PhaaNe commited on
Commit
a0f8e44
·
verified ·
1 Parent(s): 92e2128

Add explore checkpoint explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23 (job 9963f3d9e87ef2521b59, factcheck-ablations)

Browse files
Files changed (21) hide show
  1. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/model.safetensors +3 -0
  2. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/model.safetensors +3 -0
  3. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/optimizer.pt +3 -0
  4. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/rng_state.pth +3 -0
  5. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/scheduler.pt +3 -0
  6. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/sentencepiece.bpe.model +3 -0
  7. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/special_tokens_map.json +15 -0
  8. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/tokenizer.json +0 -0
  9. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/tokenizer_config.json +56 -0
  10. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/trainer_state.json +435 -0
  11. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/training_args.bin +3 -0
  12. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/model.safetensors +3 -0
  13. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/optimizer.pt +3 -0
  14. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/rng_state.pth +3 -0
  15. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/scheduler.pt +3 -0
  16. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/sentencepiece.bpe.model +3 -0
  17. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/special_tokens_map.json +15 -0
  18. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/tokenizer.json +0 -0
  19. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/tokenizer_config.json +56 -0
  20. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/trainer_state.json +608 -0
  21. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/training_args.bin +3 -0
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bcef32cc0ee8f1d32a8c52a61e7e951d5adfb7e4bafff7ea47a9e515795e7f23
3
+ size 390297060
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d69c4dd1b1b46ddbf64587e45abded851bbd98f69560ba33a3edb79ab49b90b
3
+ size 390297092
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c623feeb6669e8b09b35386b5264b53c1d288f9d1071dc7b87dfc82cdfb7de5
3
+ size 775987770
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:136257dc61addaf8c1d3702edf36f54eed52bd776033aa265c91e681f5bd4b7c
3
+ size 14244
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de0ddcd1c7002c3feb012aac545f82daba512f7da257c6d1e7d47264713fa9e9
3
+ size 1064
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02aaf05cda4db99e86b7c76eba6258867ce4d043da0fed19c87a7d46c8b53a65
3
+ size 470732
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/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
+ }
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/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
+ }
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/trainer_state.json ADDED
@@ -0,0 +1,435 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1141,
3
+ "best_metric": 0.5304348773644195,
4
+ "best_model_checkpoint": "/root/VNEncoder/explore/workspaces/9963f3d9e87ef2521b59/results/checkpoints/explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141",
5
+ "epoch": 7.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1141,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.15408320493066255,
14
+ "grad_norm": 24.169546127319336,
15
+ "learning_rate": 1.47239263803681e-06,
16
+ "loss": 4.686,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.3081664098613251,
21
+ "grad_norm": 27.795276641845703,
22
+ "learning_rate": 3.0061349693251535e-06,
23
+ "loss": 4.5156,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.4622496147919877,
28
+ "grad_norm": 24.485998153686523,
29
+ "learning_rate": 4.539877300613497e-06,
30
+ "loss": 4.4996,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.6163328197226502,
35
+ "grad_norm": 26.885868072509766,
36
+ "learning_rate": 6.073619631901841e-06,
37
+ "loss": 4.3957,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.7704160246533128,
42
+ "grad_norm": 35.59861373901367,
43
+ "learning_rate": 7.6073619631901856e-06,
44
+ "loss": 4.4215,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.9244992295839753,
49
+ "grad_norm": 16.04206657409668,
50
+ "learning_rate": 9.14110429447853e-06,
51
+ "loss": 4.4406,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 1.0,
56
+ "eval_accuracy": 0.4375,
57
+ "eval_loss": 1.061625599861145,
58
+ "eval_macro_f1": 0.4336375068640515,
59
+ "eval_runtime": 0.5186,
60
+ "eval_samples_per_second": 1326.713,
61
+ "eval_steps_per_second": 82.92,
62
+ "eval_weighted_f1": 0.43439253618328705,
63
+ "step": 163
64
+ },
65
+ {
66
+ "epoch": 1.073959938366718,
67
+ "grad_norm": 23.72640609741211,
68
+ "learning_rate": 1.0674846625766873e-05,
69
+ "loss": 4.1978,
70
+ "step": 175
71
+ },
72
+ {
73
+ "epoch": 1.2280431432973806,
74
+ "grad_norm": 19.785627365112305,
75
+ "learning_rate": 1.2208588957055216e-05,
76
+ "loss": 4.1947,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.3821263482280433,
81
+ "grad_norm": 28.096067428588867,
82
+ "learning_rate": 1.374233128834356e-05,
83
+ "loss": 4.2188,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.5362095531587057,
88
+ "grad_norm": 18.727331161499023,
89
+ "learning_rate": 1.5276073619631903e-05,
90
+ "loss": 4.2631,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.6902927580893683,
95
+ "grad_norm": 33.542518615722656,
96
+ "learning_rate": 1.6809815950920248e-05,
97
+ "loss": 4.2123,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.8443759630200307,
102
+ "grad_norm": 18.366376876831055,
103
+ "learning_rate": 1.834355828220859e-05,
104
+ "loss": 4.259,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.9984591679506933,
109
+ "grad_norm": 31.03815269470215,
110
+ "learning_rate": 1.9877300613496935e-05,
111
+ "loss": 4.2454,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 2.0,
116
+ "eval_accuracy": 0.4680232558139535,
117
+ "eval_loss": 1.0228641033172607,
118
+ "eval_macro_f1": 0.43939666021652374,
119
+ "eval_runtime": 0.5091,
120
+ "eval_samples_per_second": 1351.275,
121
+ "eval_steps_per_second": 84.455,
122
+ "eval_weighted_f1": 0.44148008394490457,
123
+ "step": 326
124
+ },
125
+ {
126
+ "epoch": 2.147919876733436,
127
+ "grad_norm": 20.479841232299805,
128
+ "learning_rate": 1.9843217450579417e-05,
129
+ "loss": 3.8271,
130
+ "step": 350
131
+ },
132
+ {
133
+ "epoch": 2.3020030816640986,
134
+ "grad_norm": 29.96990966796875,
135
+ "learning_rate": 1.967280163599182e-05,
136
+ "loss": 3.8301,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.4560862865947612,
141
+ "grad_norm": 29.412343978881836,
142
+ "learning_rate": 1.9502385821404228e-05,
143
+ "loss": 3.7406,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.610169491525424,
148
+ "grad_norm": 31.202903747558594,
149
+ "learning_rate": 1.9331970006816635e-05,
150
+ "loss": 3.5667,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.7642526964560865,
155
+ "grad_norm": 30.862821578979492,
156
+ "learning_rate": 1.916155419222904e-05,
157
+ "loss": 3.7121,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.9183359013867487,
162
+ "grad_norm": 33.535804748535156,
163
+ "learning_rate": 1.8991138377641446e-05,
164
+ "loss": 3.676,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 3.0,
169
+ "eval_accuracy": 0.4636627906976744,
170
+ "eval_loss": 1.0977932214736938,
171
+ "eval_macro_f1": 0.43252897200265616,
172
+ "eval_runtime": 0.5362,
173
+ "eval_samples_per_second": 1283.079,
174
+ "eval_steps_per_second": 80.192,
175
+ "eval_weighted_f1": 0.4355374497663359,
176
+ "step": 489
177
+ },
178
+ {
179
+ "epoch": 3.0677966101694913,
180
+ "grad_norm": 28.798425674438477,
181
+ "learning_rate": 1.8820722563053853e-05,
182
+ "loss": 3.3501,
183
+ "step": 500
184
+ },
185
+ {
186
+ "epoch": 3.221879815100154,
187
+ "grad_norm": 40.66737365722656,
188
+ "learning_rate": 1.8650306748466257e-05,
189
+ "loss": 2.9628,
190
+ "step": 525
191
+ },
192
+ {
193
+ "epoch": 3.3759630200308166,
194
+ "grad_norm": 30.825912475585938,
195
+ "learning_rate": 1.8479890933878664e-05,
196
+ "loss": 2.8444,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.5300462249614792,
201
+ "grad_norm": 31.344446182250977,
202
+ "learning_rate": 1.8309475119291072e-05,
203
+ "loss": 2.8893,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.684129429892142,
208
+ "grad_norm": 37.38386917114258,
209
+ "learning_rate": 1.8139059304703476e-05,
210
+ "loss": 2.9412,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.8382126348228045,
215
+ "grad_norm": 30.051149368286133,
216
+ "learning_rate": 1.7968643490115886e-05,
217
+ "loss": 2.9008,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.992295839753467,
222
+ "grad_norm": 46.07156753540039,
223
+ "learning_rate": 1.779822767552829e-05,
224
+ "loss": 2.9271,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 4.0,
229
+ "eval_accuracy": 0.5319767441860465,
230
+ "eval_loss": 1.0258407592773438,
231
+ "eval_macro_f1": 0.5293401869445336,
232
+ "eval_runtime": 0.5371,
233
+ "eval_samples_per_second": 1281.029,
234
+ "eval_steps_per_second": 80.064,
235
+ "eval_weighted_f1": 0.5288047204131197,
236
+ "step": 652
237
+ },
238
+ {
239
+ "epoch": 4.141756548536209,
240
+ "grad_norm": 35.36317443847656,
241
+ "learning_rate": 1.7627811860940697e-05,
242
+ "loss": 2.1226,
243
+ "step": 675
244
+ },
245
+ {
246
+ "epoch": 4.295839753466872,
247
+ "grad_norm": 42.33112716674805,
248
+ "learning_rate": 1.7457396046353105e-05,
249
+ "loss": 1.8968,
250
+ "step": 700
251
+ },
252
+ {
253
+ "epoch": 4.449922958397535,
254
+ "grad_norm": 67.32442474365234,
255
+ "learning_rate": 1.728698023176551e-05,
256
+ "loss": 2.0063,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.604006163328197,
261
+ "grad_norm": 48.348548889160156,
262
+ "learning_rate": 1.7116564417177916e-05,
263
+ "loss": 2.0946,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.75808936825886,
268
+ "grad_norm": 37.93932342529297,
269
+ "learning_rate": 1.6946148602590323e-05,
270
+ "loss": 1.8833,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.9121725731895225,
275
+ "grad_norm": 41.16898727416992,
276
+ "learning_rate": 1.6775732788002727e-05,
277
+ "loss": 2.0521,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 5.0,
282
+ "eval_accuracy": 0.5363372093023255,
283
+ "eval_loss": 1.2476704120635986,
284
+ "eval_macro_f1": 0.5170588114084537,
285
+ "eval_runtime": 0.5076,
286
+ "eval_samples_per_second": 1355.344,
287
+ "eval_steps_per_second": 84.709,
288
+ "eval_weighted_f1": 0.5174632259693357,
289
+ "step": 815
290
+ },
291
+ {
292
+ "epoch": 5.061633281972265,
293
+ "grad_norm": 25.506328582763672,
294
+ "learning_rate": 1.6605316973415134e-05,
295
+ "loss": 1.5853,
296
+ "step": 825
297
+ },
298
+ {
299
+ "epoch": 5.215716486902927,
300
+ "grad_norm": 72.44429016113281,
301
+ "learning_rate": 1.643490115882754e-05,
302
+ "loss": 1.137,
303
+ "step": 850
304
+ },
305
+ {
306
+ "epoch": 5.36979969183359,
307
+ "grad_norm": 56.91807174682617,
308
+ "learning_rate": 1.6264485344239945e-05,
309
+ "loss": 1.0425,
310
+ "step": 875
311
+ },
312
+ {
313
+ "epoch": 5.523882896764253,
314
+ "grad_norm": 49.060516357421875,
315
+ "learning_rate": 1.6094069529652356e-05,
316
+ "loss": 1.1692,
317
+ "step": 900
318
+ },
319
+ {
320
+ "epoch": 5.677966101694915,
321
+ "grad_norm": 66.7464599609375,
322
+ "learning_rate": 1.592365371506476e-05,
323
+ "loss": 1.0896,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.832049306625578,
328
+ "grad_norm": 43.30105209350586,
329
+ "learning_rate": 1.5753237900477167e-05,
330
+ "loss": 1.095,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.9861325115562405,
335
+ "grad_norm": 65.80976104736328,
336
+ "learning_rate": 1.5582822085889574e-05,
337
+ "loss": 1.1717,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 6.0,
342
+ "eval_accuracy": 0.4941860465116279,
343
+ "eval_loss": 1.8342998027801514,
344
+ "eval_macro_f1": 0.47510319462539713,
345
+ "eval_runtime": 0.5021,
346
+ "eval_samples_per_second": 1370.296,
347
+ "eval_steps_per_second": 85.643,
348
+ "eval_weighted_f1": 0.4712748564934589,
349
+ "step": 978
350
+ },
351
+ {
352
+ "epoch": 6.135593220338983,
353
+ "grad_norm": 39.15993118286133,
354
+ "learning_rate": 1.5412406271301978e-05,
355
+ "loss": 0.8963,
356
+ "step": 1000
357
+ },
358
+ {
359
+ "epoch": 6.289676425269645,
360
+ "grad_norm": 72.50019073486328,
361
+ "learning_rate": 1.5241990456714383e-05,
362
+ "loss": 0.6562,
363
+ "step": 1025
364
+ },
365
+ {
366
+ "epoch": 6.443759630200308,
367
+ "grad_norm": 27.87003517150879,
368
+ "learning_rate": 1.5071574642126792e-05,
369
+ "loss": 0.5902,
370
+ "step": 1050
371
+ },
372
+ {
373
+ "epoch": 6.597842835130971,
374
+ "grad_norm": 73.38944244384766,
375
+ "learning_rate": 1.4901158827539198e-05,
376
+ "loss": 0.5439,
377
+ "step": 1075
378
+ },
379
+ {
380
+ "epoch": 6.751926040061633,
381
+ "grad_norm": 64.54303741455078,
382
+ "learning_rate": 1.4730743012951603e-05,
383
+ "loss": 0.607,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.906009244992296,
388
+ "grad_norm": 81.66258239746094,
389
+ "learning_rate": 1.456032719836401e-05,
390
+ "loss": 0.6815,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 7.0,
395
+ "eval_accuracy": 0.5406976744186046,
396
+ "eval_loss": 1.8944121599197388,
397
+ "eval_macro_f1": 0.5304348773644195,
398
+ "eval_runtime": 0.5112,
399
+ "eval_samples_per_second": 1345.786,
400
+ "eval_steps_per_second": 84.112,
401
+ "eval_weighted_f1": 0.5293449915431478,
402
+ "step": 1141
403
+ }
404
+ ],
405
+ "logging_steps": 25,
406
+ "max_steps": 3260,
407
+ "num_input_tokens_seen": 0,
408
+ "num_train_epochs": 20,
409
+ "save_steps": 500,
410
+ "stateful_callbacks": {
411
+ "EarlyStoppingCallback": {
412
+ "args": {
413
+ "early_stopping_patience": 3,
414
+ "early_stopping_threshold": 0.0
415
+ },
416
+ "attributes": {
417
+ "early_stopping_patience_counter": 0
418
+ }
419
+ },
420
+ "TrainerControl": {
421
+ "args": {
422
+ "should_epoch_stop": false,
423
+ "should_evaluate": false,
424
+ "should_log": false,
425
+ "should_save": true,
426
+ "should_training_stop": false
427
+ },
428
+ "attributes": {}
429
+ }
430
+ },
431
+ "total_flos": 0.0,
432
+ "train_batch_size": 8,
433
+ "trial_name": null,
434
+ "trial_params": null
435
+ }
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd74124b85dfaf79ef5694b739606087d18befc4bee11a030503b4441be7f855
3
+ size 5752
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30f07116e84e985879399eac8c382551799380ff515141dd45b332f15bd4d886
3
+ size 390297092
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cc02b48e2121d5b98a5c9c13cbe10bddec7f4c84a81b465a45d8cffe64cf3a0
3
+ size 775987770
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a80a50db0f90fb36f673f1e5f515074fca50f8916ee40bd965c69213ce55970
3
+ size 14244
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9aeb06943d6c965ed80285dde181a2ea5786114ccd33bd03e59ee291d9eb99dc
3
+ size 1064
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02aaf05cda4db99e86b7c76eba6258867ce4d043da0fed19c87a7d46c8b53a65
3
+ size 470732
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/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
+ }
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/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
+ }
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/trainer_state.json ADDED
@@ -0,0 +1,608 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1141,
3
+ "best_metric": 0.5304348773644195,
4
+ "best_model_checkpoint": "/root/VNEncoder/explore/workspaces/9963f3d9e87ef2521b59/results/checkpoints/explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1141",
5
+ "epoch": 10.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1630,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.15408320493066255,
14
+ "grad_norm": 24.169546127319336,
15
+ "learning_rate": 1.47239263803681e-06,
16
+ "loss": 4.686,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.3081664098613251,
21
+ "grad_norm": 27.795276641845703,
22
+ "learning_rate": 3.0061349693251535e-06,
23
+ "loss": 4.5156,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.4622496147919877,
28
+ "grad_norm": 24.485998153686523,
29
+ "learning_rate": 4.539877300613497e-06,
30
+ "loss": 4.4996,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.6163328197226502,
35
+ "grad_norm": 26.885868072509766,
36
+ "learning_rate": 6.073619631901841e-06,
37
+ "loss": 4.3957,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.7704160246533128,
42
+ "grad_norm": 35.59861373901367,
43
+ "learning_rate": 7.6073619631901856e-06,
44
+ "loss": 4.4215,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.9244992295839753,
49
+ "grad_norm": 16.04206657409668,
50
+ "learning_rate": 9.14110429447853e-06,
51
+ "loss": 4.4406,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 1.0,
56
+ "eval_accuracy": 0.4375,
57
+ "eval_loss": 1.061625599861145,
58
+ "eval_macro_f1": 0.4336375068640515,
59
+ "eval_runtime": 0.5186,
60
+ "eval_samples_per_second": 1326.713,
61
+ "eval_steps_per_second": 82.92,
62
+ "eval_weighted_f1": 0.43439253618328705,
63
+ "step": 163
64
+ },
65
+ {
66
+ "epoch": 1.073959938366718,
67
+ "grad_norm": 23.72640609741211,
68
+ "learning_rate": 1.0674846625766873e-05,
69
+ "loss": 4.1978,
70
+ "step": 175
71
+ },
72
+ {
73
+ "epoch": 1.2280431432973806,
74
+ "grad_norm": 19.785627365112305,
75
+ "learning_rate": 1.2208588957055216e-05,
76
+ "loss": 4.1947,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.3821263482280433,
81
+ "grad_norm": 28.096067428588867,
82
+ "learning_rate": 1.374233128834356e-05,
83
+ "loss": 4.2188,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.5362095531587057,
88
+ "grad_norm": 18.727331161499023,
89
+ "learning_rate": 1.5276073619631903e-05,
90
+ "loss": 4.2631,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.6902927580893683,
95
+ "grad_norm": 33.542518615722656,
96
+ "learning_rate": 1.6809815950920248e-05,
97
+ "loss": 4.2123,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.8443759630200307,
102
+ "grad_norm": 18.366376876831055,
103
+ "learning_rate": 1.834355828220859e-05,
104
+ "loss": 4.259,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.9984591679506933,
109
+ "grad_norm": 31.03815269470215,
110
+ "learning_rate": 1.9877300613496935e-05,
111
+ "loss": 4.2454,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 2.0,
116
+ "eval_accuracy": 0.4680232558139535,
117
+ "eval_loss": 1.0228641033172607,
118
+ "eval_macro_f1": 0.43939666021652374,
119
+ "eval_runtime": 0.5091,
120
+ "eval_samples_per_second": 1351.275,
121
+ "eval_steps_per_second": 84.455,
122
+ "eval_weighted_f1": 0.44148008394490457,
123
+ "step": 326
124
+ },
125
+ {
126
+ "epoch": 2.147919876733436,
127
+ "grad_norm": 20.479841232299805,
128
+ "learning_rate": 1.9843217450579417e-05,
129
+ "loss": 3.8271,
130
+ "step": 350
131
+ },
132
+ {
133
+ "epoch": 2.3020030816640986,
134
+ "grad_norm": 29.96990966796875,
135
+ "learning_rate": 1.967280163599182e-05,
136
+ "loss": 3.8301,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.4560862865947612,
141
+ "grad_norm": 29.412343978881836,
142
+ "learning_rate": 1.9502385821404228e-05,
143
+ "loss": 3.7406,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.610169491525424,
148
+ "grad_norm": 31.202903747558594,
149
+ "learning_rate": 1.9331970006816635e-05,
150
+ "loss": 3.5667,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.7642526964560865,
155
+ "grad_norm": 30.862821578979492,
156
+ "learning_rate": 1.916155419222904e-05,
157
+ "loss": 3.7121,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.9183359013867487,
162
+ "grad_norm": 33.535804748535156,
163
+ "learning_rate": 1.8991138377641446e-05,
164
+ "loss": 3.676,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 3.0,
169
+ "eval_accuracy": 0.4636627906976744,
170
+ "eval_loss": 1.0977932214736938,
171
+ "eval_macro_f1": 0.43252897200265616,
172
+ "eval_runtime": 0.5362,
173
+ "eval_samples_per_second": 1283.079,
174
+ "eval_steps_per_second": 80.192,
175
+ "eval_weighted_f1": 0.4355374497663359,
176
+ "step": 489
177
+ },
178
+ {
179
+ "epoch": 3.0677966101694913,
180
+ "grad_norm": 28.798425674438477,
181
+ "learning_rate": 1.8820722563053853e-05,
182
+ "loss": 3.3501,
183
+ "step": 500
184
+ },
185
+ {
186
+ "epoch": 3.221879815100154,
187
+ "grad_norm": 40.66737365722656,
188
+ "learning_rate": 1.8650306748466257e-05,
189
+ "loss": 2.9628,
190
+ "step": 525
191
+ },
192
+ {
193
+ "epoch": 3.3759630200308166,
194
+ "grad_norm": 30.825912475585938,
195
+ "learning_rate": 1.8479890933878664e-05,
196
+ "loss": 2.8444,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.5300462249614792,
201
+ "grad_norm": 31.344446182250977,
202
+ "learning_rate": 1.8309475119291072e-05,
203
+ "loss": 2.8893,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.684129429892142,
208
+ "grad_norm": 37.38386917114258,
209
+ "learning_rate": 1.8139059304703476e-05,
210
+ "loss": 2.9412,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.8382126348228045,
215
+ "grad_norm": 30.051149368286133,
216
+ "learning_rate": 1.7968643490115886e-05,
217
+ "loss": 2.9008,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.992295839753467,
222
+ "grad_norm": 46.07156753540039,
223
+ "learning_rate": 1.779822767552829e-05,
224
+ "loss": 2.9271,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 4.0,
229
+ "eval_accuracy": 0.5319767441860465,
230
+ "eval_loss": 1.0258407592773438,
231
+ "eval_macro_f1": 0.5293401869445336,
232
+ "eval_runtime": 0.5371,
233
+ "eval_samples_per_second": 1281.029,
234
+ "eval_steps_per_second": 80.064,
235
+ "eval_weighted_f1": 0.5288047204131197,
236
+ "step": 652
237
+ },
238
+ {
239
+ "epoch": 4.141756548536209,
240
+ "grad_norm": 35.36317443847656,
241
+ "learning_rate": 1.7627811860940697e-05,
242
+ "loss": 2.1226,
243
+ "step": 675
244
+ },
245
+ {
246
+ "epoch": 4.295839753466872,
247
+ "grad_norm": 42.33112716674805,
248
+ "learning_rate": 1.7457396046353105e-05,
249
+ "loss": 1.8968,
250
+ "step": 700
251
+ },
252
+ {
253
+ "epoch": 4.449922958397535,
254
+ "grad_norm": 67.32442474365234,
255
+ "learning_rate": 1.728698023176551e-05,
256
+ "loss": 2.0063,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.604006163328197,
261
+ "grad_norm": 48.348548889160156,
262
+ "learning_rate": 1.7116564417177916e-05,
263
+ "loss": 2.0946,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.75808936825886,
268
+ "grad_norm": 37.93932342529297,
269
+ "learning_rate": 1.6946148602590323e-05,
270
+ "loss": 1.8833,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.9121725731895225,
275
+ "grad_norm": 41.16898727416992,
276
+ "learning_rate": 1.6775732788002727e-05,
277
+ "loss": 2.0521,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 5.0,
282
+ "eval_accuracy": 0.5363372093023255,
283
+ "eval_loss": 1.2476704120635986,
284
+ "eval_macro_f1": 0.5170588114084537,
285
+ "eval_runtime": 0.5076,
286
+ "eval_samples_per_second": 1355.344,
287
+ "eval_steps_per_second": 84.709,
288
+ "eval_weighted_f1": 0.5174632259693357,
289
+ "step": 815
290
+ },
291
+ {
292
+ "epoch": 5.061633281972265,
293
+ "grad_norm": 25.506328582763672,
294
+ "learning_rate": 1.6605316973415134e-05,
295
+ "loss": 1.5853,
296
+ "step": 825
297
+ },
298
+ {
299
+ "epoch": 5.215716486902927,
300
+ "grad_norm": 72.44429016113281,
301
+ "learning_rate": 1.643490115882754e-05,
302
+ "loss": 1.137,
303
+ "step": 850
304
+ },
305
+ {
306
+ "epoch": 5.36979969183359,
307
+ "grad_norm": 56.91807174682617,
308
+ "learning_rate": 1.6264485344239945e-05,
309
+ "loss": 1.0425,
310
+ "step": 875
311
+ },
312
+ {
313
+ "epoch": 5.523882896764253,
314
+ "grad_norm": 49.060516357421875,
315
+ "learning_rate": 1.6094069529652356e-05,
316
+ "loss": 1.1692,
317
+ "step": 900
318
+ },
319
+ {
320
+ "epoch": 5.677966101694915,
321
+ "grad_norm": 66.7464599609375,
322
+ "learning_rate": 1.592365371506476e-05,
323
+ "loss": 1.0896,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.832049306625578,
328
+ "grad_norm": 43.30105209350586,
329
+ "learning_rate": 1.5753237900477167e-05,
330
+ "loss": 1.095,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.9861325115562405,
335
+ "grad_norm": 65.80976104736328,
336
+ "learning_rate": 1.5582822085889574e-05,
337
+ "loss": 1.1717,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 6.0,
342
+ "eval_accuracy": 0.4941860465116279,
343
+ "eval_loss": 1.8342998027801514,
344
+ "eval_macro_f1": 0.47510319462539713,
345
+ "eval_runtime": 0.5021,
346
+ "eval_samples_per_second": 1370.296,
347
+ "eval_steps_per_second": 85.643,
348
+ "eval_weighted_f1": 0.4712748564934589,
349
+ "step": 978
350
+ },
351
+ {
352
+ "epoch": 6.135593220338983,
353
+ "grad_norm": 39.15993118286133,
354
+ "learning_rate": 1.5412406271301978e-05,
355
+ "loss": 0.8963,
356
+ "step": 1000
357
+ },
358
+ {
359
+ "epoch": 6.289676425269645,
360
+ "grad_norm": 72.50019073486328,
361
+ "learning_rate": 1.5241990456714383e-05,
362
+ "loss": 0.6562,
363
+ "step": 1025
364
+ },
365
+ {
366
+ "epoch": 6.443759630200308,
367
+ "grad_norm": 27.87003517150879,
368
+ "learning_rate": 1.5071574642126792e-05,
369
+ "loss": 0.5902,
370
+ "step": 1050
371
+ },
372
+ {
373
+ "epoch": 6.597842835130971,
374
+ "grad_norm": 73.38944244384766,
375
+ "learning_rate": 1.4901158827539198e-05,
376
+ "loss": 0.5439,
377
+ "step": 1075
378
+ },
379
+ {
380
+ "epoch": 6.751926040061633,
381
+ "grad_norm": 64.54303741455078,
382
+ "learning_rate": 1.4730743012951603e-05,
383
+ "loss": 0.607,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.906009244992296,
388
+ "grad_norm": 81.66258239746094,
389
+ "learning_rate": 1.456032719836401e-05,
390
+ "loss": 0.6815,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 7.0,
395
+ "eval_accuracy": 0.5406976744186046,
396
+ "eval_loss": 1.8944121599197388,
397
+ "eval_macro_f1": 0.5304348773644195,
398
+ "eval_runtime": 0.5112,
399
+ "eval_samples_per_second": 1345.786,
400
+ "eval_steps_per_second": 84.112,
401
+ "eval_weighted_f1": 0.5293449915431478,
402
+ "step": 1141
403
+ },
404
+ {
405
+ "epoch": 7.055469953775039,
406
+ "grad_norm": 38.628562927246094,
407
+ "learning_rate": 1.4389911383776416e-05,
408
+ "loss": 0.5729,
409
+ "step": 1150
410
+ },
411
+ {
412
+ "epoch": 7.209553158705701,
413
+ "grad_norm": 73.1359634399414,
414
+ "learning_rate": 1.4219495569188822e-05,
415
+ "loss": 0.3277,
416
+ "step": 1175
417
+ },
418
+ {
419
+ "epoch": 7.363636363636363,
420
+ "grad_norm": 66.63118743896484,
421
+ "learning_rate": 1.4049079754601229e-05,
422
+ "loss": 0.3122,
423
+ "step": 1200
424
+ },
425
+ {
426
+ "epoch": 7.517719568567026,
427
+ "grad_norm": 115.604248046875,
428
+ "learning_rate": 1.3878663940013635e-05,
429
+ "loss": 0.4011,
430
+ "step": 1225
431
+ },
432
+ {
433
+ "epoch": 7.671802773497689,
434
+ "grad_norm": 60.14704132080078,
435
+ "learning_rate": 1.370824812542604e-05,
436
+ "loss": 0.4226,
437
+ "step": 1250
438
+ },
439
+ {
440
+ "epoch": 7.825885978428351,
441
+ "grad_norm": 20.83880043029785,
442
+ "learning_rate": 1.3537832310838447e-05,
443
+ "loss": 0.6027,
444
+ "step": 1275
445
+ },
446
+ {
447
+ "epoch": 7.979969183359014,
448
+ "grad_norm": 79.25132751464844,
449
+ "learning_rate": 1.3367416496250853e-05,
450
+ "loss": 0.4274,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 8.0,
455
+ "eval_accuracy": 0.5174418604651163,
456
+ "eval_loss": 2.585486650466919,
457
+ "eval_macro_f1": 0.5037820512820513,
458
+ "eval_runtime": 0.4999,
459
+ "eval_samples_per_second": 1376.198,
460
+ "eval_steps_per_second": 86.012,
461
+ "eval_weighted_f1": 0.5023155187835421,
462
+ "step": 1304
463
+ },
464
+ {
465
+ "epoch": 8.129429892141756,
466
+ "grad_norm": 84.29957580566406,
467
+ "learning_rate": 1.3197000681663258e-05,
468
+ "loss": 0.2248,
469
+ "step": 1325
470
+ },
471
+ {
472
+ "epoch": 8.283513097072419,
473
+ "grad_norm": 73.11257934570312,
474
+ "learning_rate": 1.3026584867075667e-05,
475
+ "loss": 0.2109,
476
+ "step": 1350
477
+ },
478
+ {
479
+ "epoch": 8.437596302003081,
480
+ "grad_norm": 102.74080657958984,
481
+ "learning_rate": 1.2856169052488073e-05,
482
+ "loss": 0.3546,
483
+ "step": 1375
484
+ },
485
+ {
486
+ "epoch": 8.591679506933744,
487
+ "grad_norm": 8.767850875854492,
488
+ "learning_rate": 1.2685753237900478e-05,
489
+ "loss": 0.3387,
490
+ "step": 1400
491
+ },
492
+ {
493
+ "epoch": 8.745762711864407,
494
+ "grad_norm": 48.210044860839844,
495
+ "learning_rate": 1.2515337423312886e-05,
496
+ "loss": 0.3095,
497
+ "step": 1425
498
+ },
499
+ {
500
+ "epoch": 8.89984591679507,
501
+ "grad_norm": 2.3532841205596924,
502
+ "learning_rate": 1.2344921608725291e-05,
503
+ "loss": 0.4081,
504
+ "step": 1450
505
+ },
506
+ {
507
+ "epoch": 9.0,
508
+ "eval_accuracy": 0.5232558139534884,
509
+ "eval_loss": 2.916186809539795,
510
+ "eval_macro_f1": 0.5119762253596094,
511
+ "eval_runtime": 0.5058,
512
+ "eval_samples_per_second": 1360.198,
513
+ "eval_steps_per_second": 85.012,
514
+ "eval_weighted_f1": 0.5103025879076712,
515
+ "step": 1467
516
+ },
517
+ {
518
+ "epoch": 9.049306625577811,
519
+ "grad_norm": 4.058147430419922,
520
+ "learning_rate": 1.2174505794137697e-05,
521
+ "loss": 0.2661,
522
+ "step": 1475
523
+ },
524
+ {
525
+ "epoch": 9.203389830508474,
526
+ "grad_norm": 19.173160552978516,
527
+ "learning_rate": 1.2004089979550104e-05,
528
+ "loss": 0.2488,
529
+ "step": 1500
530
+ },
531
+ {
532
+ "epoch": 9.357473035439137,
533
+ "grad_norm": 30.102962493896484,
534
+ "learning_rate": 1.183367416496251e-05,
535
+ "loss": 0.1897,
536
+ "step": 1525
537
+ },
538
+ {
539
+ "epoch": 9.5115562403698,
540
+ "grad_norm": 15.41912841796875,
541
+ "learning_rate": 1.1663258350374915e-05,
542
+ "loss": 0.1972,
543
+ "step": 1550
544
+ },
545
+ {
546
+ "epoch": 9.665639445300462,
547
+ "grad_norm": 92.14854431152344,
548
+ "learning_rate": 1.1492842535787322e-05,
549
+ "loss": 0.18,
550
+ "step": 1575
551
+ },
552
+ {
553
+ "epoch": 9.819722650231125,
554
+ "grad_norm": 26.154415130615234,
555
+ "learning_rate": 1.1322426721199728e-05,
556
+ "loss": 0.183,
557
+ "step": 1600
558
+ },
559
+ {
560
+ "epoch": 9.973805855161787,
561
+ "grad_norm": 26.264793395996094,
562
+ "learning_rate": 1.1152010906612133e-05,
563
+ "loss": 0.4154,
564
+ "step": 1625
565
+ },
566
+ {
567
+ "epoch": 10.0,
568
+ "eval_accuracy": 0.5218023255813954,
569
+ "eval_loss": 3.254833221435547,
570
+ "eval_macro_f1": 0.5190916851157522,
571
+ "eval_runtime": 0.4982,
572
+ "eval_samples_per_second": 1381.032,
573
+ "eval_steps_per_second": 86.315,
574
+ "eval_weighted_f1": 0.517986714773896,
575
+ "step": 1630
576
+ }
577
+ ],
578
+ "logging_steps": 25,
579
+ "max_steps": 3260,
580
+ "num_input_tokens_seen": 0,
581
+ "num_train_epochs": 20,
582
+ "save_steps": 500,
583
+ "stateful_callbacks": {
584
+ "EarlyStoppingCallback": {
585
+ "args": {
586
+ "early_stopping_patience": 3,
587
+ "early_stopping_threshold": 0.0
588
+ },
589
+ "attributes": {
590
+ "early_stopping_patience_counter": 3
591
+ }
592
+ },
593
+ "TrainerControl": {
594
+ "args": {
595
+ "should_epoch_stop": false,
596
+ "should_evaluate": false,
597
+ "should_log": false,
598
+ "should_save": true,
599
+ "should_training_stop": true
600
+ },
601
+ "attributes": {}
602
+ }
603
+ },
604
+ "total_flos": 0.0,
605
+ "train_batch_size": 8,
606
+ "trial_name": null,
607
+ "trial_params": null
608
+ }
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114015Z-b0980c23/trainer/checkpoint-1630/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd74124b85dfaf79ef5694b739606087d18befc4bee11a030503b4441be7f855
3
+ size 5752