PhaaNe commited on
Commit
1a1acf8
·
verified ·
1 Parent(s): 904a87b

Add files using upload-large-folder tool

Browse files
Files changed (20) hide show
  1. ood-vifactcheck-visobert-source-source-06-s42-20260831T003231Z-f21e2f7e/trainer/checkpoint-2035/special_tokens_map.json +15 -0
  2. ood-vifactcheck-visobert-source-source-06-s42-20260831T003231Z-f21e2f7e/trainer/checkpoint-2035/tokenizer.json +0 -0
  3. ood-vifactcheck-visobert-source-source-06-s42-20260831T003231Z-f21e2f7e/trainer/checkpoint-2035/tokenizer_config.json +56 -0
  4. ood-vifactcheck-visobert-source-source-06-s42-20260831T003231Z-f21e2f7e/trainer/checkpoint-2035/trainer_state.json +731 -0
  5. ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1110/special_tokens_map.json +15 -0
  6. ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1110/tokenizer.json +0 -0
  7. ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1110/tokenizer_config.json +56 -0
  8. ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1110/trainer_state.json +417 -0
  9. ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1665/special_tokens_map.json +15 -0
  10. ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1665/tokenizer.json +0 -0
  11. ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1665/tokenizer_config.json +56 -0
  12. ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1665/trainer_state.json +604 -0
  13. ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1140/special_tokens_map.json +15 -0
  14. ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1140/tokenizer.json +0 -0
  15. ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1140/tokenizer_config.json +56 -0
  16. ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1140/trainer_state.json +424 -0
  17. ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1710/special_tokens_map.json +15 -0
  18. ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1710/tokenizer.json +0 -0
  19. ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1710/tokenizer_config.json +56 -0
  20. ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1710/trainer_state.json +618 -0
ood-vifactcheck-visobert-source-source-06-s42-20260831T003231Z-f21e2f7e/trainer/checkpoint-2035/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-06-s42-20260831T003231Z-f21e2f7e/trainer/checkpoint-2035/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-06-s42-20260831T003231Z-f21e2f7e/trainer/checkpoint-2035/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-06-s42-20260831T003231Z-f21e2f7e/trainer/checkpoint-2035/trainer_state.json ADDED
@@ -0,0 +1,731 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1480,
3
+ "best_metric": 0.58596362354947,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-06-s42-20260831T003231Z-f21e2f7e/trainer/checkpoint-1480",
5
+ "epoch": 11.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2035,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13550135501355012,
14
+ "grad_norm": 20.357709884643555,
15
+ "learning_rate": 1.2972972972972974e-06,
16
+ "loss": 4.6716,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.27100271002710025,
21
+ "grad_norm": 25.524555206298828,
22
+ "learning_rate": 2.648648648648649e-06,
23
+ "loss": 4.4931,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.4065040650406504,
28
+ "grad_norm": 30.60616111755371,
29
+ "learning_rate": 4.000000000000001e-06,
30
+ "loss": 4.451,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5420054200542005,
35
+ "grad_norm": 28.622480392456055,
36
+ "learning_rate": 5.351351351351351e-06,
37
+ "loss": 4.4259,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6775067750677507,
42
+ "grad_norm": 27.390562057495117,
43
+ "learning_rate": 6.702702702702704e-06,
44
+ "loss": 4.4619,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8130081300813008,
49
+ "grad_norm": 28.70516586303711,
50
+ "learning_rate": 8.054054054054055e-06,
51
+ "loss": 4.3448,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.948509485094851,
56
+ "grad_norm": 20.170974731445312,
57
+ "learning_rate": 9.405405405405407e-06,
58
+ "loss": 4.4515,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.4159891598915989,
64
+ "eval_loss": 1.0661125183105469,
65
+ "eval_macro_f1": 0.3927685716277139,
66
+ "eval_runtime": 1.5313,
67
+ "eval_samples_per_second": 481.955,
68
+ "eval_steps_per_second": 30.694,
69
+ "eval_weighted_f1": 0.3954841313234317,
70
+ "step": 185
71
+ },
72
+ {
73
+ "epoch": 1.08130081300813,
74
+ "grad_norm": 19.588459014892578,
75
+ "learning_rate": 1.0756756756756757e-05,
76
+ "loss": 4.194,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2168021680216803,
81
+ "grad_norm": 22.15048599243164,
82
+ "learning_rate": 1.210810810810811e-05,
83
+ "loss": 4.3269,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3523035230352303,
88
+ "grad_norm": 20.58547592163086,
89
+ "learning_rate": 1.3459459459459461e-05,
90
+ "loss": 4.1278,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.4878048780487805,
95
+ "grad_norm": 47.9154052734375,
96
+ "learning_rate": 1.4810810810810811e-05,
97
+ "loss": 4.2197,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6233062330623307,
102
+ "grad_norm": 26.602832794189453,
103
+ "learning_rate": 1.6162162162162163e-05,
104
+ "loss": 4.1971,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.758807588075881,
109
+ "grad_norm": 16.841196060180664,
110
+ "learning_rate": 1.7513513513513517e-05,
111
+ "loss": 4.1891,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.8943089430894309,
116
+ "grad_norm": 20.704299926757812,
117
+ "learning_rate": 1.8864864864864866e-05,
118
+ "loss": 4.3157,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.39295392953929537,
124
+ "eval_loss": 1.1406210660934448,
125
+ "eval_macro_f1": 0.33649005900799467,
126
+ "eval_runtime": 1.5201,
127
+ "eval_samples_per_second": 485.486,
128
+ "eval_steps_per_second": 30.918,
129
+ "eval_weighted_f1": 0.3303605175364009,
130
+ "step": 370
131
+ },
132
+ {
133
+ "epoch": 2.02710027100271,
134
+ "grad_norm": 23.664655685424805,
135
+ "learning_rate": 1.9975975975975978e-05,
136
+ "loss": 3.9594,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.16260162601626,
141
+ "grad_norm": 31.630964279174805,
142
+ "learning_rate": 1.9825825825825828e-05,
143
+ "loss": 3.7926,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.2981029810298104,
148
+ "grad_norm": 46.86356735229492,
149
+ "learning_rate": 1.967567567567568e-05,
150
+ "loss": 3.797,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.4336043360433606,
155
+ "grad_norm": 35.555259704589844,
156
+ "learning_rate": 1.9525525525525525e-05,
157
+ "loss": 3.6979,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.569105691056911,
162
+ "grad_norm": 21.405046463012695,
163
+ "learning_rate": 1.9375375375375376e-05,
164
+ "loss": 3.5943,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.7046070460704605,
169
+ "grad_norm": 44.80155944824219,
170
+ "learning_rate": 1.9225225225225226e-05,
171
+ "loss": 3.4968,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.8401084010840107,
176
+ "grad_norm": 51.13383483886719,
177
+ "learning_rate": 1.9075075075075076e-05,
178
+ "loss": 3.7023,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 2.975609756097561,
183
+ "grad_norm": 22.87540626525879,
184
+ "learning_rate": 1.8924924924924926e-05,
185
+ "loss": 3.6388,
186
+ "step": 550
187
+ },
188
+ {
189
+ "epoch": 3.0,
190
+ "eval_accuracy": 0.5365853658536586,
191
+ "eval_loss": 0.9346652030944824,
192
+ "eval_macro_f1": 0.5358602523969763,
193
+ "eval_runtime": 1.4809,
194
+ "eval_samples_per_second": 498.353,
195
+ "eval_steps_per_second": 31.738,
196
+ "eval_weighted_f1": 0.5349633509782014,
197
+ "step": 555
198
+ },
199
+ {
200
+ "epoch": 3.1084010840108403,
201
+ "grad_norm": 46.90874481201172,
202
+ "learning_rate": 1.8774774774774776e-05,
203
+ "loss": 2.9077,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.2439024390243905,
208
+ "grad_norm": 63.010780334472656,
209
+ "learning_rate": 1.8624624624624627e-05,
210
+ "loss": 2.7618,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.3794037940379402,
215
+ "grad_norm": 28.250125885009766,
216
+ "learning_rate": 1.8474474474474477e-05,
217
+ "loss": 2.7634,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.5149051490514904,
222
+ "grad_norm": 43.522178649902344,
223
+ "learning_rate": 1.8324324324324324e-05,
224
+ "loss": 2.965,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.6504065040650406,
229
+ "grad_norm": 44.616695404052734,
230
+ "learning_rate": 1.8174174174174174e-05,
231
+ "loss": 2.7232,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.785907859078591,
236
+ "grad_norm": 64.59855651855469,
237
+ "learning_rate": 1.8024024024024024e-05,
238
+ "loss": 2.7811,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.921409214092141,
243
+ "grad_norm": 48.08187484741211,
244
+ "learning_rate": 1.7873873873873874e-05,
245
+ "loss": 2.7656,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 4.0,
250
+ "eval_accuracy": 0.5718157181571816,
251
+ "eval_loss": 0.9289685487747192,
252
+ "eval_macro_f1": 0.5694014592940978,
253
+ "eval_runtime": 1.6006,
254
+ "eval_samples_per_second": 461.087,
255
+ "eval_steps_per_second": 29.365,
256
+ "eval_weighted_f1": 0.5699293802086568,
257
+ "step": 740
258
+ },
259
+ {
260
+ "epoch": 4.05420054200542,
261
+ "grad_norm": 37.47281265258789,
262
+ "learning_rate": 1.7723723723723724e-05,
263
+ "loss": 2.4828,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.1897018970189706,
268
+ "grad_norm": 50.268280029296875,
269
+ "learning_rate": 1.7573573573573575e-05,
270
+ "loss": 1.8789,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.32520325203252,
275
+ "grad_norm": 49.67667007446289,
276
+ "learning_rate": 1.7423423423423425e-05,
277
+ "loss": 2.0543,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.46070460704607,
282
+ "grad_norm": 43.392860412597656,
283
+ "learning_rate": 1.7273273273273275e-05,
284
+ "loss": 1.9123,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.596205962059621,
289
+ "grad_norm": 69.00615692138672,
290
+ "learning_rate": 1.7123123123123125e-05,
291
+ "loss": 1.8992,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.7317073170731705,
296
+ "grad_norm": 38.3200569152832,
297
+ "learning_rate": 1.6972972972972975e-05,
298
+ "loss": 1.9871,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.867208672086721,
303
+ "grad_norm": 62.512229919433594,
304
+ "learning_rate": 1.6822822822822826e-05,
305
+ "loss": 1.9823,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 5.0,
310
+ "grad_norm": 22.450998306274414,
311
+ "learning_rate": 1.6672672672672673e-05,
312
+ "loss": 2.102,
313
+ "step": 925
314
+ },
315
+ {
316
+ "epoch": 5.0,
317
+ "eval_accuracy": 0.5772357723577236,
318
+ "eval_loss": 1.02829909324646,
319
+ "eval_macro_f1": 0.5762566137566137,
320
+ "eval_runtime": 1.5422,
321
+ "eval_samples_per_second": 478.545,
322
+ "eval_steps_per_second": 30.476,
323
+ "eval_weighted_f1": 0.5769282057899944,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.13550135501355,
328
+ "grad_norm": 50.8524169921875,
329
+ "learning_rate": 1.6522522522522523e-05,
330
+ "loss": 1.2148,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.2710027100271,
335
+ "grad_norm": 29.982929229736328,
336
+ "learning_rate": 1.6372372372372373e-05,
337
+ "loss": 1.1953,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.40650406504065,
342
+ "grad_norm": 44.76668930053711,
343
+ "learning_rate": 1.6222222222222223e-05,
344
+ "loss": 1.0117,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.542005420054201,
349
+ "grad_norm": 70.34624481201172,
350
+ "learning_rate": 1.6072072072072073e-05,
351
+ "loss": 1.2226,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.677506775067751,
356
+ "grad_norm": 90.42357635498047,
357
+ "learning_rate": 1.5921921921921924e-05,
358
+ "loss": 1.0973,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.8130081300813,
363
+ "grad_norm": 79.31895446777344,
364
+ "learning_rate": 1.5771771771771774e-05,
365
+ "loss": 1.2688,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 5.948509485094851,
370
+ "grad_norm": 70.91158294677734,
371
+ "learning_rate": 1.5621621621621624e-05,
372
+ "loss": 1.1066,
373
+ "step": 1100
374
+ },
375
+ {
376
+ "epoch": 6.0,
377
+ "eval_accuracy": 0.5704607046070461,
378
+ "eval_loss": 1.3678768873214722,
379
+ "eval_macro_f1": 0.5727491315228105,
380
+ "eval_runtime": 1.5767,
381
+ "eval_samples_per_second": 468.062,
382
+ "eval_steps_per_second": 29.809,
383
+ "eval_weighted_f1": 0.5731608927877845,
384
+ "step": 1110
385
+ },
386
+ {
387
+ "epoch": 6.08130081300813,
388
+ "grad_norm": 25.119922637939453,
389
+ "learning_rate": 1.5471471471471474e-05,
390
+ "loss": 0.9272,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.2168021680216805,
395
+ "grad_norm": 79.80097961425781,
396
+ "learning_rate": 1.5321321321321324e-05,
397
+ "loss": 0.4529,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.35230352303523,
402
+ "grad_norm": 58.90565872192383,
403
+ "learning_rate": 1.5171171171171173e-05,
404
+ "loss": 0.624,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.487804878048781,
409
+ "grad_norm": 47.078086853027344,
410
+ "learning_rate": 1.5021021021021023e-05,
411
+ "loss": 0.8501,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.623306233062331,
416
+ "grad_norm": 77.01166534423828,
417
+ "learning_rate": 1.4870870870870873e-05,
418
+ "loss": 0.5812,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.7588075880758804,
423
+ "grad_norm": 43.48472213745117,
424
+ "learning_rate": 1.4720720720720722e-05,
425
+ "loss": 0.8517,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 6.894308943089431,
430
+ "grad_norm": 21.808334350585938,
431
+ "learning_rate": 1.4570570570570572e-05,
432
+ "loss": 0.5582,
433
+ "step": 1275
434
+ },
435
+ {
436
+ "epoch": 7.0,
437
+ "eval_accuracy": 0.5623306233062331,
438
+ "eval_loss": 1.7512412071228027,
439
+ "eval_macro_f1": 0.5648117316305371,
440
+ "eval_runtime": 1.559,
441
+ "eval_samples_per_second": 473.381,
442
+ "eval_steps_per_second": 30.148,
443
+ "eval_weighted_f1": 0.5644118241533768,
444
+ "step": 1295
445
+ },
446
+ {
447
+ "epoch": 7.02710027100271,
448
+ "grad_norm": 15.119675636291504,
449
+ "learning_rate": 1.4420420420420422e-05,
450
+ "loss": 0.6254,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.16260162601626,
455
+ "grad_norm": 31.58051300048828,
456
+ "learning_rate": 1.4270270270270272e-05,
457
+ "loss": 0.3593,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.29810298102981,
462
+ "grad_norm": 47.139869689941406,
463
+ "learning_rate": 1.4120120120120123e-05,
464
+ "loss": 0.3728,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.43360433604336,
469
+ "grad_norm": 42.259700775146484,
470
+ "learning_rate": 1.3969969969969971e-05,
471
+ "loss": 0.3558,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.569105691056911,
476
+ "grad_norm": 43.38686752319336,
477
+ "learning_rate": 1.3819819819819821e-05,
478
+ "loss": 0.401,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 7.7046070460704605,
483
+ "grad_norm": 9.104645729064941,
484
+ "learning_rate": 1.3669669669669672e-05,
485
+ "loss": 0.3715,
486
+ "step": 1425
487
+ },
488
+ {
489
+ "epoch": 7.840108401084011,
490
+ "grad_norm": 24.254037857055664,
491
+ "learning_rate": 1.3519519519519522e-05,
492
+ "loss": 0.3486,
493
+ "step": 1450
494
+ },
495
+ {
496
+ "epoch": 7.975609756097561,
497
+ "grad_norm": 62.172889709472656,
498
+ "learning_rate": 1.3369369369369369e-05,
499
+ "loss": 0.331,
500
+ "step": 1475
501
+ },
502
+ {
503
+ "epoch": 8.0,
504
+ "eval_accuracy": 0.5880758807588076,
505
+ "eval_loss": 2.284409999847412,
506
+ "eval_macro_f1": 0.58596362354947,
507
+ "eval_runtime": 1.5115,
508
+ "eval_samples_per_second": 488.244,
509
+ "eval_steps_per_second": 31.094,
510
+ "eval_weighted_f1": 0.5861948528770883,
511
+ "step": 1480
512
+ },
513
+ {
514
+ "epoch": 8.10840108401084,
515
+ "grad_norm": 24.42830467224121,
516
+ "learning_rate": 1.3219219219219219e-05,
517
+ "loss": 0.3302,
518
+ "step": 1500
519
+ },
520
+ {
521
+ "epoch": 8.24390243902439,
522
+ "grad_norm": 15.898283004760742,
523
+ "learning_rate": 1.3069069069069069e-05,
524
+ "loss": 0.2846,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 8.379403794037941,
529
+ "grad_norm": 68.95182037353516,
530
+ "learning_rate": 1.291891891891892e-05,
531
+ "loss": 0.2291,
532
+ "step": 1550
533
+ },
534
+ {
535
+ "epoch": 8.51490514905149,
536
+ "grad_norm": 52.14080810546875,
537
+ "learning_rate": 1.276876876876877e-05,
538
+ "loss": 0.308,
539
+ "step": 1575
540
+ },
541
+ {
542
+ "epoch": 8.65040650406504,
543
+ "grad_norm": 84.54850769042969,
544
+ "learning_rate": 1.2618618618618618e-05,
545
+ "loss": 0.2172,
546
+ "step": 1600
547
+ },
548
+ {
549
+ "epoch": 8.785907859078591,
550
+ "grad_norm": 70.29634094238281,
551
+ "learning_rate": 1.2468468468468468e-05,
552
+ "loss": 0.2516,
553
+ "step": 1625
554
+ },
555
+ {
556
+ "epoch": 8.92140921409214,
557
+ "grad_norm": 87.83186340332031,
558
+ "learning_rate": 1.2318318318318318e-05,
559
+ "loss": 0.2679,
560
+ "step": 1650
561
+ },
562
+ {
563
+ "epoch": 9.0,
564
+ "eval_accuracy": 0.5826558265582655,
565
+ "eval_loss": 2.6146693229675293,
566
+ "eval_macro_f1": 0.5834660256518099,
567
+ "eval_runtime": 1.5295,
568
+ "eval_samples_per_second": 482.526,
569
+ "eval_steps_per_second": 30.73,
570
+ "eval_weighted_f1": 0.5823490452376827,
571
+ "step": 1665
572
+ },
573
+ {
574
+ "epoch": 9.05420054200542,
575
+ "grad_norm": 4.580981731414795,
576
+ "learning_rate": 1.2168168168168169e-05,
577
+ "loss": 0.2569,
578
+ "step": 1675
579
+ },
580
+ {
581
+ "epoch": 9.18970189701897,
582
+ "grad_norm": 6.3205060958862305,
583
+ "learning_rate": 1.2018018018018019e-05,
584
+ "loss": 0.2589,
585
+ "step": 1700
586
+ },
587
+ {
588
+ "epoch": 9.32520325203252,
589
+ "grad_norm": 1.3349652290344238,
590
+ "learning_rate": 1.1867867867867869e-05,
591
+ "loss": 0.0668,
592
+ "step": 1725
593
+ },
594
+ {
595
+ "epoch": 9.46070460704607,
596
+ "grad_norm": 58.42913055419922,
597
+ "learning_rate": 1.1717717717717718e-05,
598
+ "loss": 0.1604,
599
+ "step": 1750
600
+ },
601
+ {
602
+ "epoch": 9.59620596205962,
603
+ "grad_norm": 3.445613145828247,
604
+ "learning_rate": 1.1567567567567568e-05,
605
+ "loss": 0.2974,
606
+ "step": 1775
607
+ },
608
+ {
609
+ "epoch": 9.731707317073171,
610
+ "grad_norm": 1.0963624715805054,
611
+ "learning_rate": 1.1417417417417418e-05,
612
+ "loss": 0.1685,
613
+ "step": 1800
614
+ },
615
+ {
616
+ "epoch": 9.86720867208672,
617
+ "grad_norm": 5.246860027313232,
618
+ "learning_rate": 1.1267267267267268e-05,
619
+ "loss": 0.3934,
620
+ "step": 1825
621
+ },
622
+ {
623
+ "epoch": 10.0,
624
+ "grad_norm": 0.009339019656181335,
625
+ "learning_rate": 1.1117117117117118e-05,
626
+ "loss": 0.304,
627
+ "step": 1850
628
+ },
629
+ {
630
+ "epoch": 10.0,
631
+ "eval_accuracy": 0.5867208672086721,
632
+ "eval_loss": 3.1766602993011475,
633
+ "eval_macro_f1": 0.5673695406979895,
634
+ "eval_runtime": 1.5904,
635
+ "eval_samples_per_second": 464.037,
636
+ "eval_steps_per_second": 29.553,
637
+ "eval_weighted_f1": 0.5677827965632406,
638
+ "step": 1850
639
+ },
640
+ {
641
+ "epoch": 10.13550135501355,
642
+ "grad_norm": 11.394216537475586,
643
+ "learning_rate": 1.0966966966966967e-05,
644
+ "loss": 0.1469,
645
+ "step": 1875
646
+ },
647
+ {
648
+ "epoch": 10.2710027100271,
649
+ "grad_norm": 7.130479335784912,
650
+ "learning_rate": 1.0816816816816817e-05,
651
+ "loss": 0.1506,
652
+ "step": 1900
653
+ },
654
+ {
655
+ "epoch": 10.40650406504065,
656
+ "grad_norm": 52.193145751953125,
657
+ "learning_rate": 1.0666666666666667e-05,
658
+ "loss": 0.2976,
659
+ "step": 1925
660
+ },
661
+ {
662
+ "epoch": 10.5420054200542,
663
+ "grad_norm": 3.1116085052490234,
664
+ "learning_rate": 1.0516516516516518e-05,
665
+ "loss": 0.1234,
666
+ "step": 1950
667
+ },
668
+ {
669
+ "epoch": 10.677506775067751,
670
+ "grad_norm": 2.1214003562927246,
671
+ "learning_rate": 1.0366366366366368e-05,
672
+ "loss": 0.1801,
673
+ "step": 1975
674
+ },
675
+ {
676
+ "epoch": 10.8130081300813,
677
+ "grad_norm": 58.83681869506836,
678
+ "learning_rate": 1.0216216216216216e-05,
679
+ "loss": 0.1299,
680
+ "step": 2000
681
+ },
682
+ {
683
+ "epoch": 10.948509485094851,
684
+ "grad_norm": 54.762779235839844,
685
+ "learning_rate": 1.0066066066066066e-05,
686
+ "loss": 0.3486,
687
+ "step": 2025
688
+ },
689
+ {
690
+ "epoch": 11.0,
691
+ "eval_accuracy": 0.5650406504065041,
692
+ "eval_loss": 3.4116227626800537,
693
+ "eval_macro_f1": 0.5688591972795952,
694
+ "eval_runtime": 1.4425,
695
+ "eval_samples_per_second": 511.61,
696
+ "eval_steps_per_second": 32.582,
697
+ "eval_weighted_f1": 0.5684347383827626,
698
+ "step": 2035
699
+ }
700
+ ],
701
+ "logging_steps": 25,
702
+ "max_steps": 3700,
703
+ "num_input_tokens_seen": 0,
704
+ "num_train_epochs": 20,
705
+ "save_steps": 500,
706
+ "stateful_callbacks": {
707
+ "EarlyStoppingCallback": {
708
+ "args": {
709
+ "early_stopping_patience": 3,
710
+ "early_stopping_threshold": 0.0
711
+ },
712
+ "attributes": {
713
+ "early_stopping_patience_counter": 3
714
+ }
715
+ },
716
+ "TrainerControl": {
717
+ "args": {
718
+ "should_epoch_stop": false,
719
+ "should_evaluate": false,
720
+ "should_log": false,
721
+ "should_save": true,
722
+ "should_training_stop": true
723
+ },
724
+ "attributes": {}
725
+ }
726
+ },
727
+ "total_flos": 0.0,
728
+ "train_batch_size": 8,
729
+ "trial_name": null,
730
+ "trial_params": null
731
+ }
ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1110/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-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1110/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1110/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-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1110/trainer_state.json ADDED
@@ -0,0 +1,417 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1110,
3
+ "best_metric": 0.5790352429467537,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1110",
5
+ "epoch": 6.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1110,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13513513513513514,
14
+ "grad_norm": 21.80800437927246,
15
+ "learning_rate": 1.2972972972972974e-06,
16
+ "loss": 4.6896,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.2702702702702703,
21
+ "grad_norm": 33.09139633178711,
22
+ "learning_rate": 2.648648648648649e-06,
23
+ "loss": 4.5582,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.40540540540540543,
28
+ "grad_norm": 19.163700103759766,
29
+ "learning_rate": 4.000000000000001e-06,
30
+ "loss": 4.4798,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5405405405405406,
35
+ "grad_norm": 19.12827491760254,
36
+ "learning_rate": 5.351351351351351e-06,
37
+ "loss": 4.4969,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6756756756756757,
42
+ "grad_norm": 34.387027740478516,
43
+ "learning_rate": 6.702702702702704e-06,
44
+ "loss": 4.4008,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8108108108108109,
49
+ "grad_norm": 23.83281898498535,
50
+ "learning_rate": 8.054054054054055e-06,
51
+ "loss": 4.3947,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9459459459459459,
56
+ "grad_norm": 21.702045440673828,
57
+ "learning_rate": 9.405405405405407e-06,
58
+ "loss": 4.2853,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.42833876221498374,
64
+ "eval_loss": 1.0765249729156494,
65
+ "eval_macro_f1": 0.4109943790186335,
66
+ "eval_runtime": 1.2316,
67
+ "eval_samples_per_second": 498.52,
68
+ "eval_steps_per_second": 31.665,
69
+ "eval_weighted_f1": 0.4117320803658142,
70
+ "step": 185
71
+ },
72
+ {
73
+ "epoch": 1.0810810810810811,
74
+ "grad_norm": 19.39019775390625,
75
+ "learning_rate": 1.0756756756756757e-05,
76
+ "loss": 4.3126,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2162162162162162,
81
+ "grad_norm": 39.22252655029297,
82
+ "learning_rate": 1.210810810810811e-05,
83
+ "loss": 4.2413,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3513513513513513,
88
+ "grad_norm": 31.64686393737793,
89
+ "learning_rate": 1.3459459459459461e-05,
90
+ "loss": 4.3433,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.4864864864864864,
95
+ "grad_norm": 16.056344985961914,
96
+ "learning_rate": 1.4810810810810811e-05,
97
+ "loss": 4.2563,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6216216216216215,
102
+ "grad_norm": 17.412839889526367,
103
+ "learning_rate": 1.6162162162162163e-05,
104
+ "loss": 4.1386,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7567567567567568,
109
+ "grad_norm": 35.531944274902344,
110
+ "learning_rate": 1.7513513513513517e-05,
111
+ "loss": 4.1201,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.8918918918918919,
116
+ "grad_norm": 26.272279739379883,
117
+ "learning_rate": 1.8864864864864866e-05,
118
+ "loss": 4.134,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.4820846905537459,
124
+ "eval_loss": 1.0206652879714966,
125
+ "eval_macro_f1": 0.4835394720899408,
126
+ "eval_runtime": 1.1729,
127
+ "eval_samples_per_second": 523.491,
128
+ "eval_steps_per_second": 33.251,
129
+ "eval_weighted_f1": 0.4835540324008778,
130
+ "step": 370
131
+ },
132
+ {
133
+ "epoch": 2.027027027027027,
134
+ "grad_norm": 45.97647476196289,
135
+ "learning_rate": 1.9975975975975978e-05,
136
+ "loss": 3.8674,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.1621621621621623,
141
+ "grad_norm": 17.95907974243164,
142
+ "learning_rate": 1.9825825825825828e-05,
143
+ "loss": 3.7436,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.2972972972972974,
148
+ "grad_norm": 60.36297607421875,
149
+ "learning_rate": 1.967567567567568e-05,
150
+ "loss": 3.7165,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.4324324324324325,
155
+ "grad_norm": 25.39438247680664,
156
+ "learning_rate": 1.9525525525525525e-05,
157
+ "loss": 3.5782,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.5675675675675675,
162
+ "grad_norm": 25.655742645263672,
163
+ "learning_rate": 1.9375375375375376e-05,
164
+ "loss": 3.6404,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.7027027027027026,
169
+ "grad_norm": 29.982927322387695,
170
+ "learning_rate": 1.9225225225225226e-05,
171
+ "loss": 3.6584,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.8378378378378377,
176
+ "grad_norm": 43.7540168762207,
177
+ "learning_rate": 1.9075075075075076e-05,
178
+ "loss": 3.4003,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 2.972972972972973,
183
+ "grad_norm": 33.52782440185547,
184
+ "learning_rate": 1.8924924924924926e-05,
185
+ "loss": 3.5571,
186
+ "step": 550
187
+ },
188
+ {
189
+ "epoch": 3.0,
190
+ "eval_accuracy": 0.5488599348534202,
191
+ "eval_loss": 0.9722922444343567,
192
+ "eval_macro_f1": 0.5440383530840297,
193
+ "eval_runtime": 1.1844,
194
+ "eval_samples_per_second": 518.403,
195
+ "eval_steps_per_second": 32.928,
196
+ "eval_weighted_f1": 0.5431619366643288,
197
+ "step": 555
198
+ },
199
+ {
200
+ "epoch": 3.108108108108108,
201
+ "grad_norm": 21.29052734375,
202
+ "learning_rate": 1.8774774774774776e-05,
203
+ "loss": 3.0971,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.2432432432432434,
208
+ "grad_norm": 56.64242935180664,
209
+ "learning_rate": 1.8624624624624627e-05,
210
+ "loss": 2.8645,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.3783783783783785,
215
+ "grad_norm": 36.68960189819336,
216
+ "learning_rate": 1.8474474474474477e-05,
217
+ "loss": 2.5796,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.5135135135135136,
222
+ "grad_norm": 34.251216888427734,
223
+ "learning_rate": 1.8324324324324324e-05,
224
+ "loss": 2.645,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.6486486486486487,
229
+ "grad_norm": 42.04529571533203,
230
+ "learning_rate": 1.8174174174174174e-05,
231
+ "loss": 3.1192,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.7837837837837838,
236
+ "grad_norm": 59.89650344848633,
237
+ "learning_rate": 1.8024024024024024e-05,
238
+ "loss": 2.6678,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.918918918918919,
243
+ "grad_norm": 24.378015518188477,
244
+ "learning_rate": 1.7873873873873874e-05,
245
+ "loss": 2.8659,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 4.0,
250
+ "eval_accuracy": 0.5342019543973942,
251
+ "eval_loss": 1.0221750736236572,
252
+ "eval_macro_f1": 0.5319167397093548,
253
+ "eval_runtime": 1.1516,
254
+ "eval_samples_per_second": 533.177,
255
+ "eval_steps_per_second": 33.866,
256
+ "eval_weighted_f1": 0.5316951729033266,
257
+ "step": 740
258
+ },
259
+ {
260
+ "epoch": 4.054054054054054,
261
+ "grad_norm": 52.451446533203125,
262
+ "learning_rate": 1.7723723723723724e-05,
263
+ "loss": 2.3238,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.1891891891891895,
268
+ "grad_norm": 31.256471633911133,
269
+ "learning_rate": 1.7573573573573575e-05,
270
+ "loss": 1.8392,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.324324324324325,
275
+ "grad_norm": 54.85249710083008,
276
+ "learning_rate": 1.7423423423423425e-05,
277
+ "loss": 2.1036,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.45945945945946,
282
+ "grad_norm": 45.86796569824219,
283
+ "learning_rate": 1.7273273273273275e-05,
284
+ "loss": 1.6843,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.594594594594595,
289
+ "grad_norm": 34.514923095703125,
290
+ "learning_rate": 1.7123123123123125e-05,
291
+ "loss": 1.9909,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.72972972972973,
296
+ "grad_norm": 38.91522979736328,
297
+ "learning_rate": 1.6972972972972975e-05,
298
+ "loss": 1.7835,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.864864864864865,
303
+ "grad_norm": 31.840133666992188,
304
+ "learning_rate": 1.6822822822822826e-05,
305
+ "loss": 1.9938,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 5.0,
310
+ "grad_norm": 37.549171447753906,
311
+ "learning_rate": 1.6672672672672673e-05,
312
+ "loss": 2.0138,
313
+ "step": 925
314
+ },
315
+ {
316
+ "epoch": 5.0,
317
+ "eval_accuracy": 0.5456026058631922,
318
+ "eval_loss": 1.1865817308425903,
319
+ "eval_macro_f1": 0.545136534421813,
320
+ "eval_runtime": 1.2099,
321
+ "eval_samples_per_second": 507.461,
322
+ "eval_steps_per_second": 32.233,
323
+ "eval_weighted_f1": 0.5445096411799859,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.135135135135135,
328
+ "grad_norm": 27.859661102294922,
329
+ "learning_rate": 1.6522522522522523e-05,
330
+ "loss": 1.0769,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.27027027027027,
335
+ "grad_norm": 82.78522491455078,
336
+ "learning_rate": 1.6372372372372373e-05,
337
+ "loss": 1.2243,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.405405405405405,
342
+ "grad_norm": 82.5791015625,
343
+ "learning_rate": 1.6222222222222223e-05,
344
+ "loss": 1.0198,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.54054054054054,
349
+ "grad_norm": 46.39576721191406,
350
+ "learning_rate": 1.6072072072072073e-05,
351
+ "loss": 1.0908,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.675675675675675,
356
+ "grad_norm": 147.5823516845703,
357
+ "learning_rate": 1.5921921921921924e-05,
358
+ "loss": 1.2331,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.8108108108108105,
363
+ "grad_norm": 123.68859100341797,
364
+ "learning_rate": 1.5771771771771774e-05,
365
+ "loss": 1.2519,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 5.945945945945946,
370
+ "grad_norm": 60.0346565246582,
371
+ "learning_rate": 1.5621621621621624e-05,
372
+ "loss": 1.11,
373
+ "step": 1100
374
+ },
375
+ {
376
+ "epoch": 6.0,
377
+ "eval_accuracy": 0.5781758957654723,
378
+ "eval_loss": 1.4552263021469116,
379
+ "eval_macro_f1": 0.5790352429467537,
380
+ "eval_runtime": 1.2569,
381
+ "eval_samples_per_second": 488.515,
382
+ "eval_steps_per_second": 31.029,
383
+ "eval_weighted_f1": 0.5785963416653497,
384
+ "step": 1110
385
+ }
386
+ ],
387
+ "logging_steps": 25,
388
+ "max_steps": 3700,
389
+ "num_input_tokens_seen": 0,
390
+ "num_train_epochs": 20,
391
+ "save_steps": 500,
392
+ "stateful_callbacks": {
393
+ "EarlyStoppingCallback": {
394
+ "args": {
395
+ "early_stopping_patience": 3,
396
+ "early_stopping_threshold": 0.0
397
+ },
398
+ "attributes": {
399
+ "early_stopping_patience_counter": 0
400
+ }
401
+ },
402
+ "TrainerControl": {
403
+ "args": {
404
+ "should_epoch_stop": false,
405
+ "should_evaluate": false,
406
+ "should_log": false,
407
+ "should_save": true,
408
+ "should_training_stop": false
409
+ },
410
+ "attributes": {}
411
+ }
412
+ },
413
+ "total_flos": 0.0,
414
+ "train_batch_size": 8,
415
+ "trial_name": null,
416
+ "trial_params": null
417
+ }
ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1665/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-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1665/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1665/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-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1665/trainer_state.json ADDED
@@ -0,0 +1,604 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1110,
3
+ "best_metric": 0.5790352429467537,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-07-s42-20260831T004433Z-f21e2f7e/trainer/checkpoint-1110",
5
+ "epoch": 9.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1665,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13513513513513514,
14
+ "grad_norm": 21.80800437927246,
15
+ "learning_rate": 1.2972972972972974e-06,
16
+ "loss": 4.6896,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.2702702702702703,
21
+ "grad_norm": 33.09139633178711,
22
+ "learning_rate": 2.648648648648649e-06,
23
+ "loss": 4.5582,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.40540540540540543,
28
+ "grad_norm": 19.163700103759766,
29
+ "learning_rate": 4.000000000000001e-06,
30
+ "loss": 4.4798,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5405405405405406,
35
+ "grad_norm": 19.12827491760254,
36
+ "learning_rate": 5.351351351351351e-06,
37
+ "loss": 4.4969,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6756756756756757,
42
+ "grad_norm": 34.387027740478516,
43
+ "learning_rate": 6.702702702702704e-06,
44
+ "loss": 4.4008,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.8108108108108109,
49
+ "grad_norm": 23.83281898498535,
50
+ "learning_rate": 8.054054054054055e-06,
51
+ "loss": 4.3947,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.9459459459459459,
56
+ "grad_norm": 21.702045440673828,
57
+ "learning_rate": 9.405405405405407e-06,
58
+ "loss": 4.2853,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.42833876221498374,
64
+ "eval_loss": 1.0765249729156494,
65
+ "eval_macro_f1": 0.4109943790186335,
66
+ "eval_runtime": 1.2316,
67
+ "eval_samples_per_second": 498.52,
68
+ "eval_steps_per_second": 31.665,
69
+ "eval_weighted_f1": 0.4117320803658142,
70
+ "step": 185
71
+ },
72
+ {
73
+ "epoch": 1.0810810810810811,
74
+ "grad_norm": 19.39019775390625,
75
+ "learning_rate": 1.0756756756756757e-05,
76
+ "loss": 4.3126,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.2162162162162162,
81
+ "grad_norm": 39.22252655029297,
82
+ "learning_rate": 1.210810810810811e-05,
83
+ "loss": 4.2413,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.3513513513513513,
88
+ "grad_norm": 31.64686393737793,
89
+ "learning_rate": 1.3459459459459461e-05,
90
+ "loss": 4.3433,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.4864864864864864,
95
+ "grad_norm": 16.056344985961914,
96
+ "learning_rate": 1.4810810810810811e-05,
97
+ "loss": 4.2563,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.6216216216216215,
102
+ "grad_norm": 17.412839889526367,
103
+ "learning_rate": 1.6162162162162163e-05,
104
+ "loss": 4.1386,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7567567567567568,
109
+ "grad_norm": 35.531944274902344,
110
+ "learning_rate": 1.7513513513513517e-05,
111
+ "loss": 4.1201,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.8918918918918919,
116
+ "grad_norm": 26.272279739379883,
117
+ "learning_rate": 1.8864864864864866e-05,
118
+ "loss": 4.134,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 2.0,
123
+ "eval_accuracy": 0.4820846905537459,
124
+ "eval_loss": 1.0206652879714966,
125
+ "eval_macro_f1": 0.4835394720899408,
126
+ "eval_runtime": 1.1729,
127
+ "eval_samples_per_second": 523.491,
128
+ "eval_steps_per_second": 33.251,
129
+ "eval_weighted_f1": 0.4835540324008778,
130
+ "step": 370
131
+ },
132
+ {
133
+ "epoch": 2.027027027027027,
134
+ "grad_norm": 45.97647476196289,
135
+ "learning_rate": 1.9975975975975978e-05,
136
+ "loss": 3.8674,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.1621621621621623,
141
+ "grad_norm": 17.95907974243164,
142
+ "learning_rate": 1.9825825825825828e-05,
143
+ "loss": 3.7436,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.2972972972972974,
148
+ "grad_norm": 60.36297607421875,
149
+ "learning_rate": 1.967567567567568e-05,
150
+ "loss": 3.7165,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.4324324324324325,
155
+ "grad_norm": 25.39438247680664,
156
+ "learning_rate": 1.9525525525525525e-05,
157
+ "loss": 3.5782,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.5675675675675675,
162
+ "grad_norm": 25.655742645263672,
163
+ "learning_rate": 1.9375375375375376e-05,
164
+ "loss": 3.6404,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.7027027027027026,
169
+ "grad_norm": 29.982927322387695,
170
+ "learning_rate": 1.9225225225225226e-05,
171
+ "loss": 3.6584,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.8378378378378377,
176
+ "grad_norm": 43.7540168762207,
177
+ "learning_rate": 1.9075075075075076e-05,
178
+ "loss": 3.4003,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 2.972972972972973,
183
+ "grad_norm": 33.52782440185547,
184
+ "learning_rate": 1.8924924924924926e-05,
185
+ "loss": 3.5571,
186
+ "step": 550
187
+ },
188
+ {
189
+ "epoch": 3.0,
190
+ "eval_accuracy": 0.5488599348534202,
191
+ "eval_loss": 0.9722922444343567,
192
+ "eval_macro_f1": 0.5440383530840297,
193
+ "eval_runtime": 1.1844,
194
+ "eval_samples_per_second": 518.403,
195
+ "eval_steps_per_second": 32.928,
196
+ "eval_weighted_f1": 0.5431619366643288,
197
+ "step": 555
198
+ },
199
+ {
200
+ "epoch": 3.108108108108108,
201
+ "grad_norm": 21.29052734375,
202
+ "learning_rate": 1.8774774774774776e-05,
203
+ "loss": 3.0971,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.2432432432432434,
208
+ "grad_norm": 56.64242935180664,
209
+ "learning_rate": 1.8624624624624627e-05,
210
+ "loss": 2.8645,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.3783783783783785,
215
+ "grad_norm": 36.68960189819336,
216
+ "learning_rate": 1.8474474474474477e-05,
217
+ "loss": 2.5796,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.5135135135135136,
222
+ "grad_norm": 34.251216888427734,
223
+ "learning_rate": 1.8324324324324324e-05,
224
+ "loss": 2.645,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.6486486486486487,
229
+ "grad_norm": 42.04529571533203,
230
+ "learning_rate": 1.8174174174174174e-05,
231
+ "loss": 3.1192,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.7837837837837838,
236
+ "grad_norm": 59.89650344848633,
237
+ "learning_rate": 1.8024024024024024e-05,
238
+ "loss": 2.6678,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.918918918918919,
243
+ "grad_norm": 24.378015518188477,
244
+ "learning_rate": 1.7873873873873874e-05,
245
+ "loss": 2.8659,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 4.0,
250
+ "eval_accuracy": 0.5342019543973942,
251
+ "eval_loss": 1.0221750736236572,
252
+ "eval_macro_f1": 0.5319167397093548,
253
+ "eval_runtime": 1.1516,
254
+ "eval_samples_per_second": 533.177,
255
+ "eval_steps_per_second": 33.866,
256
+ "eval_weighted_f1": 0.5316951729033266,
257
+ "step": 740
258
+ },
259
+ {
260
+ "epoch": 4.054054054054054,
261
+ "grad_norm": 52.451446533203125,
262
+ "learning_rate": 1.7723723723723724e-05,
263
+ "loss": 2.3238,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.1891891891891895,
268
+ "grad_norm": 31.256471633911133,
269
+ "learning_rate": 1.7573573573573575e-05,
270
+ "loss": 1.8392,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.324324324324325,
275
+ "grad_norm": 54.85249710083008,
276
+ "learning_rate": 1.7423423423423425e-05,
277
+ "loss": 2.1036,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.45945945945946,
282
+ "grad_norm": 45.86796569824219,
283
+ "learning_rate": 1.7273273273273275e-05,
284
+ "loss": 1.6843,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.594594594594595,
289
+ "grad_norm": 34.514923095703125,
290
+ "learning_rate": 1.7123123123123125e-05,
291
+ "loss": 1.9909,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.72972972972973,
296
+ "grad_norm": 38.91522979736328,
297
+ "learning_rate": 1.6972972972972975e-05,
298
+ "loss": 1.7835,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.864864864864865,
303
+ "grad_norm": 31.840133666992188,
304
+ "learning_rate": 1.6822822822822826e-05,
305
+ "loss": 1.9938,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 5.0,
310
+ "grad_norm": 37.549171447753906,
311
+ "learning_rate": 1.6672672672672673e-05,
312
+ "loss": 2.0138,
313
+ "step": 925
314
+ },
315
+ {
316
+ "epoch": 5.0,
317
+ "eval_accuracy": 0.5456026058631922,
318
+ "eval_loss": 1.1865817308425903,
319
+ "eval_macro_f1": 0.545136534421813,
320
+ "eval_runtime": 1.2099,
321
+ "eval_samples_per_second": 507.461,
322
+ "eval_steps_per_second": 32.233,
323
+ "eval_weighted_f1": 0.5445096411799859,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.135135135135135,
328
+ "grad_norm": 27.859661102294922,
329
+ "learning_rate": 1.6522522522522523e-05,
330
+ "loss": 1.0769,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.27027027027027,
335
+ "grad_norm": 82.78522491455078,
336
+ "learning_rate": 1.6372372372372373e-05,
337
+ "loss": 1.2243,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.405405405405405,
342
+ "grad_norm": 82.5791015625,
343
+ "learning_rate": 1.6222222222222223e-05,
344
+ "loss": 1.0198,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.54054054054054,
349
+ "grad_norm": 46.39576721191406,
350
+ "learning_rate": 1.6072072072072073e-05,
351
+ "loss": 1.0908,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.675675675675675,
356
+ "grad_norm": 147.5823516845703,
357
+ "learning_rate": 1.5921921921921924e-05,
358
+ "loss": 1.2331,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.8108108108108105,
363
+ "grad_norm": 123.68859100341797,
364
+ "learning_rate": 1.5771771771771774e-05,
365
+ "loss": 1.2519,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 5.945945945945946,
370
+ "grad_norm": 60.0346565246582,
371
+ "learning_rate": 1.5621621621621624e-05,
372
+ "loss": 1.11,
373
+ "step": 1100
374
+ },
375
+ {
376
+ "epoch": 6.0,
377
+ "eval_accuracy": 0.5781758957654723,
378
+ "eval_loss": 1.4552263021469116,
379
+ "eval_macro_f1": 0.5790352429467537,
380
+ "eval_runtime": 1.2569,
381
+ "eval_samples_per_second": 488.515,
382
+ "eval_steps_per_second": 31.029,
383
+ "eval_weighted_f1": 0.5785963416653497,
384
+ "step": 1110
385
+ },
386
+ {
387
+ "epoch": 6.081081081081081,
388
+ "grad_norm": 29.71009635925293,
389
+ "learning_rate": 1.5471471471471474e-05,
390
+ "loss": 0.7726,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 6.216216216216216,
395
+ "grad_norm": 56.28390121459961,
396
+ "learning_rate": 1.5321321321321324e-05,
397
+ "loss": 0.6517,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.351351351351352,
402
+ "grad_norm": 35.8258171081543,
403
+ "learning_rate": 1.5171171171171173e-05,
404
+ "loss": 0.6698,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.486486486486487,
409
+ "grad_norm": 59.16859436035156,
410
+ "learning_rate": 1.5021021021021023e-05,
411
+ "loss": 0.606,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.621621621621622,
416
+ "grad_norm": 54.234947204589844,
417
+ "learning_rate": 1.4870870870870873e-05,
418
+ "loss": 1.0303,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.756756756756757,
423
+ "grad_norm": 95.7240982055664,
424
+ "learning_rate": 1.4720720720720722e-05,
425
+ "loss": 0.8509,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 6.891891891891892,
430
+ "grad_norm": 112.42892456054688,
431
+ "learning_rate": 1.4570570570570572e-05,
432
+ "loss": 0.8145,
433
+ "step": 1275
434
+ },
435
+ {
436
+ "epoch": 7.0,
437
+ "eval_accuracy": 0.5472312703583062,
438
+ "eval_loss": 1.90389084815979,
439
+ "eval_macro_f1": 0.5385594723103413,
440
+ "eval_runtime": 1.2193,
441
+ "eval_samples_per_second": 503.579,
442
+ "eval_steps_per_second": 31.986,
443
+ "eval_weighted_f1": 0.5377525498257008,
444
+ "step": 1295
445
+ },
446
+ {
447
+ "epoch": 7.027027027027027,
448
+ "grad_norm": 53.15543746948242,
449
+ "learning_rate": 1.4420420420420422e-05,
450
+ "loss": 0.7282,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 7.162162162162162,
455
+ "grad_norm": 28.10799789428711,
456
+ "learning_rate": 1.4270270270270272e-05,
457
+ "loss": 0.4029,
458
+ "step": 1325
459
+ },
460
+ {
461
+ "epoch": 7.297297297297297,
462
+ "grad_norm": 15.045949935913086,
463
+ "learning_rate": 1.4120120120120123e-05,
464
+ "loss": 0.3345,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.4324324324324325,
469
+ "grad_norm": 51.5257453918457,
470
+ "learning_rate": 1.3969969969969971e-05,
471
+ "loss": 0.4979,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.5675675675675675,
476
+ "grad_norm": 112.37361907958984,
477
+ "learning_rate": 1.3819819819819821e-05,
478
+ "loss": 0.5292,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 7.702702702702703,
483
+ "grad_norm": 38.81922149658203,
484
+ "learning_rate": 1.3669669669669672e-05,
485
+ "loss": 0.5107,
486
+ "step": 1425
487
+ },
488
+ {
489
+ "epoch": 7.837837837837838,
490
+ "grad_norm": 41.259605407714844,
491
+ "learning_rate": 1.3519519519519522e-05,
492
+ "loss": 0.545,
493
+ "step": 1450
494
+ },
495
+ {
496
+ "epoch": 7.972972972972973,
497
+ "grad_norm": 12.271411895751953,
498
+ "learning_rate": 1.3369369369369369e-05,
499
+ "loss": 0.399,
500
+ "step": 1475
501
+ },
502
+ {
503
+ "epoch": 8.0,
504
+ "eval_accuracy": 0.5472312703583062,
505
+ "eval_loss": 2.2575490474700928,
506
+ "eval_macro_f1": 0.5509190263288625,
507
+ "eval_runtime": 1.1668,
508
+ "eval_samples_per_second": 526.208,
509
+ "eval_steps_per_second": 33.424,
510
+ "eval_weighted_f1": 0.5504081543489885,
511
+ "step": 1480
512
+ },
513
+ {
514
+ "epoch": 8.108108108108109,
515
+ "grad_norm": 91.93668365478516,
516
+ "learning_rate": 1.3219219219219219e-05,
517
+ "loss": 0.2564,
518
+ "step": 1500
519
+ },
520
+ {
521
+ "epoch": 8.243243243243244,
522
+ "grad_norm": 139.20042419433594,
523
+ "learning_rate": 1.3069069069069069e-05,
524
+ "loss": 0.4057,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 8.378378378378379,
529
+ "grad_norm": 12.768433570861816,
530
+ "learning_rate": 1.291891891891892e-05,
531
+ "loss": 0.3897,
532
+ "step": 1550
533
+ },
534
+ {
535
+ "epoch": 8.513513513513514,
536
+ "grad_norm": 55.428565979003906,
537
+ "learning_rate": 1.276876876876877e-05,
538
+ "loss": 0.2585,
539
+ "step": 1575
540
+ },
541
+ {
542
+ "epoch": 8.64864864864865,
543
+ "grad_norm": 4.836165428161621,
544
+ "learning_rate": 1.2618618618618618e-05,
545
+ "loss": 0.1912,
546
+ "step": 1600
547
+ },
548
+ {
549
+ "epoch": 8.783783783783784,
550
+ "grad_norm": 39.113346099853516,
551
+ "learning_rate": 1.2468468468468468e-05,
552
+ "loss": 0.2446,
553
+ "step": 1625
554
+ },
555
+ {
556
+ "epoch": 8.91891891891892,
557
+ "grad_norm": 30.387699127197266,
558
+ "learning_rate": 1.2318318318318318e-05,
559
+ "loss": 0.336,
560
+ "step": 1650
561
+ },
562
+ {
563
+ "epoch": 9.0,
564
+ "eval_accuracy": 0.5439739413680782,
565
+ "eval_loss": 2.8290627002716064,
566
+ "eval_macro_f1": 0.5454531729745379,
567
+ "eval_runtime": 1.2189,
568
+ "eval_samples_per_second": 503.741,
569
+ "eval_steps_per_second": 31.997,
570
+ "eval_weighted_f1": 0.5450242798509424,
571
+ "step": 1665
572
+ }
573
+ ],
574
+ "logging_steps": 25,
575
+ "max_steps": 3700,
576
+ "num_input_tokens_seen": 0,
577
+ "num_train_epochs": 20,
578
+ "save_steps": 500,
579
+ "stateful_callbacks": {
580
+ "EarlyStoppingCallback": {
581
+ "args": {
582
+ "early_stopping_patience": 3,
583
+ "early_stopping_threshold": 0.0
584
+ },
585
+ "attributes": {
586
+ "early_stopping_patience_counter": 3
587
+ }
588
+ },
589
+ "TrainerControl": {
590
+ "args": {
591
+ "should_epoch_stop": false,
592
+ "should_evaluate": false,
593
+ "should_log": false,
594
+ "should_save": true,
595
+ "should_training_stop": true
596
+ },
597
+ "attributes": {}
598
+ }
599
+ },
600
+ "total_flos": 0.0,
601
+ "train_batch_size": 8,
602
+ "trial_name": null,
603
+ "trial_params": null
604
+ }
ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1140/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-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1140/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1140/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-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1140/trainer_state.json ADDED
@@ -0,0 +1,424 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1140,
3
+ "best_metric": 0.576171588391268,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1140",
5
+ "epoch": 6.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1140,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13175230566534915,
14
+ "grad_norm": 19.988004684448242,
15
+ "learning_rate": 1.2631578947368422e-06,
16
+ "loss": 4.6825,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.2635046113306983,
21
+ "grad_norm": 25.267366409301758,
22
+ "learning_rate": 2.578947368421053e-06,
23
+ "loss": 4.5142,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.3952569169960474,
28
+ "grad_norm": 18.971866607666016,
29
+ "learning_rate": 3.894736842105263e-06,
30
+ "loss": 4.4316,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5270092226613966,
35
+ "grad_norm": 23.528650283813477,
36
+ "learning_rate": 5.210526315789474e-06,
37
+ "loss": 4.3666,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6587615283267457,
42
+ "grad_norm": 17.08769416809082,
43
+ "learning_rate": 6.526315789473685e-06,
44
+ "loss": 4.3749,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.7905138339920948,
49
+ "grad_norm": 19.257701873779297,
50
+ "learning_rate": 7.842105263157895e-06,
51
+ "loss": 4.3829,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.922266139657444,
56
+ "grad_norm": 17.924440383911133,
57
+ "learning_rate": 9.157894736842105e-06,
58
+ "loss": 4.3979,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.46326836581709147,
64
+ "eval_loss": 1.0485528707504272,
65
+ "eval_macro_f1": 0.43630370745575897,
66
+ "eval_runtime": 1.4707,
67
+ "eval_samples_per_second": 453.533,
68
+ "eval_steps_per_second": 28.558,
69
+ "eval_weighted_f1": 0.4382310074294181,
70
+ "step": 190
71
+ },
72
+ {
73
+ "epoch": 1.0527009222661396,
74
+ "grad_norm": 20.339780807495117,
75
+ "learning_rate": 1.0473684210526316e-05,
76
+ "loss": 4.2093,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.1844532279314888,
81
+ "grad_norm": 20.283437728881836,
82
+ "learning_rate": 1.1789473684210527e-05,
83
+ "loss": 4.1858,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.316205533596838,
88
+ "grad_norm": 30.267011642456055,
89
+ "learning_rate": 1.310526315789474e-05,
90
+ "loss": 4.1999,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.447957839262187,
95
+ "grad_norm": 23.36249542236328,
96
+ "learning_rate": 1.4421052631578948e-05,
97
+ "loss": 4.2122,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.5797101449275361,
102
+ "grad_norm": 26.577882766723633,
103
+ "learning_rate": 1.573684210526316e-05,
104
+ "loss": 4.3122,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7114624505928853,
109
+ "grad_norm": 34.899383544921875,
110
+ "learning_rate": 1.705263157894737e-05,
111
+ "loss": 4.3825,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.8432147562582344,
116
+ "grad_norm": 28.601293563842773,
117
+ "learning_rate": 1.836842105263158e-05,
118
+ "loss": 4.1944,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 1.9749670619235835,
123
+ "grad_norm": 22.760112762451172,
124
+ "learning_rate": 1.968421052631579e-05,
125
+ "loss": 4.2169,
126
+ "step": 375
127
+ },
128
+ {
129
+ "epoch": 2.0,
130
+ "eval_accuracy": 0.4617691154422789,
131
+ "eval_loss": 1.0304769277572632,
132
+ "eval_macro_f1": 0.4374997164954299,
133
+ "eval_runtime": 1.4433,
134
+ "eval_samples_per_second": 462.121,
135
+ "eval_steps_per_second": 29.099,
136
+ "eval_weighted_f1": 0.4377734994515398,
137
+ "step": 380
138
+ },
139
+ {
140
+ "epoch": 2.1054018445322793,
141
+ "grad_norm": 31.96175193786621,
142
+ "learning_rate": 1.988888888888889e-05,
143
+ "loss": 3.8482,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.2371541501976284,
148
+ "grad_norm": 41.56166076660156,
149
+ "learning_rate": 1.9742690058479533e-05,
150
+ "loss": 3.6964,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.3689064558629775,
155
+ "grad_norm": 21.94202995300293,
156
+ "learning_rate": 1.9596491228070177e-05,
157
+ "loss": 3.8187,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.5006587615283267,
162
+ "grad_norm": 38.86528015136719,
163
+ "learning_rate": 1.945029239766082e-05,
164
+ "loss": 3.69,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.632411067193676,
169
+ "grad_norm": 30.323911666870117,
170
+ "learning_rate": 1.9304093567251464e-05,
171
+ "loss": 3.7303,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.764163372859025,
176
+ "grad_norm": 47.82787322998047,
177
+ "learning_rate": 1.9157894736842108e-05,
178
+ "loss": 3.5243,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 2.895915678524374,
183
+ "grad_norm": 26.83512306213379,
184
+ "learning_rate": 1.901169590643275e-05,
185
+ "loss": 3.728,
186
+ "step": 550
187
+ },
188
+ {
189
+ "epoch": 3.0,
190
+ "eval_accuracy": 0.5172413793103449,
191
+ "eval_loss": 1.0077382326126099,
192
+ "eval_macro_f1": 0.508957142495257,
193
+ "eval_runtime": 1.3587,
194
+ "eval_samples_per_second": 490.915,
195
+ "eval_steps_per_second": 30.912,
196
+ "eval_weighted_f1": 0.5101501689469062,
197
+ "step": 570
198
+ },
199
+ {
200
+ "epoch": 3.02635046113307,
201
+ "grad_norm": 44.640281677246094,
202
+ "learning_rate": 1.8865497076023395e-05,
203
+ "loss": 3.407,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.158102766798419,
208
+ "grad_norm": 32.64724349975586,
209
+ "learning_rate": 1.871929824561404e-05,
210
+ "loss": 2.8667,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.289855072463768,
215
+ "grad_norm": 60.53120422363281,
216
+ "learning_rate": 1.857309941520468e-05,
217
+ "loss": 2.8141,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.421607378129117,
222
+ "grad_norm": 43.83180236816406,
223
+ "learning_rate": 1.8426900584795323e-05,
224
+ "loss": 2.7859,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.5533596837944663,
229
+ "grad_norm": 33.522403717041016,
230
+ "learning_rate": 1.8280701754385966e-05,
231
+ "loss": 2.9015,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.6851119894598154,
236
+ "grad_norm": 26.346681594848633,
237
+ "learning_rate": 1.813450292397661e-05,
238
+ "loss": 2.8719,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.8168642951251646,
243
+ "grad_norm": 40.303890228271484,
244
+ "learning_rate": 1.7988304093567253e-05,
245
+ "loss": 2.8547,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 3.9486166007905137,
250
+ "grad_norm": 89.58963775634766,
251
+ "learning_rate": 1.7842105263157897e-05,
252
+ "loss": 2.6832,
253
+ "step": 750
254
+ },
255
+ {
256
+ "epoch": 4.0,
257
+ "eval_accuracy": 0.5112443778110944,
258
+ "eval_loss": 1.2015987634658813,
259
+ "eval_macro_f1": 0.4963979167713946,
260
+ "eval_runtime": 1.3221,
261
+ "eval_samples_per_second": 504.51,
262
+ "eval_steps_per_second": 31.768,
263
+ "eval_weighted_f1": 0.497864935907554,
264
+ "step": 760
265
+ },
266
+ {
267
+ "epoch": 4.07905138339921,
268
+ "grad_norm": 44.19499969482422,
269
+ "learning_rate": 1.769590643274854e-05,
270
+ "loss": 2.5193,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.210803689064559,
275
+ "grad_norm": 29.569128036499023,
276
+ "learning_rate": 1.7549707602339184e-05,
277
+ "loss": 1.877,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.342555994729908,
282
+ "grad_norm": 67.0626449584961,
283
+ "learning_rate": 1.7403508771929828e-05,
284
+ "loss": 1.7765,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.474308300395257,
289
+ "grad_norm": 38.277992248535156,
290
+ "learning_rate": 1.725730994152047e-05,
291
+ "loss": 1.7627,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.606060606060606,
296
+ "grad_norm": 45.10826873779297,
297
+ "learning_rate": 1.7111111111111112e-05,
298
+ "loss": 1.8423,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.737812911725955,
303
+ "grad_norm": 37.751338958740234,
304
+ "learning_rate": 1.6964912280701755e-05,
305
+ "loss": 1.8323,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 4.869565217391305,
310
+ "grad_norm": 46.556373596191406,
311
+ "learning_rate": 1.68187134502924e-05,
312
+ "loss": 1.8851,
313
+ "step": 925
314
+ },
315
+ {
316
+ "epoch": 5.0,
317
+ "grad_norm": 44.333099365234375,
318
+ "learning_rate": 1.6672514619883043e-05,
319
+ "loss": 2.0318,
320
+ "step": 950
321
+ },
322
+ {
323
+ "epoch": 5.0,
324
+ "eval_accuracy": 0.527736131934033,
325
+ "eval_loss": 1.276639699935913,
326
+ "eval_macro_f1": 0.5084139910768476,
327
+ "eval_runtime": 1.3473,
328
+ "eval_samples_per_second": 495.077,
329
+ "eval_steps_per_second": 31.174,
330
+ "eval_weighted_f1": 0.5101904415127336,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.13175230566535,
335
+ "grad_norm": 69.827880859375,
336
+ "learning_rate": 1.6526315789473686e-05,
337
+ "loss": 1.4367,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.263504611330698,
342
+ "grad_norm": 50.486148834228516,
343
+ "learning_rate": 1.638011695906433e-05,
344
+ "loss": 1.1245,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.395256916996048,
349
+ "grad_norm": 49.142784118652344,
350
+ "learning_rate": 1.6233918128654974e-05,
351
+ "loss": 1.0049,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.5270092226613965,
356
+ "grad_norm": 49.65623474121094,
357
+ "learning_rate": 1.6087719298245614e-05,
358
+ "loss": 1.0796,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.658761528326746,
363
+ "grad_norm": 65.07655334472656,
364
+ "learning_rate": 1.5941520467836257e-05,
365
+ "loss": 0.8955,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 5.790513833992095,
370
+ "grad_norm": 60.225040435791016,
371
+ "learning_rate": 1.57953216374269e-05,
372
+ "loss": 1.2389,
373
+ "step": 1100
374
+ },
375
+ {
376
+ "epoch": 5.922266139657444,
377
+ "grad_norm": 30.38602638244629,
378
+ "learning_rate": 1.5649122807017545e-05,
379
+ "loss": 1.1696,
380
+ "step": 1125
381
+ },
382
+ {
383
+ "epoch": 6.0,
384
+ "eval_accuracy": 0.5727136431784108,
385
+ "eval_loss": 1.4081095457077026,
386
+ "eval_macro_f1": 0.576171588391268,
387
+ "eval_runtime": 1.3061,
388
+ "eval_samples_per_second": 510.665,
389
+ "eval_steps_per_second": 32.156,
390
+ "eval_weighted_f1": 0.5755697894759714,
391
+ "step": 1140
392
+ }
393
+ ],
394
+ "logging_steps": 25,
395
+ "max_steps": 3800,
396
+ "num_input_tokens_seen": 0,
397
+ "num_train_epochs": 20,
398
+ "save_steps": 500,
399
+ "stateful_callbacks": {
400
+ "EarlyStoppingCallback": {
401
+ "args": {
402
+ "early_stopping_patience": 3,
403
+ "early_stopping_threshold": 0.0
404
+ },
405
+ "attributes": {
406
+ "early_stopping_patience_counter": 0
407
+ }
408
+ },
409
+ "TrainerControl": {
410
+ "args": {
411
+ "should_epoch_stop": false,
412
+ "should_evaluate": false,
413
+ "should_log": false,
414
+ "should_save": true,
415
+ "should_training_stop": false
416
+ },
417
+ "attributes": {}
418
+ }
419
+ },
420
+ "total_flos": 0.0,
421
+ "train_batch_size": 8,
422
+ "trial_name": null,
423
+ "trial_params": null
424
+ }
ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1710/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1710/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1710/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "15001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1710/trainer_state.json ADDED
@@ -0,0 +1,618 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1140,
3
+ "best_metric": 0.576171588391268,
4
+ "best_model_checkpoint": "/root/VNEncoder/results/checkpoints/ood-vifactcheck-visobert-source-source-08-s42-20260831T005402Z-f21e2f7e/trainer/checkpoint-1140",
5
+ "epoch": 9.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1710,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.13175230566534915,
14
+ "grad_norm": 19.988004684448242,
15
+ "learning_rate": 1.2631578947368422e-06,
16
+ "loss": 4.6825,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.2635046113306983,
21
+ "grad_norm": 25.267366409301758,
22
+ "learning_rate": 2.578947368421053e-06,
23
+ "loss": 4.5142,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.3952569169960474,
28
+ "grad_norm": 18.971866607666016,
29
+ "learning_rate": 3.894736842105263e-06,
30
+ "loss": 4.4316,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.5270092226613966,
35
+ "grad_norm": 23.528650283813477,
36
+ "learning_rate": 5.210526315789474e-06,
37
+ "loss": 4.3666,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.6587615283267457,
42
+ "grad_norm": 17.08769416809082,
43
+ "learning_rate": 6.526315789473685e-06,
44
+ "loss": 4.3749,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.7905138339920948,
49
+ "grad_norm": 19.257701873779297,
50
+ "learning_rate": 7.842105263157895e-06,
51
+ "loss": 4.3829,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 0.922266139657444,
56
+ "grad_norm": 17.924440383911133,
57
+ "learning_rate": 9.157894736842105e-06,
58
+ "loss": 4.3979,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_accuracy": 0.46326836581709147,
64
+ "eval_loss": 1.0485528707504272,
65
+ "eval_macro_f1": 0.43630370745575897,
66
+ "eval_runtime": 1.4707,
67
+ "eval_samples_per_second": 453.533,
68
+ "eval_steps_per_second": 28.558,
69
+ "eval_weighted_f1": 0.4382310074294181,
70
+ "step": 190
71
+ },
72
+ {
73
+ "epoch": 1.0527009222661396,
74
+ "grad_norm": 20.339780807495117,
75
+ "learning_rate": 1.0473684210526316e-05,
76
+ "loss": 4.2093,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.1844532279314888,
81
+ "grad_norm": 20.283437728881836,
82
+ "learning_rate": 1.1789473684210527e-05,
83
+ "loss": 4.1858,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.316205533596838,
88
+ "grad_norm": 30.267011642456055,
89
+ "learning_rate": 1.310526315789474e-05,
90
+ "loss": 4.1999,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.447957839262187,
95
+ "grad_norm": 23.36249542236328,
96
+ "learning_rate": 1.4421052631578948e-05,
97
+ "loss": 4.2122,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.5797101449275361,
102
+ "grad_norm": 26.577882766723633,
103
+ "learning_rate": 1.573684210526316e-05,
104
+ "loss": 4.3122,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.7114624505928853,
109
+ "grad_norm": 34.899383544921875,
110
+ "learning_rate": 1.705263157894737e-05,
111
+ "loss": 4.3825,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 1.8432147562582344,
116
+ "grad_norm": 28.601293563842773,
117
+ "learning_rate": 1.836842105263158e-05,
118
+ "loss": 4.1944,
119
+ "step": 350
120
+ },
121
+ {
122
+ "epoch": 1.9749670619235835,
123
+ "grad_norm": 22.760112762451172,
124
+ "learning_rate": 1.968421052631579e-05,
125
+ "loss": 4.2169,
126
+ "step": 375
127
+ },
128
+ {
129
+ "epoch": 2.0,
130
+ "eval_accuracy": 0.4617691154422789,
131
+ "eval_loss": 1.0304769277572632,
132
+ "eval_macro_f1": 0.4374997164954299,
133
+ "eval_runtime": 1.4433,
134
+ "eval_samples_per_second": 462.121,
135
+ "eval_steps_per_second": 29.099,
136
+ "eval_weighted_f1": 0.4377734994515398,
137
+ "step": 380
138
+ },
139
+ {
140
+ "epoch": 2.1054018445322793,
141
+ "grad_norm": 31.96175193786621,
142
+ "learning_rate": 1.988888888888889e-05,
143
+ "loss": 3.8482,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.2371541501976284,
148
+ "grad_norm": 41.56166076660156,
149
+ "learning_rate": 1.9742690058479533e-05,
150
+ "loss": 3.6964,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.3689064558629775,
155
+ "grad_norm": 21.94202995300293,
156
+ "learning_rate": 1.9596491228070177e-05,
157
+ "loss": 3.8187,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.5006587615283267,
162
+ "grad_norm": 38.86528015136719,
163
+ "learning_rate": 1.945029239766082e-05,
164
+ "loss": 3.69,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 2.632411067193676,
169
+ "grad_norm": 30.323911666870117,
170
+ "learning_rate": 1.9304093567251464e-05,
171
+ "loss": 3.7303,
172
+ "step": 500
173
+ },
174
+ {
175
+ "epoch": 2.764163372859025,
176
+ "grad_norm": 47.82787322998047,
177
+ "learning_rate": 1.9157894736842108e-05,
178
+ "loss": 3.5243,
179
+ "step": 525
180
+ },
181
+ {
182
+ "epoch": 2.895915678524374,
183
+ "grad_norm": 26.83512306213379,
184
+ "learning_rate": 1.901169590643275e-05,
185
+ "loss": 3.728,
186
+ "step": 550
187
+ },
188
+ {
189
+ "epoch": 3.0,
190
+ "eval_accuracy": 0.5172413793103449,
191
+ "eval_loss": 1.0077382326126099,
192
+ "eval_macro_f1": 0.508957142495257,
193
+ "eval_runtime": 1.3587,
194
+ "eval_samples_per_second": 490.915,
195
+ "eval_steps_per_second": 30.912,
196
+ "eval_weighted_f1": 0.5101501689469062,
197
+ "step": 570
198
+ },
199
+ {
200
+ "epoch": 3.02635046113307,
201
+ "grad_norm": 44.640281677246094,
202
+ "learning_rate": 1.8865497076023395e-05,
203
+ "loss": 3.407,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.158102766798419,
208
+ "grad_norm": 32.64724349975586,
209
+ "learning_rate": 1.871929824561404e-05,
210
+ "loss": 2.8667,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.289855072463768,
215
+ "grad_norm": 60.53120422363281,
216
+ "learning_rate": 1.857309941520468e-05,
217
+ "loss": 2.8141,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.421607378129117,
222
+ "grad_norm": 43.83180236816406,
223
+ "learning_rate": 1.8426900584795323e-05,
224
+ "loss": 2.7859,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 3.5533596837944663,
229
+ "grad_norm": 33.522403717041016,
230
+ "learning_rate": 1.8280701754385966e-05,
231
+ "loss": 2.9015,
232
+ "step": 675
233
+ },
234
+ {
235
+ "epoch": 3.6851119894598154,
236
+ "grad_norm": 26.346681594848633,
237
+ "learning_rate": 1.813450292397661e-05,
238
+ "loss": 2.8719,
239
+ "step": 700
240
+ },
241
+ {
242
+ "epoch": 3.8168642951251646,
243
+ "grad_norm": 40.303890228271484,
244
+ "learning_rate": 1.7988304093567253e-05,
245
+ "loss": 2.8547,
246
+ "step": 725
247
+ },
248
+ {
249
+ "epoch": 3.9486166007905137,
250
+ "grad_norm": 89.58963775634766,
251
+ "learning_rate": 1.7842105263157897e-05,
252
+ "loss": 2.6832,
253
+ "step": 750
254
+ },
255
+ {
256
+ "epoch": 4.0,
257
+ "eval_accuracy": 0.5112443778110944,
258
+ "eval_loss": 1.2015987634658813,
259
+ "eval_macro_f1": 0.4963979167713946,
260
+ "eval_runtime": 1.3221,
261
+ "eval_samples_per_second": 504.51,
262
+ "eval_steps_per_second": 31.768,
263
+ "eval_weighted_f1": 0.497864935907554,
264
+ "step": 760
265
+ },
266
+ {
267
+ "epoch": 4.07905138339921,
268
+ "grad_norm": 44.19499969482422,
269
+ "learning_rate": 1.769590643274854e-05,
270
+ "loss": 2.5193,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.210803689064559,
275
+ "grad_norm": 29.569128036499023,
276
+ "learning_rate": 1.7549707602339184e-05,
277
+ "loss": 1.877,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 4.342555994729908,
282
+ "grad_norm": 67.0626449584961,
283
+ "learning_rate": 1.7403508771929828e-05,
284
+ "loss": 1.7765,
285
+ "step": 825
286
+ },
287
+ {
288
+ "epoch": 4.474308300395257,
289
+ "grad_norm": 38.277992248535156,
290
+ "learning_rate": 1.725730994152047e-05,
291
+ "loss": 1.7627,
292
+ "step": 850
293
+ },
294
+ {
295
+ "epoch": 4.606060606060606,
296
+ "grad_norm": 45.10826873779297,
297
+ "learning_rate": 1.7111111111111112e-05,
298
+ "loss": 1.8423,
299
+ "step": 875
300
+ },
301
+ {
302
+ "epoch": 4.737812911725955,
303
+ "grad_norm": 37.751338958740234,
304
+ "learning_rate": 1.6964912280701755e-05,
305
+ "loss": 1.8323,
306
+ "step": 900
307
+ },
308
+ {
309
+ "epoch": 4.869565217391305,
310
+ "grad_norm": 46.556373596191406,
311
+ "learning_rate": 1.68187134502924e-05,
312
+ "loss": 1.8851,
313
+ "step": 925
314
+ },
315
+ {
316
+ "epoch": 5.0,
317
+ "grad_norm": 44.333099365234375,
318
+ "learning_rate": 1.6672514619883043e-05,
319
+ "loss": 2.0318,
320
+ "step": 950
321
+ },
322
+ {
323
+ "epoch": 5.0,
324
+ "eval_accuracy": 0.527736131934033,
325
+ "eval_loss": 1.276639699935913,
326
+ "eval_macro_f1": 0.5084139910768476,
327
+ "eval_runtime": 1.3473,
328
+ "eval_samples_per_second": 495.077,
329
+ "eval_steps_per_second": 31.174,
330
+ "eval_weighted_f1": 0.5101904415127336,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.13175230566535,
335
+ "grad_norm": 69.827880859375,
336
+ "learning_rate": 1.6526315789473686e-05,
337
+ "loss": 1.4367,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 5.263504611330698,
342
+ "grad_norm": 50.486148834228516,
343
+ "learning_rate": 1.638011695906433e-05,
344
+ "loss": 1.1245,
345
+ "step": 1000
346
+ },
347
+ {
348
+ "epoch": 5.395256916996048,
349
+ "grad_norm": 49.142784118652344,
350
+ "learning_rate": 1.6233918128654974e-05,
351
+ "loss": 1.0049,
352
+ "step": 1025
353
+ },
354
+ {
355
+ "epoch": 5.5270092226613965,
356
+ "grad_norm": 49.65623474121094,
357
+ "learning_rate": 1.6087719298245614e-05,
358
+ "loss": 1.0796,
359
+ "step": 1050
360
+ },
361
+ {
362
+ "epoch": 5.658761528326746,
363
+ "grad_norm": 65.07655334472656,
364
+ "learning_rate": 1.5941520467836257e-05,
365
+ "loss": 0.8955,
366
+ "step": 1075
367
+ },
368
+ {
369
+ "epoch": 5.790513833992095,
370
+ "grad_norm": 60.225040435791016,
371
+ "learning_rate": 1.57953216374269e-05,
372
+ "loss": 1.2389,
373
+ "step": 1100
374
+ },
375
+ {
376
+ "epoch": 5.922266139657444,
377
+ "grad_norm": 30.38602638244629,
378
+ "learning_rate": 1.5649122807017545e-05,
379
+ "loss": 1.1696,
380
+ "step": 1125
381
+ },
382
+ {
383
+ "epoch": 6.0,
384
+ "eval_accuracy": 0.5727136431784108,
385
+ "eval_loss": 1.4081095457077026,
386
+ "eval_macro_f1": 0.576171588391268,
387
+ "eval_runtime": 1.3061,
388
+ "eval_samples_per_second": 510.665,
389
+ "eval_steps_per_second": 32.156,
390
+ "eval_weighted_f1": 0.5755697894759714,
391
+ "step": 1140
392
+ },
393
+ {
394
+ "epoch": 6.05270092226614,
395
+ "grad_norm": 27.564939498901367,
396
+ "learning_rate": 1.5502923976608188e-05,
397
+ "loss": 0.7303,
398
+ "step": 1150
399
+ },
400
+ {
401
+ "epoch": 6.184453227931489,
402
+ "grad_norm": 50.20222473144531,
403
+ "learning_rate": 1.5356725146198832e-05,
404
+ "loss": 0.7224,
405
+ "step": 1175
406
+ },
407
+ {
408
+ "epoch": 6.316205533596838,
409
+ "grad_norm": 38.52382278442383,
410
+ "learning_rate": 1.5210526315789476e-05,
411
+ "loss": 0.4701,
412
+ "step": 1200
413
+ },
414
+ {
415
+ "epoch": 6.4479578392621875,
416
+ "grad_norm": 77.14038848876953,
417
+ "learning_rate": 1.5064327485380119e-05,
418
+ "loss": 0.7475,
419
+ "step": 1225
420
+ },
421
+ {
422
+ "epoch": 6.579710144927536,
423
+ "grad_norm": 114.14265441894531,
424
+ "learning_rate": 1.4918128654970761e-05,
425
+ "loss": 0.6534,
426
+ "step": 1250
427
+ },
428
+ {
429
+ "epoch": 6.711462450592886,
430
+ "grad_norm": 78.8375473022461,
431
+ "learning_rate": 1.4771929824561405e-05,
432
+ "loss": 0.6285,
433
+ "step": 1275
434
+ },
435
+ {
436
+ "epoch": 6.843214756258234,
437
+ "grad_norm": 39.12944412231445,
438
+ "learning_rate": 1.4625730994152048e-05,
439
+ "loss": 0.6822,
440
+ "step": 1300
441
+ },
442
+ {
443
+ "epoch": 6.974967061923584,
444
+ "grad_norm": 45.70170974731445,
445
+ "learning_rate": 1.447953216374269e-05,
446
+ "loss": 0.5287,
447
+ "step": 1325
448
+ },
449
+ {
450
+ "epoch": 7.0,
451
+ "eval_accuracy": 0.5517241379310345,
452
+ "eval_loss": 1.8658064603805542,
453
+ "eval_macro_f1": 0.5510129251856938,
454
+ "eval_runtime": 1.3036,
455
+ "eval_samples_per_second": 511.672,
456
+ "eval_steps_per_second": 32.219,
457
+ "eval_weighted_f1": 0.5519634897968083,
458
+ "step": 1330
459
+ },
460
+ {
461
+ "epoch": 7.105401844532279,
462
+ "grad_norm": 24.08159828186035,
463
+ "learning_rate": 1.4333333333333334e-05,
464
+ "loss": 0.4743,
465
+ "step": 1350
466
+ },
467
+ {
468
+ "epoch": 7.237154150197629,
469
+ "grad_norm": 39.90761947631836,
470
+ "learning_rate": 1.4187134502923978e-05,
471
+ "loss": 0.3069,
472
+ "step": 1375
473
+ },
474
+ {
475
+ "epoch": 7.3689064558629775,
476
+ "grad_norm": 89.20237731933594,
477
+ "learning_rate": 1.4040935672514621e-05,
478
+ "loss": 0.2668,
479
+ "step": 1400
480
+ },
481
+ {
482
+ "epoch": 7.500658761528327,
483
+ "grad_norm": 91.07672119140625,
484
+ "learning_rate": 1.3894736842105265e-05,
485
+ "loss": 0.3427,
486
+ "step": 1425
487
+ },
488
+ {
489
+ "epoch": 7.632411067193676,
490
+ "grad_norm": 19.943893432617188,
491
+ "learning_rate": 1.3748538011695907e-05,
492
+ "loss": 0.3075,
493
+ "step": 1450
494
+ },
495
+ {
496
+ "epoch": 7.764163372859025,
497
+ "grad_norm": 49.538089752197266,
498
+ "learning_rate": 1.360233918128655e-05,
499
+ "loss": 0.4815,
500
+ "step": 1475
501
+ },
502
+ {
503
+ "epoch": 7.895915678524374,
504
+ "grad_norm": 70.27389526367188,
505
+ "learning_rate": 1.3456140350877194e-05,
506
+ "loss": 0.3766,
507
+ "step": 1500
508
+ },
509
+ {
510
+ "epoch": 8.0,
511
+ "eval_accuracy": 0.5517241379310345,
512
+ "eval_loss": 2.2899744510650635,
513
+ "eval_macro_f1": 0.5518911105372716,
514
+ "eval_runtime": 1.3518,
515
+ "eval_samples_per_second": 493.421,
516
+ "eval_steps_per_second": 31.07,
517
+ "eval_weighted_f1": 0.5521476486201882,
518
+ "step": 1520
519
+ },
520
+ {
521
+ "epoch": 8.02635046113307,
522
+ "grad_norm": 28.080217361450195,
523
+ "learning_rate": 1.3309941520467838e-05,
524
+ "loss": 0.5053,
525
+ "step": 1525
526
+ },
527
+ {
528
+ "epoch": 8.15810276679842,
529
+ "grad_norm": 6.912923812866211,
530
+ "learning_rate": 1.3163742690058481e-05,
531
+ "loss": 0.2343,
532
+ "step": 1550
533
+ },
534
+ {
535
+ "epoch": 8.289855072463768,
536
+ "grad_norm": 51.6353645324707,
537
+ "learning_rate": 1.3017543859649125e-05,
538
+ "loss": 0.2245,
539
+ "step": 1575
540
+ },
541
+ {
542
+ "epoch": 8.421607378129117,
543
+ "grad_norm": 74.35002136230469,
544
+ "learning_rate": 1.2871345029239768e-05,
545
+ "loss": 0.2491,
546
+ "step": 1600
547
+ },
548
+ {
549
+ "epoch": 8.553359683794467,
550
+ "grad_norm": 72.16822052001953,
551
+ "learning_rate": 1.2725146198830412e-05,
552
+ "loss": 0.4967,
553
+ "step": 1625
554
+ },
555
+ {
556
+ "epoch": 8.685111989459816,
557
+ "grad_norm": 3.117950201034546,
558
+ "learning_rate": 1.2578947368421052e-05,
559
+ "loss": 0.3127,
560
+ "step": 1650
561
+ },
562
+ {
563
+ "epoch": 8.816864295125164,
564
+ "grad_norm": 83.23318481445312,
565
+ "learning_rate": 1.2432748538011696e-05,
566
+ "loss": 0.208,
567
+ "step": 1675
568
+ },
569
+ {
570
+ "epoch": 8.948616600790514,
571
+ "grad_norm": 70.69136047363281,
572
+ "learning_rate": 1.228654970760234e-05,
573
+ "loss": 0.5268,
574
+ "step": 1700
575
+ },
576
+ {
577
+ "epoch": 9.0,
578
+ "eval_accuracy": 0.5562218890554723,
579
+ "eval_loss": 2.5926525592803955,
580
+ "eval_macro_f1": 0.5582084525539238,
581
+ "eval_runtime": 1.3838,
582
+ "eval_samples_per_second": 481.993,
583
+ "eval_steps_per_second": 30.35,
584
+ "eval_weighted_f1": 0.5576762237561994,
585
+ "step": 1710
586
+ }
587
+ ],
588
+ "logging_steps": 25,
589
+ "max_steps": 3800,
590
+ "num_input_tokens_seen": 0,
591
+ "num_train_epochs": 20,
592
+ "save_steps": 500,
593
+ "stateful_callbacks": {
594
+ "EarlyStoppingCallback": {
595
+ "args": {
596
+ "early_stopping_patience": 3,
597
+ "early_stopping_threshold": 0.0
598
+ },
599
+ "attributes": {
600
+ "early_stopping_patience_counter": 3
601
+ }
602
+ },
603
+ "TrainerControl": {
604
+ "args": {
605
+ "should_epoch_stop": false,
606
+ "should_evaluate": false,
607
+ "should_log": false,
608
+ "should_save": true,
609
+ "should_training_stop": true
610
+ },
611
+ "attributes": {}
612
+ }
613
+ },
614
+ "total_flos": 0.0,
615
+ "train_batch_size": 8,
616
+ "trial_name": null,
617
+ "trial_params": null
618
+ }