PhaaNe commited on
Commit
92e2128
·
verified ·
1 Parent(s): 6ed8677

Add explore checkpoint explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e (job 857da9454d254b3ed5de, factcheck-ablations)

Browse files
Files changed (21) hide show
  1. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/model.safetensors +3 -0
  2. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/model.safetensors +3 -0
  3. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/optimizer.pt +3 -0
  4. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/rng_state.pth +3 -0
  5. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/scheduler.pt +3 -0
  6. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/sentencepiece.bpe.model +3 -0
  7. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/special_tokens_map.json +15 -0
  8. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/tokenizer.json +0 -0
  9. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/tokenizer_config.json +56 -0
  10. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/trainer_state.json +548 -0
  11. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/training_args.bin +3 -0
  12. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/model.safetensors +3 -0
  13. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/optimizer.pt +3 -0
  14. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/rng_state.pth +3 -0
  15. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/scheduler.pt +3 -0
  16. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/sentencepiece.bpe.model +3 -0
  17. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/special_tokens_map.json +15 -0
  18. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/tokenizer.json +0 -0
  19. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/tokenizer_config.json +56 -0
  20. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/trainer_state.json +721 -0
  21. explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/training_args.bin +3 -0
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c8a326b9a3e57d16672403992e44ee46905f73bf0ca01212654a25d4767675e
3
+ size 390297060
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b90feb66e002a7877f235df4a7c634fe132924ea9b36613315b6855784e3943
3
+ size 390297092
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67ba4cf249be885091c314f25754995174f127d8d84e43ea9bdd9b01efbbf31c
3
+ size 775987770
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ed8324653825fabecd5feaa81fc940adfb6f9c068ac9949f7e91b8d82c16292
3
+ size 14244
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6547e97d775b99e301e74072e254c63728fa9f978ba1a733d5ed03438125706e
3
+ size 1064
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/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-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/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-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/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-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/trainer_state.json ADDED
@@ -0,0 +1,548 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1467,
3
+ "best_metric": 0.5430152133707621,
4
+ "best_model_checkpoint": "/root/VNEncoder/explore/workspaces/857da9454d254b3ed5de/results/checkpoints/explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467",
5
+ "epoch": 9.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1467,
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": 24.508975982666016,
15
+ "learning_rate": 1.47239263803681e-06,
16
+ "loss": 4.6751,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.3081664098613251,
21
+ "grad_norm": 29.337697982788086,
22
+ "learning_rate": 3.0061349693251535e-06,
23
+ "loss": 4.5009,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.4622496147919877,
28
+ "grad_norm": 25.202960968017578,
29
+ "learning_rate": 4.539877300613497e-06,
30
+ "loss": 4.4413,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.6163328197226502,
35
+ "grad_norm": 26.34480857849121,
36
+ "learning_rate": 6.073619631901841e-06,
37
+ "loss": 4.4354,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.7704160246533128,
42
+ "grad_norm": 40.48666000366211,
43
+ "learning_rate": 7.6073619631901856e-06,
44
+ "loss": 4.4403,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.9244992295839753,
49
+ "grad_norm": 17.46730613708496,
50
+ "learning_rate": 9.14110429447853e-06,
51
+ "loss": 4.4035,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 1.0,
56
+ "eval_accuracy": 0.44476744186046513,
57
+ "eval_loss": 1.0685410499572754,
58
+ "eval_macro_f1": 0.44361881079499743,
59
+ "eval_runtime": 0.7184,
60
+ "eval_samples_per_second": 957.618,
61
+ "eval_steps_per_second": 59.851,
62
+ "eval_weighted_f1": 0.44286949836632716,
63
+ "step": 163
64
+ },
65
+ {
66
+ "epoch": 1.073959938366718,
67
+ "grad_norm": 27.43915557861328,
68
+ "learning_rate": 1.0674846625766873e-05,
69
+ "loss": 4.2891,
70
+ "step": 175
71
+ },
72
+ {
73
+ "epoch": 1.2280431432973806,
74
+ "grad_norm": 21.196197509765625,
75
+ "learning_rate": 1.2208588957055216e-05,
76
+ "loss": 4.2913,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.3821263482280433,
81
+ "grad_norm": 37.47271728515625,
82
+ "learning_rate": 1.374233128834356e-05,
83
+ "loss": 4.2384,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.5362095531587057,
88
+ "grad_norm": 19.074737548828125,
89
+ "learning_rate": 1.5276073619631903e-05,
90
+ "loss": 4.2496,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.6902927580893683,
95
+ "grad_norm": 34.223453521728516,
96
+ "learning_rate": 1.6809815950920248e-05,
97
+ "loss": 4.2202,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.8443759630200307,
102
+ "grad_norm": 16.862079620361328,
103
+ "learning_rate": 1.834355828220859e-05,
104
+ "loss": 4.2642,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.9984591679506933,
109
+ "grad_norm": 33.562049865722656,
110
+ "learning_rate": 1.9877300613496935e-05,
111
+ "loss": 4.3175,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 2.0,
116
+ "eval_accuracy": 0.4796511627906977,
117
+ "eval_loss": 1.0326141119003296,
118
+ "eval_macro_f1": 0.442770639491951,
119
+ "eval_runtime": 0.754,
120
+ "eval_samples_per_second": 912.441,
121
+ "eval_steps_per_second": 57.028,
122
+ "eval_weighted_f1": 0.4451030901278709,
123
+ "step": 326
124
+ },
125
+ {
126
+ "epoch": 2.147919876733436,
127
+ "grad_norm": 24.802732467651367,
128
+ "learning_rate": 1.9843217450579417e-05,
129
+ "loss": 3.8293,
130
+ "step": 350
131
+ },
132
+ {
133
+ "epoch": 2.3020030816640986,
134
+ "grad_norm": 33.17911911010742,
135
+ "learning_rate": 1.967280163599182e-05,
136
+ "loss": 3.7957,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.4560862865947612,
141
+ "grad_norm": 25.208017349243164,
142
+ "learning_rate": 1.9502385821404228e-05,
143
+ "loss": 3.8292,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.610169491525424,
148
+ "grad_norm": 33.43124771118164,
149
+ "learning_rate": 1.9331970006816635e-05,
150
+ "loss": 3.6697,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.7642526964560865,
155
+ "grad_norm": 33.61294174194336,
156
+ "learning_rate": 1.916155419222904e-05,
157
+ "loss": 3.6829,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.9183359013867487,
162
+ "grad_norm": 35.11015319824219,
163
+ "learning_rate": 1.8991138377641446e-05,
164
+ "loss": 3.7169,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 3.0,
169
+ "eval_accuracy": 0.4883720930232558,
170
+ "eval_loss": 1.0557606220245361,
171
+ "eval_macro_f1": 0.4708796564172643,
172
+ "eval_runtime": 0.5406,
173
+ "eval_samples_per_second": 1272.554,
174
+ "eval_steps_per_second": 79.535,
175
+ "eval_weighted_f1": 0.47293187151118143,
176
+ "step": 489
177
+ },
178
+ {
179
+ "epoch": 3.0677966101694913,
180
+ "grad_norm": 33.60896301269531,
181
+ "learning_rate": 1.8820722563053853e-05,
182
+ "loss": 3.3454,
183
+ "step": 500
184
+ },
185
+ {
186
+ "epoch": 3.221879815100154,
187
+ "grad_norm": 32.06923294067383,
188
+ "learning_rate": 1.8650306748466257e-05,
189
+ "loss": 3.1223,
190
+ "step": 525
191
+ },
192
+ {
193
+ "epoch": 3.3759630200308166,
194
+ "grad_norm": 30.909337997436523,
195
+ "learning_rate": 1.8479890933878664e-05,
196
+ "loss": 2.8725,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.5300462249614792,
201
+ "grad_norm": 33.19773483276367,
202
+ "learning_rate": 1.8309475119291072e-05,
203
+ "loss": 2.8872,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.684129429892142,
208
+ "grad_norm": 29.102251052856445,
209
+ "learning_rate": 1.8139059304703476e-05,
210
+ "loss": 2.8382,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.8382126348228045,
215
+ "grad_norm": 39.861305236816406,
216
+ "learning_rate": 1.7968643490115886e-05,
217
+ "loss": 2.8136,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.992295839753467,
222
+ "grad_norm": 54.482730865478516,
223
+ "learning_rate": 1.779822767552829e-05,
224
+ "loss": 2.9815,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 4.0,
229
+ "eval_accuracy": 0.5465116279069767,
230
+ "eval_loss": 1.0208972692489624,
231
+ "eval_macro_f1": 0.5334543263473636,
232
+ "eval_runtime": 0.53,
233
+ "eval_samples_per_second": 1298.02,
234
+ "eval_steps_per_second": 81.126,
235
+ "eval_weighted_f1": 0.5335384018997117,
236
+ "step": 652
237
+ },
238
+ {
239
+ "epoch": 4.141756548536209,
240
+ "grad_norm": 44.77860641479492,
241
+ "learning_rate": 1.7627811860940697e-05,
242
+ "loss": 2.1073,
243
+ "step": 675
244
+ },
245
+ {
246
+ "epoch": 4.295839753466872,
247
+ "grad_norm": 45.415504455566406,
248
+ "learning_rate": 1.7457396046353105e-05,
249
+ "loss": 1.9004,
250
+ "step": 700
251
+ },
252
+ {
253
+ "epoch": 4.449922958397535,
254
+ "grad_norm": 85.03325653076172,
255
+ "learning_rate": 1.728698023176551e-05,
256
+ "loss": 1.7899,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.604006163328197,
261
+ "grad_norm": 39.80827331542969,
262
+ "learning_rate": 1.7116564417177916e-05,
263
+ "loss": 2.0386,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.75808936825886,
268
+ "grad_norm": 54.64188003540039,
269
+ "learning_rate": 1.6946148602590323e-05,
270
+ "loss": 1.8445,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.9121725731895225,
275
+ "grad_norm": 55.0621223449707,
276
+ "learning_rate": 1.6775732788002727e-05,
277
+ "loss": 2.22,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 5.0,
282
+ "eval_accuracy": 0.5319767441860465,
283
+ "eval_loss": 1.4829277992248535,
284
+ "eval_macro_f1": 0.49842630775196134,
285
+ "eval_runtime": 0.5664,
286
+ "eval_samples_per_second": 1214.734,
287
+ "eval_steps_per_second": 75.921,
288
+ "eval_weighted_f1": 0.4993038473738914,
289
+ "step": 815
290
+ },
291
+ {
292
+ "epoch": 5.061633281972265,
293
+ "grad_norm": 37.069908142089844,
294
+ "learning_rate": 1.6605316973415134e-05,
295
+ "loss": 1.7564,
296
+ "step": 825
297
+ },
298
+ {
299
+ "epoch": 5.215716486902927,
300
+ "grad_norm": 71.45074462890625,
301
+ "learning_rate": 1.643490115882754e-05,
302
+ "loss": 1.1292,
303
+ "step": 850
304
+ },
305
+ {
306
+ "epoch": 5.36979969183359,
307
+ "grad_norm": 84.66228485107422,
308
+ "learning_rate": 1.6264485344239945e-05,
309
+ "loss": 1.0182,
310
+ "step": 875
311
+ },
312
+ {
313
+ "epoch": 5.523882896764253,
314
+ "grad_norm": 146.02886962890625,
315
+ "learning_rate": 1.6094069529652356e-05,
316
+ "loss": 1.4498,
317
+ "step": 900
318
+ },
319
+ {
320
+ "epoch": 5.677966101694915,
321
+ "grad_norm": 41.28833770751953,
322
+ "learning_rate": 1.592365371506476e-05,
323
+ "loss": 1.3913,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.832049306625578,
328
+ "grad_norm": 32.541603088378906,
329
+ "learning_rate": 1.5753237900477167e-05,
330
+ "loss": 1.1054,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.9861325115562405,
335
+ "grad_norm": 68.97745513916016,
336
+ "learning_rate": 1.5582822085889574e-05,
337
+ "loss": 1.3864,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 6.0,
342
+ "eval_accuracy": 0.502906976744186,
343
+ "eval_loss": 1.7800474166870117,
344
+ "eval_macro_f1": 0.48896951675304745,
345
+ "eval_runtime": 0.5519,
346
+ "eval_samples_per_second": 1246.599,
347
+ "eval_steps_per_second": 77.912,
348
+ "eval_weighted_f1": 0.4853569431870788,
349
+ "step": 978
350
+ },
351
+ {
352
+ "epoch": 6.135593220338983,
353
+ "grad_norm": 22.984542846679688,
354
+ "learning_rate": 1.5412406271301978e-05,
355
+ "loss": 0.9296,
356
+ "step": 1000
357
+ },
358
+ {
359
+ "epoch": 6.289676425269645,
360
+ "grad_norm": 55.60026550292969,
361
+ "learning_rate": 1.5241990456714383e-05,
362
+ "loss": 0.686,
363
+ "step": 1025
364
+ },
365
+ {
366
+ "epoch": 6.443759630200308,
367
+ "grad_norm": 67.12779235839844,
368
+ "learning_rate": 1.5071574642126792e-05,
369
+ "loss": 0.7364,
370
+ "step": 1050
371
+ },
372
+ {
373
+ "epoch": 6.597842835130971,
374
+ "grad_norm": 41.4534797668457,
375
+ "learning_rate": 1.4901158827539198e-05,
376
+ "loss": 0.5871,
377
+ "step": 1075
378
+ },
379
+ {
380
+ "epoch": 6.751926040061633,
381
+ "grad_norm": 39.721866607666016,
382
+ "learning_rate": 1.4730743012951603e-05,
383
+ "loss": 0.7935,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.906009244992296,
388
+ "grad_norm": 46.155452728271484,
389
+ "learning_rate": 1.456032719836401e-05,
390
+ "loss": 0.728,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 7.0,
395
+ "eval_accuracy": 0.5348837209302325,
396
+ "eval_loss": 1.8709527254104614,
397
+ "eval_macro_f1": 0.535061587196157,
398
+ "eval_runtime": 0.5132,
399
+ "eval_samples_per_second": 1340.497,
400
+ "eval_steps_per_second": 83.781,
401
+ "eval_weighted_f1": 0.5338957807282007,
402
+ "step": 1141
403
+ },
404
+ {
405
+ "epoch": 7.055469953775039,
406
+ "grad_norm": 26.76979637145996,
407
+ "learning_rate": 1.4389911383776416e-05,
408
+ "loss": 0.5329,
409
+ "step": 1150
410
+ },
411
+ {
412
+ "epoch": 7.209553158705701,
413
+ "grad_norm": 75.73136901855469,
414
+ "learning_rate": 1.4219495569188822e-05,
415
+ "loss": 0.294,
416
+ "step": 1175
417
+ },
418
+ {
419
+ "epoch": 7.363636363636363,
420
+ "grad_norm": 104.0707015991211,
421
+ "learning_rate": 1.4049079754601229e-05,
422
+ "loss": 0.3715,
423
+ "step": 1200
424
+ },
425
+ {
426
+ "epoch": 7.517719568567026,
427
+ "grad_norm": 65.98831176757812,
428
+ "learning_rate": 1.3878663940013635e-05,
429
+ "loss": 0.3996,
430
+ "step": 1225
431
+ },
432
+ {
433
+ "epoch": 7.671802773497689,
434
+ "grad_norm": 34.55245590209961,
435
+ "learning_rate": 1.370824812542604e-05,
436
+ "loss": 0.5399,
437
+ "step": 1250
438
+ },
439
+ {
440
+ "epoch": 7.825885978428351,
441
+ "grad_norm": 29.771799087524414,
442
+ "learning_rate": 1.3537832310838447e-05,
443
+ "loss": 0.4579,
444
+ "step": 1275
445
+ },
446
+ {
447
+ "epoch": 7.979969183359014,
448
+ "grad_norm": 41.84404754638672,
449
+ "learning_rate": 1.3367416496250853e-05,
450
+ "loss": 0.4825,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 8.0,
455
+ "eval_accuracy": 0.5436046511627907,
456
+ "eval_loss": 2.202044725418091,
457
+ "eval_macro_f1": 0.5386058292162003,
458
+ "eval_runtime": 0.5298,
459
+ "eval_samples_per_second": 1298.606,
460
+ "eval_steps_per_second": 81.163,
461
+ "eval_weighted_f1": 0.5376471868109608,
462
+ "step": 1304
463
+ },
464
+ {
465
+ "epoch": 8.129429892141756,
466
+ "grad_norm": 13.249227523803711,
467
+ "learning_rate": 1.3197000681663258e-05,
468
+ "loss": 0.1814,
469
+ "step": 1325
470
+ },
471
+ {
472
+ "epoch": 8.283513097072419,
473
+ "grad_norm": 88.17157745361328,
474
+ "learning_rate": 1.3026584867075667e-05,
475
+ "loss": 0.3641,
476
+ "step": 1350
477
+ },
478
+ {
479
+ "epoch": 8.437596302003081,
480
+ "grad_norm": 65.5908432006836,
481
+ "learning_rate": 1.2856169052488073e-05,
482
+ "loss": 0.2651,
483
+ "step": 1375
484
+ },
485
+ {
486
+ "epoch": 8.591679506933744,
487
+ "grad_norm": 117.96095275878906,
488
+ "learning_rate": 1.2685753237900478e-05,
489
+ "loss": 0.2696,
490
+ "step": 1400
491
+ },
492
+ {
493
+ "epoch": 8.745762711864407,
494
+ "grad_norm": 11.11864948272705,
495
+ "learning_rate": 1.2515337423312886e-05,
496
+ "loss": 0.322,
497
+ "step": 1425
498
+ },
499
+ {
500
+ "epoch": 8.89984591679507,
501
+ "grad_norm": 52.752716064453125,
502
+ "learning_rate": 1.2344921608725291e-05,
503
+ "loss": 0.2328,
504
+ "step": 1450
505
+ },
506
+ {
507
+ "epoch": 9.0,
508
+ "eval_accuracy": 0.5494186046511628,
509
+ "eval_loss": 2.6548359394073486,
510
+ "eval_macro_f1": 0.5430152133707621,
511
+ "eval_runtime": 0.53,
512
+ "eval_samples_per_second": 1298.196,
513
+ "eval_steps_per_second": 81.137,
514
+ "eval_weighted_f1": 0.541976774441087,
515
+ "step": 1467
516
+ }
517
+ ],
518
+ "logging_steps": 25,
519
+ "max_steps": 3260,
520
+ "num_input_tokens_seen": 0,
521
+ "num_train_epochs": 20,
522
+ "save_steps": 500,
523
+ "stateful_callbacks": {
524
+ "EarlyStoppingCallback": {
525
+ "args": {
526
+ "early_stopping_patience": 3,
527
+ "early_stopping_threshold": 0.0
528
+ },
529
+ "attributes": {
530
+ "early_stopping_patience_counter": 0
531
+ }
532
+ },
533
+ "TrainerControl": {
534
+ "args": {
535
+ "should_epoch_stop": false,
536
+ "should_evaluate": false,
537
+ "should_log": false,
538
+ "should_save": true,
539
+ "should_training_stop": false
540
+ },
541
+ "attributes": {}
542
+ }
543
+ },
544
+ "total_flos": 0.0,
545
+ "train_batch_size": 8,
546
+ "trial_name": null,
547
+ "trial_params": null
548
+ }
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05e2bb28099c0cdbd8b455fd78d68c154d18d9422ae40f4a69d2ef165ca41028
3
+ size 5752
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d73eea8a42cd4290f232c0301e30079e5d0aaf0118e466e0541c8b0267b9ec6
3
+ size 390297092
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c2b588635cfbeb14c128b3e6ae29a78bf21934acd54cd8246f401e19fe75338
3
+ size 775987770
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f501779be7085c60ba517593908a233c0df348fb4d83c2b728858bb2a26e122
3
+ size 14244
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4ee466b906d9827e10488372e1a869ea79129c224cae9d22411515f3be28fbe
3
+ size 1064
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/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-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/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-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/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-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/trainer_state.json ADDED
@@ -0,0 +1,721 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1467,
3
+ "best_metric": 0.5430152133707621,
4
+ "best_model_checkpoint": "/root/VNEncoder/explore/workspaces/857da9454d254b3ed5de/results/checkpoints/explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1467",
5
+ "epoch": 12.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1956,
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": 24.508975982666016,
15
+ "learning_rate": 1.47239263803681e-06,
16
+ "loss": 4.6751,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.3081664098613251,
21
+ "grad_norm": 29.337697982788086,
22
+ "learning_rate": 3.0061349693251535e-06,
23
+ "loss": 4.5009,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.4622496147919877,
28
+ "grad_norm": 25.202960968017578,
29
+ "learning_rate": 4.539877300613497e-06,
30
+ "loss": 4.4413,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 0.6163328197226502,
35
+ "grad_norm": 26.34480857849121,
36
+ "learning_rate": 6.073619631901841e-06,
37
+ "loss": 4.4354,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 0.7704160246533128,
42
+ "grad_norm": 40.48666000366211,
43
+ "learning_rate": 7.6073619631901856e-06,
44
+ "loss": 4.4403,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 0.9244992295839753,
49
+ "grad_norm": 17.46730613708496,
50
+ "learning_rate": 9.14110429447853e-06,
51
+ "loss": 4.4035,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 1.0,
56
+ "eval_accuracy": 0.44476744186046513,
57
+ "eval_loss": 1.0685410499572754,
58
+ "eval_macro_f1": 0.44361881079499743,
59
+ "eval_runtime": 0.7184,
60
+ "eval_samples_per_second": 957.618,
61
+ "eval_steps_per_second": 59.851,
62
+ "eval_weighted_f1": 0.44286949836632716,
63
+ "step": 163
64
+ },
65
+ {
66
+ "epoch": 1.073959938366718,
67
+ "grad_norm": 27.43915557861328,
68
+ "learning_rate": 1.0674846625766873e-05,
69
+ "loss": 4.2891,
70
+ "step": 175
71
+ },
72
+ {
73
+ "epoch": 1.2280431432973806,
74
+ "grad_norm": 21.196197509765625,
75
+ "learning_rate": 1.2208588957055216e-05,
76
+ "loss": 4.2913,
77
+ "step": 200
78
+ },
79
+ {
80
+ "epoch": 1.3821263482280433,
81
+ "grad_norm": 37.47271728515625,
82
+ "learning_rate": 1.374233128834356e-05,
83
+ "loss": 4.2384,
84
+ "step": 225
85
+ },
86
+ {
87
+ "epoch": 1.5362095531587057,
88
+ "grad_norm": 19.074737548828125,
89
+ "learning_rate": 1.5276073619631903e-05,
90
+ "loss": 4.2496,
91
+ "step": 250
92
+ },
93
+ {
94
+ "epoch": 1.6902927580893683,
95
+ "grad_norm": 34.223453521728516,
96
+ "learning_rate": 1.6809815950920248e-05,
97
+ "loss": 4.2202,
98
+ "step": 275
99
+ },
100
+ {
101
+ "epoch": 1.8443759630200307,
102
+ "grad_norm": 16.862079620361328,
103
+ "learning_rate": 1.834355828220859e-05,
104
+ "loss": 4.2642,
105
+ "step": 300
106
+ },
107
+ {
108
+ "epoch": 1.9984591679506933,
109
+ "grad_norm": 33.562049865722656,
110
+ "learning_rate": 1.9877300613496935e-05,
111
+ "loss": 4.3175,
112
+ "step": 325
113
+ },
114
+ {
115
+ "epoch": 2.0,
116
+ "eval_accuracy": 0.4796511627906977,
117
+ "eval_loss": 1.0326141119003296,
118
+ "eval_macro_f1": 0.442770639491951,
119
+ "eval_runtime": 0.754,
120
+ "eval_samples_per_second": 912.441,
121
+ "eval_steps_per_second": 57.028,
122
+ "eval_weighted_f1": 0.4451030901278709,
123
+ "step": 326
124
+ },
125
+ {
126
+ "epoch": 2.147919876733436,
127
+ "grad_norm": 24.802732467651367,
128
+ "learning_rate": 1.9843217450579417e-05,
129
+ "loss": 3.8293,
130
+ "step": 350
131
+ },
132
+ {
133
+ "epoch": 2.3020030816640986,
134
+ "grad_norm": 33.17911911010742,
135
+ "learning_rate": 1.967280163599182e-05,
136
+ "loss": 3.7957,
137
+ "step": 375
138
+ },
139
+ {
140
+ "epoch": 2.4560862865947612,
141
+ "grad_norm": 25.208017349243164,
142
+ "learning_rate": 1.9502385821404228e-05,
143
+ "loss": 3.8292,
144
+ "step": 400
145
+ },
146
+ {
147
+ "epoch": 2.610169491525424,
148
+ "grad_norm": 33.43124771118164,
149
+ "learning_rate": 1.9331970006816635e-05,
150
+ "loss": 3.6697,
151
+ "step": 425
152
+ },
153
+ {
154
+ "epoch": 2.7642526964560865,
155
+ "grad_norm": 33.61294174194336,
156
+ "learning_rate": 1.916155419222904e-05,
157
+ "loss": 3.6829,
158
+ "step": 450
159
+ },
160
+ {
161
+ "epoch": 2.9183359013867487,
162
+ "grad_norm": 35.11015319824219,
163
+ "learning_rate": 1.8991138377641446e-05,
164
+ "loss": 3.7169,
165
+ "step": 475
166
+ },
167
+ {
168
+ "epoch": 3.0,
169
+ "eval_accuracy": 0.4883720930232558,
170
+ "eval_loss": 1.0557606220245361,
171
+ "eval_macro_f1": 0.4708796564172643,
172
+ "eval_runtime": 0.5406,
173
+ "eval_samples_per_second": 1272.554,
174
+ "eval_steps_per_second": 79.535,
175
+ "eval_weighted_f1": 0.47293187151118143,
176
+ "step": 489
177
+ },
178
+ {
179
+ "epoch": 3.0677966101694913,
180
+ "grad_norm": 33.60896301269531,
181
+ "learning_rate": 1.8820722563053853e-05,
182
+ "loss": 3.3454,
183
+ "step": 500
184
+ },
185
+ {
186
+ "epoch": 3.221879815100154,
187
+ "grad_norm": 32.06923294067383,
188
+ "learning_rate": 1.8650306748466257e-05,
189
+ "loss": 3.1223,
190
+ "step": 525
191
+ },
192
+ {
193
+ "epoch": 3.3759630200308166,
194
+ "grad_norm": 30.909337997436523,
195
+ "learning_rate": 1.8479890933878664e-05,
196
+ "loss": 2.8725,
197
+ "step": 550
198
+ },
199
+ {
200
+ "epoch": 3.5300462249614792,
201
+ "grad_norm": 33.19773483276367,
202
+ "learning_rate": 1.8309475119291072e-05,
203
+ "loss": 2.8872,
204
+ "step": 575
205
+ },
206
+ {
207
+ "epoch": 3.684129429892142,
208
+ "grad_norm": 29.102251052856445,
209
+ "learning_rate": 1.8139059304703476e-05,
210
+ "loss": 2.8382,
211
+ "step": 600
212
+ },
213
+ {
214
+ "epoch": 3.8382126348228045,
215
+ "grad_norm": 39.861305236816406,
216
+ "learning_rate": 1.7968643490115886e-05,
217
+ "loss": 2.8136,
218
+ "step": 625
219
+ },
220
+ {
221
+ "epoch": 3.992295839753467,
222
+ "grad_norm": 54.482730865478516,
223
+ "learning_rate": 1.779822767552829e-05,
224
+ "loss": 2.9815,
225
+ "step": 650
226
+ },
227
+ {
228
+ "epoch": 4.0,
229
+ "eval_accuracy": 0.5465116279069767,
230
+ "eval_loss": 1.0208972692489624,
231
+ "eval_macro_f1": 0.5334543263473636,
232
+ "eval_runtime": 0.53,
233
+ "eval_samples_per_second": 1298.02,
234
+ "eval_steps_per_second": 81.126,
235
+ "eval_weighted_f1": 0.5335384018997117,
236
+ "step": 652
237
+ },
238
+ {
239
+ "epoch": 4.141756548536209,
240
+ "grad_norm": 44.77860641479492,
241
+ "learning_rate": 1.7627811860940697e-05,
242
+ "loss": 2.1073,
243
+ "step": 675
244
+ },
245
+ {
246
+ "epoch": 4.295839753466872,
247
+ "grad_norm": 45.415504455566406,
248
+ "learning_rate": 1.7457396046353105e-05,
249
+ "loss": 1.9004,
250
+ "step": 700
251
+ },
252
+ {
253
+ "epoch": 4.449922958397535,
254
+ "grad_norm": 85.03325653076172,
255
+ "learning_rate": 1.728698023176551e-05,
256
+ "loss": 1.7899,
257
+ "step": 725
258
+ },
259
+ {
260
+ "epoch": 4.604006163328197,
261
+ "grad_norm": 39.80827331542969,
262
+ "learning_rate": 1.7116564417177916e-05,
263
+ "loss": 2.0386,
264
+ "step": 750
265
+ },
266
+ {
267
+ "epoch": 4.75808936825886,
268
+ "grad_norm": 54.64188003540039,
269
+ "learning_rate": 1.6946148602590323e-05,
270
+ "loss": 1.8445,
271
+ "step": 775
272
+ },
273
+ {
274
+ "epoch": 4.9121725731895225,
275
+ "grad_norm": 55.0621223449707,
276
+ "learning_rate": 1.6775732788002727e-05,
277
+ "loss": 2.22,
278
+ "step": 800
279
+ },
280
+ {
281
+ "epoch": 5.0,
282
+ "eval_accuracy": 0.5319767441860465,
283
+ "eval_loss": 1.4829277992248535,
284
+ "eval_macro_f1": 0.49842630775196134,
285
+ "eval_runtime": 0.5664,
286
+ "eval_samples_per_second": 1214.734,
287
+ "eval_steps_per_second": 75.921,
288
+ "eval_weighted_f1": 0.4993038473738914,
289
+ "step": 815
290
+ },
291
+ {
292
+ "epoch": 5.061633281972265,
293
+ "grad_norm": 37.069908142089844,
294
+ "learning_rate": 1.6605316973415134e-05,
295
+ "loss": 1.7564,
296
+ "step": 825
297
+ },
298
+ {
299
+ "epoch": 5.215716486902927,
300
+ "grad_norm": 71.45074462890625,
301
+ "learning_rate": 1.643490115882754e-05,
302
+ "loss": 1.1292,
303
+ "step": 850
304
+ },
305
+ {
306
+ "epoch": 5.36979969183359,
307
+ "grad_norm": 84.66228485107422,
308
+ "learning_rate": 1.6264485344239945e-05,
309
+ "loss": 1.0182,
310
+ "step": 875
311
+ },
312
+ {
313
+ "epoch": 5.523882896764253,
314
+ "grad_norm": 146.02886962890625,
315
+ "learning_rate": 1.6094069529652356e-05,
316
+ "loss": 1.4498,
317
+ "step": 900
318
+ },
319
+ {
320
+ "epoch": 5.677966101694915,
321
+ "grad_norm": 41.28833770751953,
322
+ "learning_rate": 1.592365371506476e-05,
323
+ "loss": 1.3913,
324
+ "step": 925
325
+ },
326
+ {
327
+ "epoch": 5.832049306625578,
328
+ "grad_norm": 32.541603088378906,
329
+ "learning_rate": 1.5753237900477167e-05,
330
+ "loss": 1.1054,
331
+ "step": 950
332
+ },
333
+ {
334
+ "epoch": 5.9861325115562405,
335
+ "grad_norm": 68.97745513916016,
336
+ "learning_rate": 1.5582822085889574e-05,
337
+ "loss": 1.3864,
338
+ "step": 975
339
+ },
340
+ {
341
+ "epoch": 6.0,
342
+ "eval_accuracy": 0.502906976744186,
343
+ "eval_loss": 1.7800474166870117,
344
+ "eval_macro_f1": 0.48896951675304745,
345
+ "eval_runtime": 0.5519,
346
+ "eval_samples_per_second": 1246.599,
347
+ "eval_steps_per_second": 77.912,
348
+ "eval_weighted_f1": 0.4853569431870788,
349
+ "step": 978
350
+ },
351
+ {
352
+ "epoch": 6.135593220338983,
353
+ "grad_norm": 22.984542846679688,
354
+ "learning_rate": 1.5412406271301978e-05,
355
+ "loss": 0.9296,
356
+ "step": 1000
357
+ },
358
+ {
359
+ "epoch": 6.289676425269645,
360
+ "grad_norm": 55.60026550292969,
361
+ "learning_rate": 1.5241990456714383e-05,
362
+ "loss": 0.686,
363
+ "step": 1025
364
+ },
365
+ {
366
+ "epoch": 6.443759630200308,
367
+ "grad_norm": 67.12779235839844,
368
+ "learning_rate": 1.5071574642126792e-05,
369
+ "loss": 0.7364,
370
+ "step": 1050
371
+ },
372
+ {
373
+ "epoch": 6.597842835130971,
374
+ "grad_norm": 41.4534797668457,
375
+ "learning_rate": 1.4901158827539198e-05,
376
+ "loss": 0.5871,
377
+ "step": 1075
378
+ },
379
+ {
380
+ "epoch": 6.751926040061633,
381
+ "grad_norm": 39.721866607666016,
382
+ "learning_rate": 1.4730743012951603e-05,
383
+ "loss": 0.7935,
384
+ "step": 1100
385
+ },
386
+ {
387
+ "epoch": 6.906009244992296,
388
+ "grad_norm": 46.155452728271484,
389
+ "learning_rate": 1.456032719836401e-05,
390
+ "loss": 0.728,
391
+ "step": 1125
392
+ },
393
+ {
394
+ "epoch": 7.0,
395
+ "eval_accuracy": 0.5348837209302325,
396
+ "eval_loss": 1.8709527254104614,
397
+ "eval_macro_f1": 0.535061587196157,
398
+ "eval_runtime": 0.5132,
399
+ "eval_samples_per_second": 1340.497,
400
+ "eval_steps_per_second": 83.781,
401
+ "eval_weighted_f1": 0.5338957807282007,
402
+ "step": 1141
403
+ },
404
+ {
405
+ "epoch": 7.055469953775039,
406
+ "grad_norm": 26.76979637145996,
407
+ "learning_rate": 1.4389911383776416e-05,
408
+ "loss": 0.5329,
409
+ "step": 1150
410
+ },
411
+ {
412
+ "epoch": 7.209553158705701,
413
+ "grad_norm": 75.73136901855469,
414
+ "learning_rate": 1.4219495569188822e-05,
415
+ "loss": 0.294,
416
+ "step": 1175
417
+ },
418
+ {
419
+ "epoch": 7.363636363636363,
420
+ "grad_norm": 104.0707015991211,
421
+ "learning_rate": 1.4049079754601229e-05,
422
+ "loss": 0.3715,
423
+ "step": 1200
424
+ },
425
+ {
426
+ "epoch": 7.517719568567026,
427
+ "grad_norm": 65.98831176757812,
428
+ "learning_rate": 1.3878663940013635e-05,
429
+ "loss": 0.3996,
430
+ "step": 1225
431
+ },
432
+ {
433
+ "epoch": 7.671802773497689,
434
+ "grad_norm": 34.55245590209961,
435
+ "learning_rate": 1.370824812542604e-05,
436
+ "loss": 0.5399,
437
+ "step": 1250
438
+ },
439
+ {
440
+ "epoch": 7.825885978428351,
441
+ "grad_norm": 29.771799087524414,
442
+ "learning_rate": 1.3537832310838447e-05,
443
+ "loss": 0.4579,
444
+ "step": 1275
445
+ },
446
+ {
447
+ "epoch": 7.979969183359014,
448
+ "grad_norm": 41.84404754638672,
449
+ "learning_rate": 1.3367416496250853e-05,
450
+ "loss": 0.4825,
451
+ "step": 1300
452
+ },
453
+ {
454
+ "epoch": 8.0,
455
+ "eval_accuracy": 0.5436046511627907,
456
+ "eval_loss": 2.202044725418091,
457
+ "eval_macro_f1": 0.5386058292162003,
458
+ "eval_runtime": 0.5298,
459
+ "eval_samples_per_second": 1298.606,
460
+ "eval_steps_per_second": 81.163,
461
+ "eval_weighted_f1": 0.5376471868109608,
462
+ "step": 1304
463
+ },
464
+ {
465
+ "epoch": 8.129429892141756,
466
+ "grad_norm": 13.249227523803711,
467
+ "learning_rate": 1.3197000681663258e-05,
468
+ "loss": 0.1814,
469
+ "step": 1325
470
+ },
471
+ {
472
+ "epoch": 8.283513097072419,
473
+ "grad_norm": 88.17157745361328,
474
+ "learning_rate": 1.3026584867075667e-05,
475
+ "loss": 0.3641,
476
+ "step": 1350
477
+ },
478
+ {
479
+ "epoch": 8.437596302003081,
480
+ "grad_norm": 65.5908432006836,
481
+ "learning_rate": 1.2856169052488073e-05,
482
+ "loss": 0.2651,
483
+ "step": 1375
484
+ },
485
+ {
486
+ "epoch": 8.591679506933744,
487
+ "grad_norm": 117.96095275878906,
488
+ "learning_rate": 1.2685753237900478e-05,
489
+ "loss": 0.2696,
490
+ "step": 1400
491
+ },
492
+ {
493
+ "epoch": 8.745762711864407,
494
+ "grad_norm": 11.11864948272705,
495
+ "learning_rate": 1.2515337423312886e-05,
496
+ "loss": 0.322,
497
+ "step": 1425
498
+ },
499
+ {
500
+ "epoch": 8.89984591679507,
501
+ "grad_norm": 52.752716064453125,
502
+ "learning_rate": 1.2344921608725291e-05,
503
+ "loss": 0.2328,
504
+ "step": 1450
505
+ },
506
+ {
507
+ "epoch": 9.0,
508
+ "eval_accuracy": 0.5494186046511628,
509
+ "eval_loss": 2.6548359394073486,
510
+ "eval_macro_f1": 0.5430152133707621,
511
+ "eval_runtime": 0.53,
512
+ "eval_samples_per_second": 1298.196,
513
+ "eval_steps_per_second": 81.137,
514
+ "eval_weighted_f1": 0.541976774441087,
515
+ "step": 1467
516
+ },
517
+ {
518
+ "epoch": 9.049306625577811,
519
+ "grad_norm": 1.2443686723709106,
520
+ "learning_rate": 1.2174505794137697e-05,
521
+ "loss": 0.1952,
522
+ "step": 1475
523
+ },
524
+ {
525
+ "epoch": 9.203389830508474,
526
+ "grad_norm": 73.73765563964844,
527
+ "learning_rate": 1.2004089979550104e-05,
528
+ "loss": 0.1259,
529
+ "step": 1500
530
+ },
531
+ {
532
+ "epoch": 9.357473035439137,
533
+ "grad_norm": 68.92479705810547,
534
+ "learning_rate": 1.183367416496251e-05,
535
+ "loss": 0.1772,
536
+ "step": 1525
537
+ },
538
+ {
539
+ "epoch": 9.5115562403698,
540
+ "grad_norm": 9.42828369140625,
541
+ "learning_rate": 1.1663258350374915e-05,
542
+ "loss": 0.2231,
543
+ "step": 1550
544
+ },
545
+ {
546
+ "epoch": 9.665639445300462,
547
+ "grad_norm": 120.51764678955078,
548
+ "learning_rate": 1.1492842535787322e-05,
549
+ "loss": 0.2227,
550
+ "step": 1575
551
+ },
552
+ {
553
+ "epoch": 9.819722650231125,
554
+ "grad_norm": 137.04330444335938,
555
+ "learning_rate": 1.1322426721199728e-05,
556
+ "loss": 0.2063,
557
+ "step": 1600
558
+ },
559
+ {
560
+ "epoch": 9.973805855161787,
561
+ "grad_norm": 15.55699634552002,
562
+ "learning_rate": 1.1152010906612133e-05,
563
+ "loss": 0.2486,
564
+ "step": 1625
565
+ },
566
+ {
567
+ "epoch": 10.0,
568
+ "eval_accuracy": 0.5261627906976745,
569
+ "eval_loss": 3.3866662979125977,
570
+ "eval_macro_f1": 0.5289705303068459,
571
+ "eval_runtime": 0.5087,
572
+ "eval_samples_per_second": 1352.398,
573
+ "eval_steps_per_second": 84.525,
574
+ "eval_weighted_f1": 0.5274775701831181,
575
+ "step": 1630
576
+ },
577
+ {
578
+ "epoch": 10.12326656394453,
579
+ "grad_norm": 1.3246686458587646,
580
+ "learning_rate": 1.0981595092024542e-05,
581
+ "loss": 0.1534,
582
+ "step": 1650
583
+ },
584
+ {
585
+ "epoch": 10.277349768875192,
586
+ "grad_norm": 46.461814880371094,
587
+ "learning_rate": 1.0811179277436948e-05,
588
+ "loss": 0.1029,
589
+ "step": 1675
590
+ },
591
+ {
592
+ "epoch": 10.431432973805855,
593
+ "grad_norm": 81.67166900634766,
594
+ "learning_rate": 1.0640763462849353e-05,
595
+ "loss": 0.1326,
596
+ "step": 1700
597
+ },
598
+ {
599
+ "epoch": 10.585516178736517,
600
+ "grad_norm": 114.39840698242188,
601
+ "learning_rate": 1.047034764826176e-05,
602
+ "loss": 0.2421,
603
+ "step": 1725
604
+ },
605
+ {
606
+ "epoch": 10.73959938366718,
607
+ "grad_norm": 107.1532211303711,
608
+ "learning_rate": 1.0299931833674166e-05,
609
+ "loss": 0.1713,
610
+ "step": 1750
611
+ },
612
+ {
613
+ "epoch": 10.893682588597843,
614
+ "grad_norm": 115.2070541381836,
615
+ "learning_rate": 1.0129516019086572e-05,
616
+ "loss": 0.2655,
617
+ "step": 1775
618
+ },
619
+ {
620
+ "epoch": 11.0,
621
+ "eval_accuracy": 0.5465116279069767,
622
+ "eval_loss": 3.7149198055267334,
623
+ "eval_macro_f1": 0.5338052424849481,
624
+ "eval_runtime": 0.6843,
625
+ "eval_samples_per_second": 1005.413,
626
+ "eval_steps_per_second": 62.838,
627
+ "eval_weighted_f1": 0.533799837067824,
628
+ "step": 1793
629
+ },
630
+ {
631
+ "epoch": 11.043143297380585,
632
+ "grad_norm": 123.80921173095703,
633
+ "learning_rate": 9.959100204498979e-06,
634
+ "loss": 0.1339,
635
+ "step": 1800
636
+ },
637
+ {
638
+ "epoch": 11.197226502311247,
639
+ "grad_norm": 135.196044921875,
640
+ "learning_rate": 9.788684389911385e-06,
641
+ "loss": 0.124,
642
+ "step": 1825
643
+ },
644
+ {
645
+ "epoch": 11.35130970724191,
646
+ "grad_norm": 0.667020857334137,
647
+ "learning_rate": 9.61826857532379e-06,
648
+ "loss": 0.1576,
649
+ "step": 1850
650
+ },
651
+ {
652
+ "epoch": 11.505392912172573,
653
+ "grad_norm": 0.04211437329649925,
654
+ "learning_rate": 9.447852760736197e-06,
655
+ "loss": 0.1948,
656
+ "step": 1875
657
+ },
658
+ {
659
+ "epoch": 11.659476117103235,
660
+ "grad_norm": 29.547664642333984,
661
+ "learning_rate": 9.277436946148603e-06,
662
+ "loss": 0.0903,
663
+ "step": 1900
664
+ },
665
+ {
666
+ "epoch": 11.813559322033898,
667
+ "grad_norm": 0.4731750190258026,
668
+ "learning_rate": 9.107021131561008e-06,
669
+ "loss": 0.0565,
670
+ "step": 1925
671
+ },
672
+ {
673
+ "epoch": 11.96764252696456,
674
+ "grad_norm": 1.8464325666427612,
675
+ "learning_rate": 8.936605316973416e-06,
676
+ "loss": 0.0813,
677
+ "step": 1950
678
+ },
679
+ {
680
+ "epoch": 12.0,
681
+ "eval_accuracy": 0.5406976744186046,
682
+ "eval_loss": 4.013413429260254,
683
+ "eval_macro_f1": 0.5385444905838144,
684
+ "eval_runtime": 0.5248,
685
+ "eval_samples_per_second": 1311.009,
686
+ "eval_steps_per_second": 81.938,
687
+ "eval_weighted_f1": 0.5375030929261781,
688
+ "step": 1956
689
+ }
690
+ ],
691
+ "logging_steps": 25,
692
+ "max_steps": 3260,
693
+ "num_input_tokens_seen": 0,
694
+ "num_train_epochs": 20,
695
+ "save_steps": 500,
696
+ "stateful_callbacks": {
697
+ "EarlyStoppingCallback": {
698
+ "args": {
699
+ "early_stopping_patience": 3,
700
+ "early_stopping_threshold": 0.0
701
+ },
702
+ "attributes": {
703
+ "early_stopping_patience_counter": 3
704
+ }
705
+ },
706
+ "TrainerControl": {
707
+ "args": {
708
+ "should_epoch_stop": false,
709
+ "should_evaluate": false,
710
+ "should_log": false,
711
+ "should_save": true,
712
+ "should_training_stop": true
713
+ },
714
+ "attributes": {}
715
+ }
716
+ },
717
+ "total_flos": 0.0,
718
+ "train_batch_size": 8,
719
+ "trial_name": null,
720
+ "trial_params": null
721
+ }
explore-factcheck-ablations-vifactcheck-visobert-s42-20261114T114317Z-8940293e/trainer/checkpoint-1956/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05e2bb28099c0cdbd8b455fd78d68c154d18d9422ae40f4a69d2ef165ca41028
3
+ size 5752