+ },
753
+ {
754
+ "epoch": 12.021947873799725,
755
+ "grad_norm": 0.03334344923496246,
756
+ "learning_rate": 8.870673952641167e-06,
757
+ "loss": 0.0936,
758
+ "step": 2200
759
+ },
760
+ {
761
+ "epoch": 12.159122085048011,
762
+ "grad_norm": 0.43032997846603394,
763
+ "learning_rate": 8.718882817243474e-06,
764
+ "loss": 0.0628,
765
+ "step": 2225
766
+ },
767
+ {
768
+ "epoch": 12.296296296296296,
769
+ "grad_norm": 0.32953521609306335,
770
+ "learning_rate": 8.567091681845781e-06,
771
+ "loss": 0.0083,
772
+ "step": 2250
773
+ },
774
+ {
775
+ "epoch": 12.433470507544582,
776
+ "grad_norm": 37.53985595703125,
777
+ "learning_rate": 8.415300546448089e-06,
778
+ "loss": 0.044,
779
+ "step": 2275
780
+ },
781
+ {
782
+ "epoch": 12.570644718792867,
783
+ "grad_norm": 0.20523951947689056,
784
+ "learning_rate": 8.263509411050396e-06,
785
+ "loss": 0.0268,
786
+ "step": 2300
787
+ },
788
+ {
789
+ "epoch": 12.707818930041153,
790
+ "grad_norm": 4.1444878578186035,
791
+ "learning_rate": 8.111718275652702e-06,
792
+ "loss": 0.0596,
793
+ "step": 2325
794
+ },
795
+ {
796
+ "epoch": 12.844993141289438,
797
+ "grad_norm": 0.061257388442754745,
798
+ "learning_rate": 7.959927140255009e-06,
799
+ "loss": 0.0659,
800
+ "step": 2350
801
+ },
802
+ {
803
+ "epoch": 12.982167352537722,
804
+ "grad_norm": 0.05395309999585152,
805
+ "learning_rate": 7.808136004857318e-06,
806
+ "loss": 0.0581,
807
+ "step": 2375
808
+ },
809
+ {
810
+ "epoch": 13.0,
811
+ "eval_accuracy": 0.5557404326123128,
812
+ "eval_loss": 4.146053314208984,
813
+ "eval_macro_f1": 0.5560115122475766,
814
+ "eval_runtime": 1.1959,
815
+ "eval_samples_per_second": 502.565,
816
+ "eval_steps_per_second": 31.776,
817
+ "eval_weighted_f1": 0.5549405339030834,
818
+ "step": 2379
819
+ },
820
+ {
821
+ "epoch": 13.11522633744856,
822
+ "grad_norm": 0.043270938098430634,
823
+ "learning_rate": 7.656344869459624e-06,
824
+ "loss": 0.0528,
825
+ "step": 2400
826
+ },
827
+ {
828
+ "epoch": 13.252400548696844,
829
+ "grad_norm": 9.04575252532959,
830
+ "learning_rate": 7.504553734061931e-06,
831
+ "loss": 0.1646,
832
+ "step": 2425
833
+ },
834
+ {
835
+ "epoch": 13.38957475994513,
836
+ "grad_norm": 0.08492506295442581,
837
+ "learning_rate": 7.352762598664238e-06,
838
+ "loss": 0.0393,
839
+ "step": 2450
840
+ },
841
+ {
842
+ "epoch": 13.526748971193415,
843
+ "grad_norm": 4.976897716522217,
844
+ "learning_rate": 7.2009714632665465e-06,
845
+ "loss": 0.0251,
846
+ "step": 2475
847
+ },
848
+ {
849
+ "epoch": 13.663923182441701,
850
+ "grad_norm": 69.86087799072266,
851
+ "learning_rate": 7.049180327868853e-06,
852
+ "loss": 0.0125,
853
+ "step": 2500
854
+ },
855
+ {
856
+ "epoch": 13.801097393689986,
857
+ "grad_norm": 8.68476390838623,
858
+ "learning_rate": 6.89738919247116e-06,
859
+ "loss": 0.0216,
860
+ "step": 2525
861
+ },
862
+ {
863
+ "epoch": 13.938271604938272,
864
+ "grad_norm": 110.86022186279297,
865
+ "learning_rate": 6.745598057073467e-06,
866
+ "loss": 0.0492,
867
+ "step": 2550
868
+ },
869
+ {
870
+ "epoch": 14.0,
871
+ "eval_accuracy": 0.5707154742096506,
872
+ "eval_loss": 4.267321586608887,
873
+ "eval_macro_f1": 0.5679075347662623,
874
+ "eval_runtime": 1.3073,
875
+ "eval_samples_per_second": 459.712,
876
+ "eval_steps_per_second": 29.067,
877
+ "eval_weighted_f1": 0.5671223382127905,
878
+ "step": 2562
879
+ }
880
+ ],
881
+ "logging_steps": 25,
882
+ "max_steps": 3660,
883
+ "num_input_tokens_seen": 0,
884
+ "num_train_epochs": 20,
885
+ "save_steps": 500,
886
+ "stateful_callbacks": {
887
+ "EarlyStoppingCallback": {
888
+ "args": {
889
+ "early_stopping_patience": 3,
890
+ "early_stopping_threshold": 0.0
891
+ },
892
+ "attributes": {
893
+ "early_stopping_patience_counter": 3
894
+ }
895
+ },
896
+ "TrainerControl": {
897
+ "args": {
898
+ "should_epoch_stop": false,
899
+ "should_evaluate": false,
900
+ "should_log": false,
901
+ "should_save": true,
902
+ "should_training_stop": true
903
+ },
904
+ "attributes": {}
905
+ }
906
+ },
907
+ "total_flos": 0.0,
908
+ "train_batch_size": 8,
909
+ "trial_name": null,
910
+ "trial_params": null
911
+ }
ood-vifactcheck-visobert-source-source-04-s42-20260831T001152Z-f21e2f7e/trainer/checkpoint-1584/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-s42-20260831T001152Z-f21e2f7e/trainer/checkpoint-1584/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-04-s42-20260831T001152Z-f21e2f7e/trainer/checkpoint-1584/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-s42-20260831T001152Z-f21e2f7e/trainer/checkpoint-1584/trainer_state.json ADDED
@@ -0,0 +1,583 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1584,
3
+ "best_metric": 0.5827959365470929,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-04-s42-20260831T001152Z-f21e2f7e/trainer/checkpoint-1584",
5
+ "epoch": 9.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1584,
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": 22.647480010986328,
15
+ "learning_rate": 1.3636363636363636e-06,
16
+ "loss": 4.6969,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.2849002849002849,
21
+ "grad_norm": 25.382366180419922,
22
+ "learning_rate": 2.784090909090909e-06,
23
+ "loss": 4.5328,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.42735042735042733,
28
+ "grad_norm": 29.886882781982422,
29
+ "learning_rate": 4.204545454545455e-06,
30
+ "loss": 4.4957,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5698005698005698,
35
+ "grad_norm": 27.658693313598633,
36
+ "learning_rate": 5.625e-06,
37
+ "loss": 4.4721,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.7122507122507122,
42
+ "grad_norm": 19.12157440185547,
43
+ "learning_rate": 7.045454545454546e-06,
44
+ "loss": 4.377,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8547008547008547,
49
+ "grad_norm": 17.77869987487793,
50
+ "learning_rate": 8.465909090909091e-06,
51
+ "loss": 4.3755,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9971509971509972,
56
+ "grad_norm": 57.97350311279297,
57
+ "learning_rate": 9.886363636363637e-06,
58
+ "loss": 4.3975,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.4281567489114659,
64
+ "eval_loss": 1.068101406097412,
65
+ "eval_macro_f1": 0.41701292397097306,
66
+ "eval_runtime": 1.3598,
67
+ "eval_samples_per_second": 506.706,
68
+ "eval_steps_per_second": 32.359,
69
+ "eval_weighted_f1": 0.416564605324622,
70
+ "step": 176
71
+ },
72
+ {
73
+ "epoch": 1.1367521367521367,
74
+ "grad_norm": 26.81343650817871,
75
+ "learning_rate": 1.1306818181818183e-05,
76
+ "loss": 4.1715,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2792022792022792,
81
+ "grad_norm": 31.017099380493164,
82
+ "learning_rate": 1.2727272727272728e-05,
83
+ "loss": 4.3255,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.4216524216524218,
88
+ "grad_norm": 31.356576919555664,
89
+ "learning_rate": 1.4147727272727274e-05,
90
+ "loss": 4.2753,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.564102564102564,
95
+ "grad_norm": 40.33323669433594,
96
+ "learning_rate": 1.5568181818181822e-05,
97
+ "loss": 4.2271,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.7065527065527064,
102
+ "grad_norm": 23.7972412109375,
103
+ "learning_rate": 1.6988636363636368e-05,
104
+ "loss": 4.2389,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.849002849002849,
109
+ "grad_norm": 39.20664978027344,
110
+ "learning_rate": 1.840909090909091e-05,
111
+ "loss": 4.3281,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.9914529914529915,
116
+ "grad_norm": 37.309452056884766,
117
+ "learning_rate": 1.9829545454545456e-05,
118
+ "loss": 4.2745,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.45573294629898403,
124
+ "eval_loss": 1.0810412168502808,
125
+ "eval_macro_f1": 0.3707354177879669,
126
+ "eval_runtime": 1.4696,
127
+ "eval_samples_per_second": 468.849,
128
+ "eval_steps_per_second": 29.941,
129
+ "eval_weighted_f1": 0.3757687195315277,
130
+ "step": 352
131
+ },
132
+ {
133
+ "epoch": 2.131054131054131,
134
+ "grad_norm": 18.360013961791992,
135
+ "learning_rate": 1.9861111111111114e-05,
136
+ "loss": 3.8592,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.2735042735042734,
141
+ "grad_norm": 19.811275482177734,
142
+ "learning_rate": 1.9703282828282832e-05,
143
+ "loss": 3.8411,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.415954415954416,
148
+ "grad_norm": 22.7065486907959,
149
+ "learning_rate": 1.9545454545454546e-05,
150
+ "loss": 3.6747,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.5584045584045585,
155
+ "grad_norm": 46.72457504272461,
156
+ "learning_rate": 1.9387626262626263e-05,
157
+ "loss": 3.6057,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.700854700854701,
162
+ "grad_norm": 27.52411651611328,
163
+ "learning_rate": 1.922979797979798e-05,
164
+ "loss": 3.5957,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.8433048433048436,
169
+ "grad_norm": 43.00493240356445,
170
+ "learning_rate": 1.90719696969697e-05,
171
+ "loss": 3.5362,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.9857549857549857,
176
+ "grad_norm": 22.293670654296875,
177
+ "learning_rate": 1.8914141414141416e-05,
178
+ "loss": 3.666,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5195936139332366,
184
+ "eval_loss": 0.9567133784294128,
185
+ "eval_macro_f1": 0.5007173131879014,
186
+ "eval_runtime": 1.6699,
187
+ "eval_samples_per_second": 412.591,
188
+ "eval_steps_per_second": 26.348,
189
+ "eval_weighted_f1": 0.5030444563452333,
190
+ "step": 528
191
+ },
192
+ {
193
+ "epoch": 3.1253561253561255,
194
+ "grad_norm": 53.87973403930664,
195
+ "learning_rate": 1.8756313131313133e-05,
196
+ "loss": 3.1279,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.267806267806268,
201
+ "grad_norm": 32.13964080810547,
202
+ "learning_rate": 1.859848484848485e-05,
203
+ "loss": 2.7206,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.41025641025641,
208
+ "grad_norm": 33.29050064086914,
209
+ "learning_rate": 1.844065656565657e-05,
210
+ "loss": 2.7132,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.5527065527065527,
215
+ "grad_norm": 76.17159271240234,
216
+ "learning_rate": 1.8282828282828286e-05,
217
+ "loss": 2.8156,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.695156695156695,
222
+ "grad_norm": 28.67092514038086,
223
+ "learning_rate": 1.8125e-05,
224
+ "loss": 2.6867,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.8376068376068377,
229
+ "grad_norm": 39.83797836303711,
230
+ "learning_rate": 1.7967171717171718e-05,
231
+ "loss": 2.7785,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.98005698005698,
236
+ "grad_norm": 67.78004455566406,
237
+ "learning_rate": 1.7809343434343435e-05,
238
+ "loss": 2.8467,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 4.0,
243
+ "eval_accuracy": 0.5239477503628447,
244
+ "eval_loss": 1.0697739124298096,
245
+ "eval_macro_f1": 0.5041346579802543,
246
+ "eval_runtime": 1.3567,
247
+ "eval_samples_per_second": 507.839,
248
+ "eval_steps_per_second": 32.431,
249
+ "eval_weighted_f1": 0.5067010462199292,
250
+ "step": 704
251
+ },
252
+ {
253
+ "epoch": 4.119658119658119,
254
+ "grad_norm": 29.25116729736328,
255
+ "learning_rate": 1.7651515151515153e-05,
256
+ "loss": 1.9684,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.262108262108262,
261
+ "grad_norm": 138.7740478515625,
262
+ "learning_rate": 1.749368686868687e-05,
263
+ "loss": 1.8833,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.404558404558404,
268
+ "grad_norm": 43.61983108520508,
269
+ "learning_rate": 1.7335858585858588e-05,
270
+ "loss": 2.0275,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.547008547008547,
275
+ "grad_norm": 43.14507293701172,
276
+ "learning_rate": 1.7178030303030305e-05,
277
+ "loss": 1.9386,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.689458689458689,
282
+ "grad_norm": 67.69963836669922,
283
+ "learning_rate": 1.7020202020202023e-05,
284
+ "loss": 1.6151,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.831908831908832,
289
+ "grad_norm": 54.038822174072266,
290
+ "learning_rate": 1.6862373737373737e-05,
291
+ "loss": 1.7144,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.9743589743589745,
296
+ "grad_norm": 68.24846649169922,
297
+ "learning_rate": 1.6704545454545454e-05,
298
+ "loss": 1.9561,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 5.0,
303
+ "eval_accuracy": 0.5660377358490566,
304
+ "eval_loss": 1.1763917207717896,
305
+ "eval_macro_f1": 0.5619178548450664,
306
+ "eval_runtime": 1.5814,
307
+ "eval_samples_per_second": 435.694,
308
+ "eval_steps_per_second": 27.824,
309
+ "eval_weighted_f1": 0.5624988377140542,
310
+ "step": 880
311
+ },
312
+ {
313
+ "epoch": 5.113960113960114,
314
+ "grad_norm": 53.75267028808594,
315
+ "learning_rate": 1.6546717171717172e-05,
316
+ "loss": 1.2223,
317
+ "step": 900
318
+ },
319
+ {
320
+ "epoch": 5.256410256410256,
321
+ "grad_norm": 74.08851623535156,
322
+ "learning_rate": 1.638888888888889e-05,
323
+ "loss": 1.141,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.398860398860399,
328
+ "grad_norm": 45.42808151245117,
329
+ "learning_rate": 1.6231060606060607e-05,
330
+ "loss": 1.0914,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.5413105413105415,
335
+ "grad_norm": 51.34950637817383,
336
+ "learning_rate": 1.6073232323232324e-05,
337
+ "loss": 0.9795,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.683760683760684,
342
+ "grad_norm": 73.9638442993164,
343
+ "learning_rate": 1.5915404040404042e-05,
344
+ "loss": 1.0929,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.8262108262108265,
349
+ "grad_norm": 41.8527717590332,
350
+ "learning_rate": 1.575757575757576e-05,
351
+ "loss": 1.1185,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.968660968660968,
356
+ "grad_norm": 109.64105224609375,
357
+ "learning_rate": 1.5599747474747477e-05,
358
+ "loss": 0.9814,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 6.0,
363
+ "eval_accuracy": 0.5529753265602322,
364
+ "eval_loss": 1.7165353298187256,
365
+ "eval_macro_f1": 0.5448369193384395,
366
+ "eval_runtime": 1.3346,
367
+ "eval_samples_per_second": 516.259,
368
+ "eval_steps_per_second": 32.969,
369
+ "eval_weighted_f1": 0.5464179491770302,
370
+ "step": 1056
371
+ },
372
+ {
373
+ "epoch": 6.1082621082621085,
374
+ "grad_norm": 57.110198974609375,
375
+ "learning_rate": 1.544191919191919e-05,
376
+ "loss": 0.811,
377
+ "step": 1075
378
+ },
379
+ {
380
+ "epoch": 6.250712250712251,
381
+ "grad_norm": 56.78047561645508,
382
+ "learning_rate": 1.528409090909091e-05,
383
+ "loss": 0.4878,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.3931623931623935,
388
+ "grad_norm": 53.50935745239258,
389
+ "learning_rate": 1.5126262626262626e-05,
390
+ "loss": 0.7522,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.535612535612536,
395
+ "grad_norm": 14.805171966552734,
396
+ "learning_rate": 1.4968434343434344e-05,
397
+ "loss": 0.619,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.678062678062678,
402
+ "grad_norm": 75.20733642578125,
403
+ "learning_rate": 1.4810606060606061e-05,
404
+ "loss": 0.5579,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.82051282051282,
409
+ "grad_norm": 46.543548583984375,
410
+ "learning_rate": 1.4652777777777779e-05,
411
+ "loss": 0.6341,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.962962962962963,
416
+ "grad_norm": 97.66300201416016,
417
+ "learning_rate": 1.4494949494949494e-05,
418
+ "loss": 0.6457,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 7.0,
423
+ "eval_accuracy": 0.5703918722786647,
424
+ "eval_loss": 1.898086667060852,
425
+ "eval_macro_f1": 0.5672731392573087,
426
+ "eval_runtime": 1.7464,
427
+ "eval_samples_per_second": 394.522,
428
+ "eval_steps_per_second": 25.194,
429
+ "eval_weighted_f1": 0.5682004750295354,
430
+ "step": 1232
431
+ },
432
+ {
433
+ "epoch": 7.102564102564102,
434
+ "grad_norm": 46.35641098022461,
435
+ "learning_rate": 1.4337121212121212e-05,
436
+ "loss": 0.4373,
437
+ "step": 1250
438
+ },
439
+ {
440
+ "epoch": 7.245014245014245,
441
+ "grad_norm": 36.77285385131836,
442
+ "learning_rate": 1.4179292929292931e-05,
443
+ "loss": 0.2564,
444
+ "step": 1275
445
+ },
446
+ {
447
+ "epoch": 7.387464387464387,
448
+ "grad_norm": 105.87223815917969,
449
+ "learning_rate": 1.4021464646464649e-05,
450
+ "loss": 0.5362,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.52991452991453,
455
+ "grad_norm": 10.142257690429688,
456
+ "learning_rate": 1.3863636363636366e-05,
457
+ "loss": 0.2968,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.672364672364672,
462
+ "grad_norm": 7.8140692710876465,
463
+ "learning_rate": 1.3705808080808084e-05,
464
+ "loss": 0.4024,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.814814814814815,
469
+ "grad_norm": 99.27178192138672,
470
+ "learning_rate": 1.35479797979798e-05,
471
+ "loss": 0.2586,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.957264957264957,
476
+ "grad_norm": 84.80584716796875,
477
+ "learning_rate": 1.3390151515151517e-05,
478
+ "loss": 0.4128,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 8.0,
483
+ "eval_accuracy": 0.579100145137881,
484
+ "eval_loss": 2.4165353775024414,
485
+ "eval_macro_f1": 0.5730352518211279,
486
+ "eval_runtime": 1.4796,
487
+ "eval_samples_per_second": 465.668,
488
+ "eval_steps_per_second": 29.738,
489
+ "eval_weighted_f1": 0.5730396492373024,
490
+ "step": 1408
491
+ },
492
+ {
493
+ "epoch": 8.096866096866098,
494
+ "grad_norm": 20.056537628173828,
495
+ "learning_rate": 1.3232323232323234e-05,
496
+ "loss": 0.4127,
497
+ "step": 1425
498
+ },
499
+ {
500
+ "epoch": 8.239316239316238,
501
+ "grad_norm": 54.585140228271484,
502
+ "learning_rate": 1.3074494949494952e-05,
503
+ "loss": 0.2561,
504
+ "step": 1450
505
+ },
506
+ {
507
+ "epoch": 8.381766381766381,
508
+ "grad_norm": 24.0261287689209,
509
+ "learning_rate": 1.2916666666666668e-05,
510
+ "loss": 0.2442,
511
+ "step": 1475
512
+ },
513
+ {
514
+ "epoch": 8.524216524216524,
515
+ "grad_norm": 121.47981262207031,
516
+ "learning_rate": 1.2758838383838385e-05,
517
+ "loss": 0.2792,
518
+ "step": 1500
519
+ },
520
+ {
521
+ "epoch": 8.666666666666666,
522
+ "grad_norm": 7.25212287902832,
523
+ "learning_rate": 1.2601010101010103e-05,
524
+ "loss": 0.3819,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 8.809116809116809,
529
+ "grad_norm": 2.060002326965332,
530
+ "learning_rate": 1.244318181818182e-05,
531
+ "loss": 0.2825,
532
+ "step": 1550
533
+ },
534
+ {
535
+ "epoch": 8.951566951566951,
536
+ "grad_norm": 25.59327507019043,
537
+ "learning_rate": 1.2285353535353536e-05,
538
+ "loss": 0.1745,
539
+ "step": 1575
540
+ },
541
+ {
542
+ "epoch": 9.0,
543
+ "eval_accuracy": 0.5863570391872278,
544
+ "eval_loss": 2.7472100257873535,
545
+ "eval_macro_f1": 0.5827959365470929,
546
+ "eval_runtime": 1.3505,
547
+ "eval_samples_per_second": 510.197,
548
+ "eval_steps_per_second": 32.582,
549
+ "eval_weighted_f1": 0.5833044385279388,
550
+ "step": 1584
551
+ }
552
+ ],
553
+ "logging_steps": 25,
554
+ "max_steps": 3520,
555
+ "num_input_tokens_seen": 0,
556
+ "num_train_epochs": 20,
557
+ "save_steps": 500,
558
+ "stateful_callbacks": {
559
+ "EarlyStoppingCallback": {
560
+ "args": {
561
+ "early_stopping_patience": 3,
562
+ "early_stopping_threshold": 0.0
563
+ },
564
+ "attributes": {
565
+ "early_stopping_patience_counter": 0
566
+ }
567
+ },
568
+ "TrainerControl": {
569
+ "args": {
570
+ "should_epoch_stop": false,
571
+ "should_evaluate": false,
572
+ "should_log": false,
573
+ "should_save": true,
574
+ "should_training_stop": false
575
+ },
576
+ "attributes": {}
577
+ }
578
+ },
579
+ "total_flos": 0.0,
580
+ "train_batch_size": 8,
581
+ "trial_name": null,
582
+ "trial_params": null
583
+ }
ood-vifactcheck-visobert-source-source-04-s42-20260831T001152Z-f21e2f7e/trainer/checkpoint-2112/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-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-1504/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-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-1504/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-1504/trainer_state.json ADDED
@@ -0,0 +1,551 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1504,
3
+ "best_metric": 0.5579227129078749,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-1504",
5
+ "epoch": 8.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1504,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13351134846461948,
14
+ "grad_norm": 22.322864532470703,
15
+ "learning_rate": 1.276595744680851e-06,
16
+ "loss": 4.7569,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.26702269692923897,
21
+ "grad_norm": 18.13532257080078,
22
+ "learning_rate": 2.6063829787234047e-06,
23
+ "loss": 4.5608,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.40053404539385845,
28
+ "grad_norm": 23.731555938720703,
29
+ "learning_rate": 3.936170212765958e-06,
30
+ "loss": 4.4613,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5340453938584779,
35
+ "grad_norm": 21.85297203063965,
36
+ "learning_rate": 5.265957446808511e-06,
37
+ "loss": 4.4263,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6675567423230975,
42
+ "grad_norm": 18.258148193359375,
43
+ "learning_rate": 6.595744680851064e-06,
44
+ "loss": 4.3928,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8010680907877169,
49
+ "grad_norm": 25.962888717651367,
50
+ "learning_rate": 7.925531914893617e-06,
51
+ "loss": 4.4004,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9345794392523364,
56
+ "grad_norm": 22.468839645385742,
57
+ "learning_rate": 9.255319148936171e-06,
58
+ "loss": 4.3929,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.44292803970223327,
64
+ "eval_loss": 1.073993444442749,
65
+ "eval_macro_f1": 0.4231966681441773,
66
+ "eval_runtime": 1.8462,
67
+ "eval_samples_per_second": 436.575,
68
+ "eval_steps_per_second": 27.624,
69
+ "eval_weighted_f1": 0.4243818632598222,
70
+ "step": 188
71
+ },
72
+ {
73
+ "epoch": 1.0640854472630175,
74
+ "grad_norm": 22.531965255737305,
75
+ "learning_rate": 1.0585106382978725e-05,
76
+ "loss": 4.1537,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.1975967957276368,
81
+ "grad_norm": 27.091325759887695,
82
+ "learning_rate": 1.1914893617021277e-05,
83
+ "loss": 4.1933,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3311081441922563,
88
+ "grad_norm": 16.70467758178711,
89
+ "learning_rate": 1.3244680851063832e-05,
90
+ "loss": 4.2545,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.4646194926568759,
95
+ "grad_norm": 21.3377742767334,
96
+ "learning_rate": 1.4574468085106384e-05,
97
+ "loss": 4.2853,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.5981308411214954,
102
+ "grad_norm": 30.86119270324707,
103
+ "learning_rate": 1.590425531914894e-05,
104
+ "loss": 4.3148,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7316421895861147,
109
+ "grad_norm": 34.764469146728516,
110
+ "learning_rate": 1.723404255319149e-05,
111
+ "loss": 4.3369,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.8651535380507345,
116
+ "grad_norm": 29.580862045288086,
117
+ "learning_rate": 1.8563829787234043e-05,
118
+ "loss": 4.2851,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 1.9986648865153538,
123
+ "grad_norm": 17.38605308532715,
124
+ "learning_rate": 1.9893617021276595e-05,
125
+ "loss": 4.1906,
126
+ "step": 375
127
+ },
128
+ {
129
+ "epoch": 2.0,
130
+ "eval_accuracy": 0.45161290322580644,
131
+ "eval_loss": 1.0458306074142456,
132
+ "eval_macro_f1": 0.4529451534773988,
133
+ "eval_runtime": 1.6213,
134
+ "eval_samples_per_second": 497.146,
135
+ "eval_steps_per_second": 31.457,
136
+ "eval_weighted_f1": 0.45298079118418205,
137
+ "step": 376
138
+ },
139
+ {
140
+ "epoch": 2.128170894526035,
141
+ "grad_norm": 62.4507942199707,
142
+ "learning_rate": 1.986406619385343e-05,
143
+ "loss": 3.8,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.2616822429906542,
148
+ "grad_norm": 58.91880798339844,
149
+ "learning_rate": 1.971631205673759e-05,
150
+ "loss": 3.6124,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.3951935914552736,
155
+ "grad_norm": 39.31308364868164,
156
+ "learning_rate": 1.956855791962175e-05,
157
+ "loss": 3.8914,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.5287049399198933,
162
+ "grad_norm": 22.640705108642578,
163
+ "learning_rate": 1.9420803782505913e-05,
164
+ "loss": 3.5547,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.6622162883845126,
169
+ "grad_norm": 41.816246032714844,
170
+ "learning_rate": 1.927304964539007e-05,
171
+ "loss": 3.7866,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.795727636849132,
176
+ "grad_norm": 42.730987548828125,
177
+ "learning_rate": 1.9125295508274234e-05,
178
+ "loss": 3.72,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 2.9292389853137517,
183
+ "grad_norm": 60.83925247192383,
184
+ "learning_rate": 1.8977541371158394e-05,
185
+ "loss": 3.5783,
186
+ "step": 550
187
+ },
188
+ {
189
+ "epoch": 3.0,
190
+ "eval_accuracy": 0.5124069478908189,
191
+ "eval_loss": 0.9625744223594666,
192
+ "eval_macro_f1": 0.5097802825080134,
193
+ "eval_runtime": 1.5073,
194
+ "eval_samples_per_second": 534.744,
195
+ "eval_steps_per_second": 33.836,
196
+ "eval_weighted_f1": 0.5085575710583636,
197
+ "step": 564
198
+ },
199
+ {
200
+ "epoch": 3.0587449933244324,
201
+ "grad_norm": 88.73998260498047,
202
+ "learning_rate": 1.8829787234042557e-05,
203
+ "loss": 3.3725,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.192256341789052,
208
+ "grad_norm": 35.2221565246582,
209
+ "learning_rate": 1.8682033096926714e-05,
210
+ "loss": 2.9961,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.3257676902536715,
215
+ "grad_norm": 84.26072692871094,
216
+ "learning_rate": 1.8534278959810878e-05,
217
+ "loss": 2.9425,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.4592790387182912,
222
+ "grad_norm": 44.57492446899414,
223
+ "learning_rate": 1.8386524822695038e-05,
224
+ "loss": 3.0794,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.5927903871829105,
229
+ "grad_norm": 58.55664825439453,
230
+ "learning_rate": 1.8238770685579198e-05,
231
+ "loss": 2.9221,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.72630173564753,
236
+ "grad_norm": 30.089426040649414,
237
+ "learning_rate": 1.8091016548463358e-05,
238
+ "loss": 2.8937,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.8598130841121496,
243
+ "grad_norm": 36.28462600708008,
244
+ "learning_rate": 1.7943262411347518e-05,
245
+ "loss": 2.7413,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 3.993324432576769,
250
+ "grad_norm": 46.17044448852539,
251
+ "learning_rate": 1.779550827423168e-05,
252
+ "loss": 2.9684,
253
+ "step": 750
254
+ },
255
+ {
256
+ "epoch": 4.0,
257
+ "eval_accuracy": 0.5409429280397022,
258
+ "eval_loss": 1.0235350131988525,
259
+ "eval_macro_f1": 0.5247269846237144,
260
+ "eval_runtime": 1.7288,
261
+ "eval_samples_per_second": 466.21,
262
+ "eval_steps_per_second": 29.5,
263
+ "eval_weighted_f1": 0.5253531877039733,
264
+ "step": 752
265
+ },
266
+ {
267
+ "epoch": 4.12283044058745,
268
+ "grad_norm": 71.43463134765625,
269
+ "learning_rate": 1.7647754137115842e-05,
270
+ "loss": 2.0595,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.25634178905207,
275
+ "grad_norm": 26.903982162475586,
276
+ "learning_rate": 1.7500000000000002e-05,
277
+ "loss": 1.9419,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.389853137516689,
282
+ "grad_norm": 49.290889739990234,
283
+ "learning_rate": 1.7352245862884162e-05,
284
+ "loss": 1.8252,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.5233644859813085,
289
+ "grad_norm": 63.213871002197266,
290
+ "learning_rate": 1.7204491725768322e-05,
291
+ "loss": 1.6869,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.656875834445928,
296
+ "grad_norm": 75.79057312011719,
297
+ "learning_rate": 1.7056737588652482e-05,
298
+ "loss": 1.8438,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.790387182910547,
303
+ "grad_norm": 53.408260345458984,
304
+ "learning_rate": 1.6908983451536646e-05,
305
+ "loss": 1.9522,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 4.923898531375167,
310
+ "grad_norm": 83.96077728271484,
311
+ "learning_rate": 1.6761229314420806e-05,
312
+ "loss": 1.9863,
313
+ "step": 925
314
+ },
315
+ {
316
+ "epoch": 5.0,
317
+ "eval_accuracy": 0.5409429280397022,
318
+ "eval_loss": 1.2202266454696655,
319
+ "eval_macro_f1": 0.5262637868549809,
320
+ "eval_runtime": 1.5508,
321
+ "eval_samples_per_second": 519.725,
322
+ "eval_steps_per_second": 32.886,
323
+ "eval_weighted_f1": 0.5271264102304224,
324
+ "step": 940
325
+ },
326
+ {
327
+ "epoch": 5.0534045393858475,
328
+ "grad_norm": 87.87933349609375,
329
+ "learning_rate": 1.6613475177304966e-05,
330
+ "loss": 1.8714,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.186915887850467,
335
+ "grad_norm": 41.41051483154297,
336
+ "learning_rate": 1.6465721040189126e-05,
337
+ "loss": 1.1742,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.320427236315087,
342
+ "grad_norm": 38.732147216796875,
343
+ "learning_rate": 1.6317966903073286e-05,
344
+ "loss": 0.9856,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.453938584779706,
349
+ "grad_norm": 48.09575653076172,
350
+ "learning_rate": 1.6170212765957446e-05,
351
+ "loss": 1.2094,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.587449933244326,
356
+ "grad_norm": 46.516414642333984,
357
+ "learning_rate": 1.602245862884161e-05,
358
+ "loss": 1.0066,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.720961281708945,
363
+ "grad_norm": 60.10972595214844,
364
+ "learning_rate": 1.587470449172577e-05,
365
+ "loss": 0.9418,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 5.854472630173564,
370
+ "grad_norm": 46.39299011230469,
371
+ "learning_rate": 1.572695035460993e-05,
372
+ "loss": 1.1445,
373
+ "step": 1100
374
+ },
375
+ {
376
+ "epoch": 5.9879839786381845,
377
+ "grad_norm": 53.410823822021484,
378
+ "learning_rate": 1.557919621749409e-05,
379
+ "loss": 1.2914,
380
+ "step": 1125
381
+ },
382
+ {
383
+ "epoch": 6.0,
384
+ "eval_accuracy": 0.5570719602977667,
385
+ "eval_loss": 1.4003506898880005,
386
+ "eval_macro_f1": 0.5574671249641955,
387
+ "eval_runtime": 1.5349,
388
+ "eval_samples_per_second": 525.12,
389
+ "eval_steps_per_second": 33.227,
390
+ "eval_weighted_f1": 0.5570597249012997,
391
+ "step": 1128
392
+ },
393
+ {
394
+ "epoch": 6.117489986648865,
395
+ "grad_norm": 18.52421760559082,
396
+ "learning_rate": 1.543144208037825e-05,
397
+ "loss": 0.5062,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.251001335113485,
402
+ "grad_norm": 25.59064292907715,
403
+ "learning_rate": 1.5283687943262414e-05,
404
+ "loss": 0.6041,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.384512683578104,
409
+ "grad_norm": 40.50505447387695,
410
+ "learning_rate": 1.5135933806146574e-05,
411
+ "loss": 0.6036,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.518024032042724,
416
+ "grad_norm": 164.75254821777344,
417
+ "learning_rate": 1.4988179669030734e-05,
418
+ "loss": 0.9985,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.651535380507343,
423
+ "grad_norm": 50.93531036376953,
424
+ "learning_rate": 1.4840425531914894e-05,
425
+ "loss": 0.5222,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 6.785046728971962,
430
+ "grad_norm": 51.81645584106445,
431
+ "learning_rate": 1.4692671394799056e-05,
432
+ "loss": 0.735,
433
+ "step": 1275
434
+ },
435
+ {
436
+ "epoch": 6.9185580774365825,
437
+ "grad_norm": 95.4217300415039,
438
+ "learning_rate": 1.4544917257683215e-05,
439
+ "loss": 0.7514,
440
+ "step": 1300
441
+ },
442
+ {
443
+ "epoch": 7.0,
444
+ "eval_accuracy": 0.5459057071960298,
445
+ "eval_loss": 2.1691551208496094,
446
+ "eval_macro_f1": 0.5357504015756832,
447
+ "eval_runtime": 1.5579,
448
+ "eval_samples_per_second": 517.368,
449
+ "eval_steps_per_second": 32.737,
450
+ "eval_weighted_f1": 0.533912544691515,
451
+ "step": 1316
452
+ },
453
+ {
454
+ "epoch": 7.048064085447263,
455
+ "grad_norm": 72.32121276855469,
456
+ "learning_rate": 1.4397163120567377e-05,
457
+ "loss": 0.6422,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.181575433911883,
462
+ "grad_norm": 33.10828399658203,
463
+ "learning_rate": 1.4249408983451538e-05,
464
+ "loss": 0.299,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.315086782376502,
469
+ "grad_norm": 72.97974395751953,
470
+ "learning_rate": 1.41016548463357e-05,
471
+ "loss": 0.2696,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.4485981308411215,
476
+ "grad_norm": 47.81031036376953,
477
+ "learning_rate": 1.3953900709219859e-05,
478
+ "loss": 0.4767,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 7.582109479305741,
483
+ "grad_norm": 46.9863166809082,
484
+ "learning_rate": 1.380614657210402e-05,
485
+ "loss": 0.4954,
486
+ "step": 1425
487
+ },
488
+ {
489
+ "epoch": 7.71562082777036,
490
+ "grad_norm": 32.43349075317383,
491
+ "learning_rate": 1.365839243498818e-05,
492
+ "loss": 0.3034,
493
+ "step": 1450
494
+ },
495
+ {
496
+ "epoch": 7.8491321762349795,
497
+ "grad_norm": 75.6369857788086,
498
+ "learning_rate": 1.351063829787234e-05,
499
+ "loss": 0.4042,
500
+ "step": 1475
501
+ },
502
+ {
503
+ "epoch": 7.9826435246996,
504
+ "grad_norm": 92.12084197998047,
505
+ "learning_rate": 1.3362884160756503e-05,
506
+ "loss": 0.3452,
507
+ "step": 1500
508
+ },
509
+ {
510
+ "epoch": 8.0,
511
+ "eval_accuracy": 0.5595533498759305,
512
+ "eval_loss": 2.3556525707244873,
513
+ "eval_macro_f1": 0.5579227129078749,
514
+ "eval_runtime": 1.5222,
515
+ "eval_samples_per_second": 529.486,
516
+ "eval_steps_per_second": 33.503,
517
+ "eval_weighted_f1": 0.5566900195425303,
518
+ "step": 1504
519
+ }
520
+ ],
521
+ "logging_steps": 25,
522
+ "max_steps": 3760,
523
+ "num_input_tokens_seen": 0,
524
+ "num_train_epochs": 20,
525
+ "save_steps": 500,
526
+ "stateful_callbacks": {
527
+ "EarlyStoppingCallback": {
528
+ "args": {
529
+ "early_stopping_patience": 3,
530
+ "early_stopping_threshold": 0.0
531
+ },
532
+ "attributes": {
533
+ "early_stopping_patience_counter": 0
534
+ }
535
+ },
536
+ "TrainerControl": {
537
+ "args": {
538
+ "should_epoch_stop": false,
539
+ "should_evaluate": false,
540
+ "should_log": false,
541
+ "should_save": true,
542
+ "should_training_stop": false
543
+ },
544
+ "attributes": {}
545
+ }
546
+ },
547
+ "total_flos": 0.0,
548
+ "train_batch_size": 8,
549
+ "trial_name": null,
550
+ "trial_params": null
551
+ }
ood-vifactcheck-visobert-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-2068/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-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-2068/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-2068/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-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-2068/trainer_state.json ADDED
@@ -0,0 +1,738 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1504,
3
+ "best_metric": 0.5579227129078749,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-topic-domain-04-s44-20260830T192314Z-59e46264/trainer/checkpoint-1504",
5
+ "epoch": 11.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2068,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13351134846461948,
14
+ "grad_norm": 22.322864532470703,
15
+ "learning_rate": 1.276595744680851e-06,
16
+ "loss": 4.7569,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.26702269692923897,
21
+ "grad_norm": 18.13532257080078,
22
+ "learning_rate": 2.6063829787234047e-06,
23
+ "loss": 4.5608,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.40053404539385845,
28
+ "grad_norm": 23.731555938720703,
29
+ "learning_rate": 3.936170212765958e-06,
30
+ "loss": 4.4613,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5340453938584779,
35
+ "grad_norm": 21.85297203063965,
36
+ "learning_rate": 5.265957446808511e-06,
37
+ "loss": 4.4263,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6675567423230975,
42
+ "grad_norm": 18.258148193359375,
43
+ "learning_rate": 6.595744680851064e-06,
44
+ "loss": 4.3928,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8010680907877169,
49
+ "grad_norm": 25.962888717651367,
50
+ "learning_rate": 7.925531914893617e-06,
51
+ "loss": 4.4004,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9345794392523364,
56
+ "grad_norm": 22.468839645385742,
57
+ "learning_rate": 9.255319148936171e-06,
58
+ "loss": 4.3929,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.44292803970223327,
64
+ "eval_loss": 1.073993444442749,
65
+ "eval_macro_f1": 0.4231966681441773,
66
+ "eval_runtime": 1.8462,
67
+ "eval_samples_per_second": 436.575,
68
+ "eval_steps_per_second": 27.624,
69
+ "eval_weighted_f1": 0.4243818632598222,
70
+ "step": 188
71
+ },
72
+ {
73
+ "epoch": 1.0640854472630175,
74
+ "grad_norm": 22.531965255737305,
75
+ "learning_rate": 1.0585106382978725e-05,
76
+ "loss": 4.1537,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.1975967957276368,
81
+ "grad_norm": 27.091325759887695,
82
+ "learning_rate": 1.1914893617021277e-05,
83
+ "loss": 4.1933,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3311081441922563,
88
+ "grad_norm": 16.70467758178711,
89
+ "learning_rate": 1.3244680851063832e-05,
90
+ "loss": 4.2545,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.4646194926568759,
95
+ "grad_norm": 21.3377742767334,
96
+ "learning_rate": 1.4574468085106384e-05,
97
+ "loss": 4.2853,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.5981308411214954,
102
+ "grad_norm": 30.86119270324707,
103
+ "learning_rate": 1.590425531914894e-05,
104
+ "loss": 4.3148,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7316421895861147,
109
+ "grad_norm": 34.764469146728516,
110
+ "learning_rate": 1.723404255319149e-05,
111
+ "loss": 4.3369,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.8651535380507345,
116
+ "grad_norm": 29.580862045288086,
117
+ "learning_rate": 1.8563829787234043e-05,
118
+ "loss": 4.2851,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 1.9986648865153538,
123
+ "grad_norm": 17.38605308532715,
124
+ "learning_rate": 1.9893617021276595e-05,
125
+ "loss": 4.1906,
126
+ "step": 375
127
+ },
128
+ {
129
+ "epoch": 2.0,
130
+ "eval_accuracy": 0.45161290322580644,
131
+ "eval_loss": 1.0458306074142456,
132
+ "eval_macro_f1": 0.4529451534773988,
133
+ "eval_runtime": 1.6213,
134
+ "eval_samples_per_second": 497.146,
135
+ "eval_steps_per_second": 31.457,
136
+ "eval_weighted_f1": 0.45298079118418205,
137
+ "step": 376
138
+ },
139
+ {
140
+ "epoch": 2.128170894526035,
141
+ "grad_norm": 62.4507942199707,
142
+ "learning_rate": 1.986406619385343e-05,
143
+ "loss": 3.8,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.2616822429906542,
148
+ "grad_norm": 58.91880798339844,
149
+ "learning_rate": 1.971631205673759e-05,
150
+ "loss": 3.6124,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.3951935914552736,
155
+ "grad_norm": 39.31308364868164,
156
+ "learning_rate": 1.956855791962175e-05,
157
+ "loss": 3.8914,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.5287049399198933,
162
+ "grad_norm": 22.640705108642578,
163
+ "learning_rate": 1.9420803782505913e-05,
164
+ "loss": 3.5547,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.6622162883845126,
169
+ "grad_norm": 41.816246032714844,
170
+ "learning_rate": 1.927304964539007e-05,
171
+ "loss": 3.7866,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.795727636849132,
176
+ "grad_norm": 42.730987548828125,
177
+ "learning_rate": 1.9125295508274234e-05,
178
+ "loss": 3.72,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 2.9292389853137517,
183
+ "grad_norm": 60.83925247192383,
184
+ "learning_rate": 1.8977541371158394e-05,
185
+ "loss": 3.5783,
186
+ "step": 550
187
+ },
188
+ {
189
+ "epoch": 3.0,
190
+ "eval_accuracy": 0.5124069478908189,
191
+ "eval_loss": 0.9625744223594666,
192
+ "eval_macro_f1": 0.5097802825080134,
193
+ "eval_runtime": 1.5073,
194
+ "eval_samples_per_second": 534.744,
195
+ "eval_steps_per_second": 33.836,
196
+ "eval_weighted_f1": 0.5085575710583636,
197
+ "step": 564
198
+ },
199
+ {
200
+ "epoch": 3.0587449933244324,
201
+ "grad_norm": 88.73998260498047,
202
+ "learning_rate": 1.8829787234042557e-05,
203
+ "loss": 3.3725,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.192256341789052,
208
+ "grad_norm": 35.2221565246582,
209
+ "learning_rate": 1.8682033096926714e-05,
210
+ "loss": 2.9961,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.3257676902536715,
215
+ "grad_norm": 84.26072692871094,
216
+ "learning_rate": 1.8534278959810878e-05,
217
+ "loss": 2.9425,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.4592790387182912,
222
+ "grad_norm": 44.57492446899414,
223
+ "learning_rate": 1.8386524822695038e-05,
224
+ "loss": 3.0794,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.5927903871829105,
229
+ "grad_norm": 58.55664825439453,
230
+ "learning_rate": 1.8238770685579198e-05,
231
+ "loss": 2.9221,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.72630173564753,
236
+ "grad_norm": 30.089426040649414,
237
+ "learning_rate": 1.8091016548463358e-05,
238
+ "loss": 2.8937,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.8598130841121496,
243
+ "grad_norm": 36.28462600708008,
244
+ "learning_rate": 1.7943262411347518e-05,
245
+ "loss": 2.7413,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 3.993324432576769,
250
+ "grad_norm": 46.17044448852539,
251
+ "learning_rate": 1.779550827423168e-05,
252
+ "loss": 2.9684,
253
+ "step": 750
254
+ },
255
+ {
256
+ "epoch": 4.0,
257
+ "eval_accuracy": 0.5409429280397022,
258
+ "eval_loss": 1.0235350131988525,
259
+ "eval_macro_f1": 0.5247269846237144,
260
+ "eval_runtime": 1.7288,
261
+ "eval_samples_per_second": 466.21,
262
+ "eval_steps_per_second": 29.5,
263
+ "eval_weighted_f1": 0.5253531877039733,
264
+ "step": 752
265
+ },
266
+ {
267
+ "epoch": 4.12283044058745,
268
+ "grad_norm": 71.43463134765625,
269
+ "learning_rate": 1.7647754137115842e-05,
270
+ "loss": 2.0595,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.25634178905207,
275
+ "grad_norm": 26.903982162475586,
276
+ "learning_rate": 1.7500000000000002e-05,
277
+ "loss": 1.9419,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.389853137516689,
282
+ "grad_norm": 49.290889739990234,
283
+ "learning_rate": 1.7352245862884162e-05,
284
+ "loss": 1.8252,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.5233644859813085,
289
+ "grad_norm": 63.213871002197266,
290
+ "learning_rate": 1.7204491725768322e-05,
291
+ "loss": 1.6869,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.656875834445928,
296
+ "grad_norm": 75.79057312011719,
297
+ "learning_rate": 1.7056737588652482e-05,
298
+ "loss": 1.8438,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.790387182910547,
303
+ "grad_norm": 53.408260345458984,
304
+ "learning_rate": 1.6908983451536646e-05,
305
+ "loss": 1.9522,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 4.923898531375167,
310
+ "grad_norm": 83.96077728271484,
311
+ "learning_rate": 1.6761229314420806e-05,
312
+ "loss": 1.9863,
313
+ "step": 925
314
+ },
315
+ {
316
+ "epoch": 5.0,
317
+ "eval_accuracy": 0.5409429280397022,
318
+ "eval_loss": 1.2202266454696655,
319
+ "eval_macro_f1": 0.5262637868549809,
320
+ "eval_runtime": 1.5508,
321
+ "eval_samples_per_second": 519.725,
322
+ "eval_steps_per_second": 32.886,
323
+ "eval_weighted_f1": 0.5271264102304224,
324
+ "step": 940
325
+ },
326
+ {
327
+ "epoch": 5.0534045393858475,
328
+ "grad_norm": 87.87933349609375,
329
+ "learning_rate": 1.6613475177304966e-05,
330
+ "loss": 1.8714,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.186915887850467,
335
+ "grad_norm": 41.41051483154297,
336
+ "learning_rate": 1.6465721040189126e-05,
337
+ "loss": 1.1742,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.320427236315087,
342
+ "grad_norm": 38.732147216796875,
343
+ "learning_rate": 1.6317966903073286e-05,
344
+ "loss": 0.9856,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.453938584779706,
349
+ "grad_norm": 48.09575653076172,
350
+ "learning_rate": 1.6170212765957446e-05,
351
+ "loss": 1.2094,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.587449933244326,
356
+ "grad_norm": 46.516414642333984,
357
+ "learning_rate": 1.602245862884161e-05,
358
+ "loss": 1.0066,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.720961281708945,
363
+ "grad_norm": 60.10972595214844,
364
+ "learning_rate": 1.587470449172577e-05,
365
+ "loss": 0.9418,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 5.854472630173564,
370
+ "grad_norm": 46.39299011230469,
371
+ "learning_rate": 1.572695035460993e-05,
372
+ "loss": 1.1445,
373
+ "step": 1100
374
+ },
375
+ {
376
+ "epoch": 5.9879839786381845,
377
+ "grad_norm": 53.410823822021484,
378
+ "learning_rate": 1.557919621749409e-05,
379
+ "loss": 1.2914,
380
+ "step": 1125
381
+ },
382
+ {
383
+ "epoch": 6.0,
384
+ "eval_accuracy": 0.5570719602977667,
385
+ "eval_loss": 1.4003506898880005,
386
+ "eval_macro_f1": 0.5574671249641955,
387
+ "eval_runtime": 1.5349,
388
+ "eval_samples_per_second": 525.12,
389
+ "eval_steps_per_second": 33.227,
390
+ "eval_weighted_f1": 0.5570597249012997,
391
+ "step": 1128
392
+ },
393
+ {
394
+ "epoch": 6.117489986648865,
395
+ "grad_norm": 18.52421760559082,
396
+ "learning_rate": 1.543144208037825e-05,
397
+ "loss": 0.5062,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.251001335113485,
402
+ "grad_norm": 25.59064292907715,
403
+ "learning_rate": 1.5283687943262414e-05,
404
+ "loss": 0.6041,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.384512683578104,
409
+ "grad_norm": 40.50505447387695,
410
+ "learning_rate": 1.5135933806146574e-05,
411
+ "loss": 0.6036,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.518024032042724,
416
+ "grad_norm": 164.75254821777344,
417
+ "learning_rate": 1.4988179669030734e-05,
418
+ "loss": 0.9985,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.651535380507343,
423
+ "grad_norm": 50.93531036376953,
424
+ "learning_rate": 1.4840425531914894e-05,
425
+ "loss": 0.5222,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 6.785046728971962,
430
+ "grad_norm": 51.81645584106445,
431
+ "learning_rate": 1.4692671394799056e-05,
432
+ "loss": 0.735,
433
+ "step": 1275
434
+ },
435
+ {
436
+ "epoch": 6.9185580774365825,
437
+ "grad_norm": 95.4217300415039,
438
+ "learning_rate": 1.4544917257683215e-05,
439
+ "loss": 0.7514,
440
+ "step": 1300
441
+ },
442
+ {
443
+ "epoch": 7.0,
444
+ "eval_accuracy": 0.5459057071960298,
445
+ "eval_loss": 2.1691551208496094,
446
+ "eval_macro_f1": 0.5357504015756832,
447
+ "eval_runtime": 1.5579,
448
+ "eval_samples_per_second": 517.368,
449
+ "eval_steps_per_second": 32.737,
450
+ "eval_weighted_f1": 0.533912544691515,
451
+ "step": 1316
452
+ },
453
+ {
454
+ "epoch": 7.048064085447263,
455
+ "grad_norm": 72.32121276855469,
456
+ "learning_rate": 1.4397163120567377e-05,
457
+ "loss": 0.6422,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.181575433911883,
462
+ "grad_norm": 33.10828399658203,
463
+ "learning_rate": 1.4249408983451538e-05,
464
+ "loss": 0.299,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.315086782376502,
469
+ "grad_norm": 72.97974395751953,
470
+ "learning_rate": 1.41016548463357e-05,
471
+ "loss": 0.2696,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.4485981308411215,
476
+ "grad_norm": 47.81031036376953,
477
+ "learning_rate": 1.3953900709219859e-05,
478
+ "loss": 0.4767,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 7.582109479305741,
483
+ "grad_norm": 46.9863166809082,
484
+ "learning_rate": 1.380614657210402e-05,
485
+ "loss": 0.4954,
486
+ "step": 1425
487
+ },
488
+ {
489
+ "epoch": 7.71562082777036,
490
+ "grad_norm": 32.43349075317383,
491
+ "learning_rate": 1.365839243498818e-05,
492
+ "loss": 0.3034,
493
+ "step": 1450
494
+ },
495
+ {
496
+ "epoch": 7.8491321762349795,
497
+ "grad_norm": 75.6369857788086,
498
+ "learning_rate": 1.351063829787234e-05,
499
+ "loss": 0.4042,
500
+ "step": 1475
501
+ },
502
+ {
503
+ "epoch": 7.9826435246996,
504
+ "grad_norm": 92.12084197998047,
505
+ "learning_rate": 1.3362884160756503e-05,
506
+ "loss": 0.3452,
507
+ "step": 1500
508
+ },
509
+ {
510
+ "epoch": 8.0,
511
+ "eval_accuracy": 0.5595533498759305,
512
+ "eval_loss": 2.3556525707244873,
513
+ "eval_macro_f1": 0.5579227129078749,
514
+ "eval_runtime": 1.5222,
515
+ "eval_samples_per_second": 529.486,
516
+ "eval_steps_per_second": 33.503,
517
+ "eval_weighted_f1": 0.5566900195425303,
518
+ "step": 1504
519
+ },
520
+ {
521
+ "epoch": 8.11214953271028,
522
+ "grad_norm": 17.82720947265625,
523
+ "learning_rate": 1.3215130023640663e-05,
524
+ "loss": 0.2589,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 8.2456608811749,
529
+ "grad_norm": 26.870107650756836,
530
+ "learning_rate": 1.3067375886524825e-05,
531
+ "loss": 0.2117,
532
+ "step": 1550
533
+ },
534
+ {
535
+ "epoch": 8.379172229639519,
536
+ "grad_norm": 38.61922073364258,
537
+ "learning_rate": 1.2919621749408983e-05,
538
+ "loss": 0.2209,
539
+ "step": 1575
540
+ },
541
+ {
542
+ "epoch": 8.51268357810414,
543
+ "grad_norm": 49.68370056152344,
544
+ "learning_rate": 1.2771867612293145e-05,
545
+ "loss": 0.2849,
546
+ "step": 1600
547
+ },
548
+ {
549
+ "epoch": 8.646194926568759,
550
+ "grad_norm": 4.231113433837891,
551
+ "learning_rate": 1.2624113475177307e-05,
552
+ "loss": 0.4228,
553
+ "step": 1625
554
+ },
555
+ {
556
+ "epoch": 8.779706275033378,
557
+ "grad_norm": 15.163440704345703,
558
+ "learning_rate": 1.2476359338061469e-05,
559
+ "loss": 0.2344,
560
+ "step": 1650
561
+ },
562
+ {
563
+ "epoch": 8.913217623497998,
564
+ "grad_norm": 16.31005096435547,
565
+ "learning_rate": 1.2328605200945627e-05,
566
+ "loss": 0.2879,
567
+ "step": 1675
568
+ },
569
+ {
570
+ "epoch": 9.0,
571
+ "eval_accuracy": 0.5545905707196029,
572
+ "eval_loss": 2.810938835144043,
573
+ "eval_macro_f1": 0.5575069651354344,
574
+ "eval_runtime": 1.6364,
575
+ "eval_samples_per_second": 492.535,
576
+ "eval_steps_per_second": 31.165,
577
+ "eval_weighted_f1": 0.556745015348677,
578
+ "step": 1692
579
+ },
580
+ {
581
+ "epoch": 9.042723631508679,
582
+ "grad_norm": 6.268764495849609,
583
+ "learning_rate": 1.2180851063829789e-05,
584
+ "loss": 0.2864,
585
+ "step": 1700
586
+ },
587
+ {
588
+ "epoch": 9.176234979973298,
589
+ "grad_norm": 44.075340270996094,
590
+ "learning_rate": 1.2033096926713949e-05,
591
+ "loss": 0.1736,
592
+ "step": 1725
593
+ },
594
+ {
595
+ "epoch": 9.309746328437917,
596
+ "grad_norm": 59.280452728271484,
597
+ "learning_rate": 1.1885342789598109e-05,
598
+ "loss": 0.1837,
599
+ "step": 1750
600
+ },
601
+ {
602
+ "epoch": 9.443257676902537,
603
+ "grad_norm": 5.006645202636719,
604
+ "learning_rate": 1.1737588652482271e-05,
605
+ "loss": 0.1814,
606
+ "step": 1775
607
+ },
608
+ {
609
+ "epoch": 9.576769025367156,
610
+ "grad_norm": 41.28781509399414,
611
+ "learning_rate": 1.1589834515366431e-05,
612
+ "loss": 0.227,
613
+ "step": 1800
614
+ },
615
+ {
616
+ "epoch": 9.710280373831775,
617
+ "grad_norm": 30.240901947021484,
618
+ "learning_rate": 1.1442080378250593e-05,
619
+ "loss": 0.1593,
620
+ "step": 1825
621
+ },
622
+ {
623
+ "epoch": 9.843791722296395,
624
+ "grad_norm": 1.5607326030731201,
625
+ "learning_rate": 1.1294326241134751e-05,
626
+ "loss": 0.1842,
627
+ "step": 1850
628
+ },
629
+ {
630
+ "epoch": 9.977303070761014,
631
+ "grad_norm": 10.858011245727539,
632
+ "learning_rate": 1.1146572104018913e-05,
633
+ "loss": 0.1931,
634
+ "step": 1875
635
+ },
636
+ {
637
+ "epoch": 10.0,
638
+ "eval_accuracy": 0.5446650124069479,
639
+ "eval_loss": 3.19215989112854,
640
+ "eval_macro_f1": 0.547668856788183,
641
+ "eval_runtime": 1.5859,
642
+ "eval_samples_per_second": 508.222,
643
+ "eval_steps_per_second": 32.158,
644
+ "eval_weighted_f1": 0.5467298221313233,
645
+ "step": 1880
646
+ },
647
+ {
648
+ "epoch": 10.106809078771695,
649
+ "grad_norm": 0.5183790922164917,
650
+ "learning_rate": 1.0998817966903075e-05,
651
+ "loss": 0.1436,
652
+ "step": 1900
653
+ },
654
+ {
655
+ "epoch": 10.240320427236314,
656
+ "grad_norm": 12.756089210510254,
657
+ "learning_rate": 1.0851063829787233e-05,
658
+ "loss": 0.1176,
659
+ "step": 1925
660
+ },
661
+ {
662
+ "epoch": 10.373831775700934,
663
+ "grad_norm": 6.136720180511475,
664
+ "learning_rate": 1.0703309692671395e-05,
665
+ "loss": 0.1036,
666
+ "step": 1950
667
+ },
668
+ {
669
+ "epoch": 10.507343124165555,
670
+ "grad_norm": 9.979118347167969,
671
+ "learning_rate": 1.0555555555555557e-05,
672
+ "loss": 0.0861,
673
+ "step": 1975
674
+ },
675
+ {
676
+ "epoch": 10.640854472630174,
677
+ "grad_norm": 12.464832305908203,
678
+ "learning_rate": 1.0407801418439717e-05,
679
+ "loss": 0.1276,
680
+ "step": 2000
681
+ },
682
+ {
683
+ "epoch": 10.774365821094793,
684
+ "grad_norm": 1.2295441627502441,
685
+ "learning_rate": 1.0260047281323877e-05,
686
+ "loss": 0.0669,
687
+ "step": 2025
688
+ },
689
+ {
690
+ "epoch": 10.907877169559413,
691
+ "grad_norm": 3.5188684463500977,
692
+ "learning_rate": 1.011229314420804e-05,
693
+ "loss": 0.1316,
694
+ "step": 2050
695
+ },
696
+ {
697
+ "epoch": 11.0,
698
+ "eval_accuracy": 0.5310173697270472,
699
+ "eval_loss": 4.0017313957214355,
700
+ "eval_macro_f1": 0.5351024947991848,
701
+ "eval_runtime": 1.7521,
702
+ "eval_samples_per_second": 460.027,
703
+ "eval_steps_per_second": 29.108,
704
+ "eval_weighted_f1": 0.5349323225139989,
705
+ "step": 2068
706
+ }
707
+ ],
708
+ "logging_steps": 25,
709
+ "max_steps": 3760,
710
+ "num_input_tokens_seen": 0,
711
+ "num_train_epochs": 20,
712
+ "save_steps": 500,
713
+ "stateful_callbacks": {
714
+ "EarlyStoppingCallback": {
715
+ "args": {
716
+ "early_stopping_patience": 3,
717
+ "early_stopping_threshold": 0.0
718
+ },
719
+ "attributes": {
720
+ "early_stopping_patience_counter": 3
721
+ }
722
+ },
723
+ "TrainerControl": {
724
+ "args": {
725
+ "should_epoch_stop": false,
726
+ "should_evaluate": false,
727
+ "should_log": false,
728
+ "should_save": true,
729
+ "should_training_stop": true
730
+ },
731
+ "attributes": {}
732
+ }
733
+ },
734
+ "total_flos": 0.0,
735
+ "train_batch_size": 8,
736
+ "trial_name": null,
737
+ "trial_params": null
738
+ }
ood-vifactcheck-visobert-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1050/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-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1050/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1050/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-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1050/trainer_state.json ADDED
@@ -0,0 +1,403 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1050,
3
+ "best_metric": 0.5807931836628414,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1050",
5
+ "epoch": 6.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1050,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.1430615164520744,
14
+ "grad_norm": 31.865772247314453,
15
+ "learning_rate": 1.3714285714285717e-06,
16
+ "loss": 4.7067,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.2861230329041488,
21
+ "grad_norm": 86.9774169921875,
22
+ "learning_rate": 2.8000000000000003e-06,
23
+ "loss": 4.5275,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.4291845493562232,
28
+ "grad_norm": 22.50299072265625,
29
+ "learning_rate": 4.228571428571429e-06,
30
+ "loss": 4.4275,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5722460658082976,
35
+ "grad_norm": 33.08814239501953,
36
+ "learning_rate": 5.6571428571428576e-06,
37
+ "loss": 4.436,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.7153075822603719,
42
+ "grad_norm": 30.650859832763672,
43
+ "learning_rate": 7.085714285714286e-06,
44
+ "loss": 4.3396,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8583690987124464,
49
+ "grad_norm": 32.624080657958984,
50
+ "learning_rate": 8.514285714285715e-06,
51
+ "loss": 4.4026,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 1.0,
56
+ "grad_norm": 18.855180740356445,
57
+ "learning_rate": 9.942857142857145e-06,
58
+ "loss": 4.4149,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.46226415094339623,
64
+ "eval_loss": 1.0523526668548584,
65
+ "eval_macro_f1": 0.45822447476873207,
66
+ "eval_runtime": 1.4487,
67
+ "eval_samples_per_second": 512.188,
68
+ "eval_steps_per_second": 32.443,
69
+ "eval_weighted_f1": 0.45965687705527497,
70
+ "step": 175
71
+ },
72
+ {
73
+ "epoch": 1.1430615164520743,
74
+ "grad_norm": 18.931419372558594,
75
+ "learning_rate": 1.1371428571428573e-05,
76
+ "loss": 4.1944,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2861230329041489,
81
+ "grad_norm": 38.163333892822266,
82
+ "learning_rate": 1.2800000000000001e-05,
83
+ "loss": 4.2518,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.4291845493562232,
88
+ "grad_norm": 21.238845825195312,
89
+ "learning_rate": 1.422857142857143e-05,
90
+ "loss": 4.2895,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.5722460658082977,
95
+ "grad_norm": 20.062454223632812,
96
+ "learning_rate": 1.5657142857142856e-05,
97
+ "loss": 4.2339,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.715307582260372,
102
+ "grad_norm": 27.799701690673828,
103
+ "learning_rate": 1.708571428571429e-05,
104
+ "loss": 4.2937,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.8583690987124464,
109
+ "grad_norm": 25.488401412963867,
110
+ "learning_rate": 1.8514285714285716e-05,
111
+ "loss": 4.2281,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 2.0,
116
+ "grad_norm": 25.27716636657715,
117
+ "learning_rate": 1.9942857142857142e-05,
118
+ "loss": 4.1326,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.4838274932614555,
124
+ "eval_loss": 1.022624135017395,
125
+ "eval_macro_f1": 0.4715840610212078,
126
+ "eval_runtime": 1.391,
127
+ "eval_samples_per_second": 533.422,
128
+ "eval_steps_per_second": 33.788,
129
+ "eval_weighted_f1": 0.47309104995458273,
130
+ "step": 350
131
+ },
132
+ {
133
+ "epoch": 2.1430615164520743,
134
+ "grad_norm": 47.266258239746094,
135
+ "learning_rate": 1.984761904761905e-05,
136
+ "loss": 3.7631,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.2861230329041486,
141
+ "grad_norm": 18.24425506591797,
142
+ "learning_rate": 1.968888888888889e-05,
143
+ "loss": 3.8939,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.429184549356223,
148
+ "grad_norm": 43.297943115234375,
149
+ "learning_rate": 1.953015873015873e-05,
150
+ "loss": 3.6679,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.5722460658082977,
155
+ "grad_norm": 26.381681442260742,
156
+ "learning_rate": 1.937142857142857e-05,
157
+ "loss": 3.7516,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.715307582260372,
162
+ "grad_norm": 41.768619537353516,
163
+ "learning_rate": 1.9212698412698414e-05,
164
+ "loss": 3.9213,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.8583690987124464,
169
+ "grad_norm": 25.291807174682617,
170
+ "learning_rate": 1.9053968253968257e-05,
171
+ "loss": 3.7031,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 3.0,
176
+ "grad_norm": 23.45393180847168,
177
+ "learning_rate": 1.8895238095238096e-05,
178
+ "loss": 3.4836,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 3.0,
183
+ "eval_accuracy": 0.5404312668463612,
184
+ "eval_loss": 1.0125032663345337,
185
+ "eval_macro_f1": 0.5066229232895899,
186
+ "eval_runtime": 1.443,
187
+ "eval_samples_per_second": 514.205,
188
+ "eval_steps_per_second": 32.571,
189
+ "eval_weighted_f1": 0.5085057807429776,
190
+ "step": 525
191
+ },
192
+ {
193
+ "epoch": 3.1430615164520743,
194
+ "grad_norm": 32.272132873535156,
195
+ "learning_rate": 1.8736507936507936e-05,
196
+ "loss": 2.9949,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.2861230329041486,
201
+ "grad_norm": 36.219486236572266,
202
+ "learning_rate": 1.857777777777778e-05,
203
+ "loss": 2.7964,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.429184549356223,
208
+ "grad_norm": 97.75125122070312,
209
+ "learning_rate": 1.8419047619047622e-05,
210
+ "loss": 2.867,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.5722460658082977,
215
+ "grad_norm": 27.99545669555664,
216
+ "learning_rate": 1.8260317460317462e-05,
217
+ "loss": 2.6819,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.715307582260372,
222
+ "grad_norm": 41.776817321777344,
223
+ "learning_rate": 1.81015873015873e-05,
224
+ "loss": 2.6838,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.8583690987124464,
229
+ "grad_norm": 26.961713790893555,
230
+ "learning_rate": 1.7942857142857144e-05,
231
+ "loss": 2.8288,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 4.0,
236
+ "grad_norm": 53.057899475097656,
237
+ "learning_rate": 1.7784126984126987e-05,
238
+ "loss": 2.8631,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 4.0,
243
+ "eval_accuracy": 0.555256064690027,
244
+ "eval_loss": 1.0714473724365234,
245
+ "eval_macro_f1": 0.5396360367825455,
246
+ "eval_runtime": 1.5886,
247
+ "eval_samples_per_second": 467.081,
248
+ "eval_steps_per_second": 29.586,
249
+ "eval_weighted_f1": 0.5416144525627564,
250
+ "step": 700
251
+ },
252
+ {
253
+ "epoch": 4.143061516452074,
254
+ "grad_norm": 44.307342529296875,
255
+ "learning_rate": 1.7625396825396827e-05,
256
+ "loss": 1.9982,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.286123032904149,
261
+ "grad_norm": 73.86514282226562,
262
+ "learning_rate": 1.7466666666666667e-05,
263
+ "loss": 1.7347,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.429184549356223,
268
+ "grad_norm": 62.523658752441406,
269
+ "learning_rate": 1.730793650793651e-05,
270
+ "loss": 1.6416,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.572246065808297,
275
+ "grad_norm": 59.32464599609375,
276
+ "learning_rate": 1.7149206349206353e-05,
277
+ "loss": 1.7604,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.715307582260372,
282
+ "grad_norm": 82.62734985351562,
283
+ "learning_rate": 1.6990476190476192e-05,
284
+ "loss": 1.9825,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.858369098712446,
289
+ "grad_norm": 38.8407096862793,
290
+ "learning_rate": 1.6831746031746032e-05,
291
+ "loss": 2.0389,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 5.0,
296
+ "grad_norm": 46.659053802490234,
297
+ "learning_rate": 1.6673015873015875e-05,
298
+ "loss": 1.7559,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 5.0,
303
+ "eval_accuracy": 0.5579514824797843,
304
+ "eval_loss": 1.1615017652511597,
305
+ "eval_macro_f1": 0.5574309934847261,
306
+ "eval_runtime": 1.6032,
307
+ "eval_samples_per_second": 462.824,
308
+ "eval_steps_per_second": 29.316,
309
+ "eval_weighted_f1": 0.5579945046954642,
310
+ "step": 875
311
+ },
312
+ {
313
+ "epoch": 5.143061516452074,
314
+ "grad_norm": 88.76544952392578,
315
+ "learning_rate": 1.6514285714285714e-05,
316
+ "loss": 1.3796,
317
+ "step": 900
318
+ },
319
+ {
320
+ "epoch": 5.286123032904149,
321
+ "grad_norm": 31.617660522460938,
322
+ "learning_rate": 1.6355555555555557e-05,
323
+ "loss": 1.1088,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.429184549356223,
328
+ "grad_norm": 38.45769119262695,
329
+ "learning_rate": 1.6196825396825397e-05,
330
+ "loss": 1.1409,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.572246065808297,
335
+ "grad_norm": 38.00873565673828,
336
+ "learning_rate": 1.603809523809524e-05,
337
+ "loss": 0.9732,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.715307582260372,
342
+ "grad_norm": 103.0644302368164,
343
+ "learning_rate": 1.587936507936508e-05,
344
+ "loss": 0.9979,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.858369098712446,
349
+ "grad_norm": 74.31143188476562,
350
+ "learning_rate": 1.5720634920634923e-05,
351
+ "loss": 1.1588,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 6.0,
356
+ "grad_norm": 52.96892547607422,
357
+ "learning_rate": 1.5561904761904762e-05,
358
+ "loss": 0.8392,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 6.0,
363
+ "eval_accuracy": 0.5795148247978437,
364
+ "eval_loss": 1.5431890487670898,
365
+ "eval_macro_f1": 0.5807931836628414,
366
+ "eval_runtime": 1.4761,
367
+ "eval_samples_per_second": 502.663,
368
+ "eval_steps_per_second": 31.84,
369
+ "eval_weighted_f1": 0.5807431132955138,
370
+ "step": 1050
371
+ }
372
+ ],
373
+ "logging_steps": 25,
374
+ "max_steps": 3500,
375
+ "num_input_tokens_seen": 0,
376
+ "num_train_epochs": 20,
377
+ "save_steps": 500,
378
+ "stateful_callbacks": {
379
+ "EarlyStoppingCallback": {
380
+ "args": {
381
+ "early_stopping_patience": 3,
382
+ "early_stopping_threshold": 0.0
383
+ },
384
+ "attributes": {
385
+ "early_stopping_patience_counter": 0
386
+ }
387
+ },
388
+ "TrainerControl": {
389
+ "args": {
390
+ "should_epoch_stop": false,
391
+ "should_evaluate": false,
392
+ "should_log": false,
393
+ "should_save": true,
394
+ "should_training_stop": false
395
+ },
396
+ "attributes": {}
397
+ }
398
+ },
399
+ "total_flos": 0.0,
400
+ "train_batch_size": 8,
401
+ "trial_name": null,
402
+ "trial_params": null
403
+ }
ood-vifactcheck-visobert-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1575/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-topic-domain-05-s44-20260830T193529Z-59e46264/trainer/checkpoint-1575/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff