Denash commited on
Commit
433c0e5
·
verified ·
1 Parent(s): decd2e0

Upload folder using huggingface_hub

Browse files
checkpoint-1098/config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "RobertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "dtype": "float32",
10
+ "eos_token_id": 2,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "safe",
16
+ "1": "vulnerable"
17
+ },
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 3072,
20
+ "is_decoder": false,
21
+ "label2id": {
22
+ "safe": 0,
23
+ "vulnerable": 1
24
+ },
25
+ "layer_norm_eps": 1e-05,
26
+ "max_position_embeddings": 514,
27
+ "model_type": "roberta",
28
+ "num_attention_heads": 12,
29
+ "num_hidden_layers": 12,
30
+ "output_past": true,
31
+ "pad_token_id": 1,
32
+ "problem_type": "single_label_classification",
33
+ "tie_word_embeddings": true,
34
+ "transformers_version": "5.13.1",
35
+ "type_vocab_size": 1,
36
+ "use_cache": false,
37
+ "vocab_size": 50265
38
+ }
checkpoint-1098/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2728b4eb8a1f3468ec296637e61b7ab25deb462d37576aa36889a5c995c212c5
3
+ size 498612824
checkpoint-1098/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b46bcc010a2807812561c5a14d63ce31eb92864a0a1f667b8dc334486121e94
3
+ size 997340299
checkpoint-1098/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ff111fdfe540b8065ce1b16a952b79150dd256469f1ea038289b296888a48ca
3
+ size 14455
checkpoint-1098/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b380ae465471816e4d19e30825b7f6419d5cce1095c8b10588c4cd345a6456e
3
+ size 1465
checkpoint-1098/trainer_state.json ADDED
@@ -0,0 +1,878 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 366,
3
+ "best_metric": 0.41911664605140686,
4
+ "best_model_checkpoint": "training_output/run_20260713_203733/model/checkpoint-366",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1098,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0273224043715847,
14
+ "grad_norm": 4.927176475524902,
15
+ "learning_rate": 7.377049180327868e-07,
16
+ "loss": 0.6899909496307373,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.0546448087431694,
21
+ "grad_norm": 6.0230512619018555,
22
+ "learning_rate": 1.557377049180328e-06,
23
+ "loss": 0.7346426486968994,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.08196721311475409,
28
+ "grad_norm": 6.2535271644592285,
29
+ "learning_rate": 2.377049180327869e-06,
30
+ "loss": 0.6693419456481934,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.1092896174863388,
35
+ "grad_norm": 10.038297653198242,
36
+ "learning_rate": 3.1967213114754097e-06,
37
+ "loss": 0.6888828277587891,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.1366120218579235,
42
+ "grad_norm": 10.041961669921875,
43
+ "learning_rate": 4.016393442622951e-06,
44
+ "loss": 0.623580026626587,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.16393442622950818,
49
+ "grad_norm": 9.107032775878906,
50
+ "learning_rate": 4.836065573770492e-06,
51
+ "loss": 0.692430830001831,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.1912568306010929,
56
+ "grad_norm": 9.090417861938477,
57
+ "learning_rate": 5.655737704918032e-06,
58
+ "loss": 0.6110644817352295,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.2185792349726776,
63
+ "grad_norm": 7.914913177490234,
64
+ "learning_rate": 6.4754098360655735e-06,
65
+ "loss": 0.6268288612365722,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.2459016393442623,
70
+ "grad_norm": 8.18041706085205,
71
+ "learning_rate": 7.2950819672131145e-06,
72
+ "loss": 0.6859075546264648,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.273224043715847,
77
+ "grad_norm": 5.7157392501831055,
78
+ "learning_rate": 8.114754098360657e-06,
79
+ "loss": 0.7717578887939454,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.3005464480874317,
84
+ "grad_norm": 11.109936714172363,
85
+ "learning_rate": 8.934426229508197e-06,
86
+ "loss": 0.6476659774780273,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.32786885245901637,
91
+ "grad_norm": 10.369423866271973,
92
+ "learning_rate": 9.754098360655738e-06,
93
+ "loss": 0.6707509994506836,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.3551912568306011,
98
+ "grad_norm": 10.730874061584473,
99
+ "learning_rate": 1.0573770491803279e-05,
100
+ "loss": 0.6787842750549317,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.3825136612021858,
105
+ "grad_norm": 8.120079040527344,
106
+ "learning_rate": 1.139344262295082e-05,
107
+ "loss": 0.6765446186065673,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.4098360655737705,
112
+ "grad_norm": 14.429889678955078,
113
+ "learning_rate": 1.221311475409836e-05,
114
+ "loss": 0.623370361328125,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.4371584699453552,
119
+ "grad_norm": 9.193517684936523,
120
+ "learning_rate": 1.3032786885245902e-05,
121
+ "loss": 0.6362186908721924,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.4644808743169399,
126
+ "grad_norm": 8.7772855758667,
127
+ "learning_rate": 1.3852459016393443e-05,
128
+ "loss": 0.6641538619995118,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.4918032786885246,
133
+ "grad_norm": 5.2151780128479,
134
+ "learning_rate": 1.4672131147540984e-05,
135
+ "loss": 0.6653701782226562,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.5191256830601093,
140
+ "grad_norm": 17.197834014892578,
141
+ "learning_rate": 1.5491803278688525e-05,
142
+ "loss": 0.6582591533660889,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.546448087431694,
147
+ "grad_norm": 7.26473331451416,
148
+ "learning_rate": 1.6311475409836068e-05,
149
+ "loss": 0.765175724029541,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.5737704918032787,
154
+ "grad_norm": 5.47290563583374,
155
+ "learning_rate": 1.7131147540983607e-05,
156
+ "loss": 0.676247501373291,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.6010928961748634,
161
+ "grad_norm": 4.55684232711792,
162
+ "learning_rate": 1.7950819672131146e-05,
163
+ "loss": 0.6748649597167968,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.6284153005464481,
168
+ "grad_norm": 9.52070140838623,
169
+ "learning_rate": 1.877049180327869e-05,
170
+ "loss": 0.7023877143859864,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.6557377049180327,
175
+ "grad_norm": 10.813623428344727,
176
+ "learning_rate": 1.959016393442623e-05,
177
+ "loss": 0.5075790405273437,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.6830601092896175,
182
+ "grad_norm": 15.77322769165039,
183
+ "learning_rate": 2.040983606557377e-05,
184
+ "loss": 0.5947454929351806,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.7103825136612022,
189
+ "grad_norm": 14.141736030578613,
190
+ "learning_rate": 2.122950819672131e-05,
191
+ "loss": 0.7629669666290283,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.7377049180327869,
196
+ "grad_norm": 0.5761239528656006,
197
+ "learning_rate": 2.2049180327868853e-05,
198
+ "loss": 0.717600679397583,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.7650273224043715,
203
+ "grad_norm": 13.026739120483398,
204
+ "learning_rate": 2.2868852459016393e-05,
205
+ "loss": 1.5919431686401366,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.7923497267759563,
210
+ "grad_norm": 12.15133285522461,
211
+ "learning_rate": 2.3688524590163936e-05,
212
+ "loss": 0.6558570861816406,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.819672131147541,
217
+ "grad_norm": 20.35292625427246,
218
+ "learning_rate": 2.4508196721311478e-05,
219
+ "loss": 0.6991987705230713,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.8469945355191257,
224
+ "grad_norm": 9.59134292602539,
225
+ "learning_rate": 2.5327868852459018e-05,
226
+ "loss": 1.1239334106445313,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.8743169398907104,
231
+ "grad_norm": 4.265232086181641,
232
+ "learning_rate": 2.6147540983606557e-05,
233
+ "loss": 0.4866987705230713,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.9016393442622951,
238
+ "grad_norm": 9.87460708618164,
239
+ "learning_rate": 2.69672131147541e-05,
240
+ "loss": 0.7435293674468995,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.9289617486338798,
245
+ "grad_norm": 12.707701683044434,
246
+ "learning_rate": 2.7786885245901642e-05,
247
+ "loss": 0.9141542434692382,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.9562841530054644,
252
+ "grad_norm": 2.9138967990875244,
253
+ "learning_rate": 2.860655737704918e-05,
254
+ "loss": 0.5319347858428956,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.9836065573770492,
259
+ "grad_norm": 6.917504787445068,
260
+ "learning_rate": 2.942622950819672e-05,
261
+ "loss": 0.5438683509826661,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 1.0,
266
+ "eval_accuracy": 0.8492,
267
+ "eval_confusion_matrix": [
268
+ [
269
+ 282,
270
+ 0
271
+ ],
272
+ [
273
+ 54,
274
+ 22
275
+ ]
276
+ ],
277
+ "eval_f1": 0.449,
278
+ "eval_loss": 0.41911664605140686,
279
+ "eval_pr_auc": 0.6187,
280
+ "eval_precision": 1.0,
281
+ "eval_recall": 0.2895,
282
+ "eval_roc_auc": 0.7485,
283
+ "eval_runtime": 138.7598,
284
+ "eval_samples_per_second": 2.58,
285
+ "eval_steps_per_second": 0.649,
286
+ "step": 366
287
+ },
288
+ {
289
+ "epoch": 1.010928961748634,
290
+ "grad_norm": 6.519931316375732,
291
+ "learning_rate": 2.9972677595628417e-05,
292
+ "loss": 0.796769380569458,
293
+ "step": 370
294
+ },
295
+ {
296
+ "epoch": 1.0382513661202186,
297
+ "grad_norm": 4.874457359313965,
298
+ "learning_rate": 2.98816029143898e-05,
299
+ "loss": 0.398636531829834,
300
+ "step": 380
301
+ },
302
+ {
303
+ "epoch": 1.0655737704918034,
304
+ "grad_norm": 15.488590240478516,
305
+ "learning_rate": 2.9790528233151184e-05,
306
+ "loss": 0.7061968803405761,
307
+ "step": 390
308
+ },
309
+ {
310
+ "epoch": 1.092896174863388,
311
+ "grad_norm": 19.74693489074707,
312
+ "learning_rate": 2.9699453551912568e-05,
313
+ "loss": 0.5211454391479492,
314
+ "step": 400
315
+ },
316
+ {
317
+ "epoch": 1.1202185792349726,
318
+ "grad_norm": 2.513188362121582,
319
+ "learning_rate": 2.9608378870673954e-05,
320
+ "loss": 0.7898036003112793,
321
+ "step": 410
322
+ },
323
+ {
324
+ "epoch": 1.1475409836065573,
325
+ "grad_norm": 25.31527328491211,
326
+ "learning_rate": 2.9517304189435338e-05,
327
+ "loss": 0.8195189476013184,
328
+ "step": 420
329
+ },
330
+ {
331
+ "epoch": 1.174863387978142,
332
+ "grad_norm": 26.276199340820312,
333
+ "learning_rate": 2.942622950819672e-05,
334
+ "loss": 0.8042145729064941,
335
+ "step": 430
336
+ },
337
+ {
338
+ "epoch": 1.2021857923497268,
339
+ "grad_norm": 19.838821411132812,
340
+ "learning_rate": 2.9335154826958105e-05,
341
+ "loss": 0.5532180786132812,
342
+ "step": 440
343
+ },
344
+ {
345
+ "epoch": 1.2295081967213115,
346
+ "grad_norm": 1.5704028606414795,
347
+ "learning_rate": 2.924408014571949e-05,
348
+ "loss": 0.46767144203186034,
349
+ "step": 450
350
+ },
351
+ {
352
+ "epoch": 1.2568306010928962,
353
+ "grad_norm": 0.24576333165168762,
354
+ "learning_rate": 2.9153005464480875e-05,
355
+ "loss": 0.7125775337219238,
356
+ "step": 460
357
+ },
358
+ {
359
+ "epoch": 1.2841530054644807,
360
+ "grad_norm": 2.8051435947418213,
361
+ "learning_rate": 2.9061930783242258e-05,
362
+ "loss": 1.4706084251403808,
363
+ "step": 470
364
+ },
365
+ {
366
+ "epoch": 1.3114754098360657,
367
+ "grad_norm": 2.1515614986419678,
368
+ "learning_rate": 2.897085610200364e-05,
369
+ "loss": 0.740934181213379,
370
+ "step": 480
371
+ },
372
+ {
373
+ "epoch": 1.3387978142076502,
374
+ "grad_norm": 5.335806846618652,
375
+ "learning_rate": 2.8879781420765028e-05,
376
+ "loss": 0.5567354679107666,
377
+ "step": 490
378
+ },
379
+ {
380
+ "epoch": 1.366120218579235,
381
+ "grad_norm": 9.499934196472168,
382
+ "learning_rate": 2.878870673952641e-05,
383
+ "loss": 0.5102160930633545,
384
+ "step": 500
385
+ },
386
+ {
387
+ "epoch": 1.3934426229508197,
388
+ "grad_norm": 0.30213791131973267,
389
+ "learning_rate": 2.8697632058287795e-05,
390
+ "loss": 0.8127809524536133,
391
+ "step": 510
392
+ },
393
+ {
394
+ "epoch": 1.4207650273224044,
395
+ "grad_norm": 17.827165603637695,
396
+ "learning_rate": 2.860655737704918e-05,
397
+ "loss": 1.3400749206542968,
398
+ "step": 520
399
+ },
400
+ {
401
+ "epoch": 1.4480874316939891,
402
+ "grad_norm": 2.58740234375,
403
+ "learning_rate": 2.8515482695810565e-05,
404
+ "loss": 0.5003597736358643,
405
+ "step": 530
406
+ },
407
+ {
408
+ "epoch": 1.4754098360655736,
409
+ "grad_norm": 22.007869720458984,
410
+ "learning_rate": 2.842440801457195e-05,
411
+ "loss": 1.6859323501586914,
412
+ "step": 540
413
+ },
414
+ {
415
+ "epoch": 1.5027322404371586,
416
+ "grad_norm": 14.658230781555176,
417
+ "learning_rate": 2.8333333333333332e-05,
418
+ "loss": 0.9195033073425293,
419
+ "step": 550
420
+ },
421
+ {
422
+ "epoch": 1.530054644808743,
423
+ "grad_norm": 1.4521483182907104,
424
+ "learning_rate": 2.8242258652094715e-05,
425
+ "loss": 0.4108741283416748,
426
+ "step": 560
427
+ },
428
+ {
429
+ "epoch": 1.5573770491803278,
430
+ "grad_norm": 10.377533912658691,
431
+ "learning_rate": 2.8151183970856102e-05,
432
+ "loss": 0.655418586730957,
433
+ "step": 570
434
+ },
435
+ {
436
+ "epoch": 1.5846994535519126,
437
+ "grad_norm": 0.5859184265136719,
438
+ "learning_rate": 2.8060109289617486e-05,
439
+ "loss": 1.2600435256958007,
440
+ "step": 580
441
+ },
442
+ {
443
+ "epoch": 1.6120218579234973,
444
+ "grad_norm": 0.8876230120658875,
445
+ "learning_rate": 2.796903460837887e-05,
446
+ "loss": 1.0982278823852538,
447
+ "step": 590
448
+ },
449
+ {
450
+ "epoch": 1.639344262295082,
451
+ "grad_norm": 13.285502433776855,
452
+ "learning_rate": 2.7877959927140256e-05,
453
+ "loss": 0.6685476779937745,
454
+ "step": 600
455
+ },
456
+ {
457
+ "epoch": 1.6666666666666665,
458
+ "grad_norm": 0.6555774211883545,
459
+ "learning_rate": 2.7786885245901642e-05,
460
+ "loss": 0.19812407493591308,
461
+ "step": 610
462
+ },
463
+ {
464
+ "epoch": 1.6939890710382515,
465
+ "grad_norm": 2.7470834255218506,
466
+ "learning_rate": 2.7695810564663026e-05,
467
+ "loss": 1.1334525108337403,
468
+ "step": 620
469
+ },
470
+ {
471
+ "epoch": 1.721311475409836,
472
+ "grad_norm": 14.42418098449707,
473
+ "learning_rate": 2.760473588342441e-05,
474
+ "loss": 0.9498401641845703,
475
+ "step": 630
476
+ },
477
+ {
478
+ "epoch": 1.748633879781421,
479
+ "grad_norm": 57.35378646850586,
480
+ "learning_rate": 2.7513661202185793e-05,
481
+ "loss": 1.2678983688354493,
482
+ "step": 640
483
+ },
484
+ {
485
+ "epoch": 1.7759562841530054,
486
+ "grad_norm": 0.7189310789108276,
487
+ "learning_rate": 2.742258652094718e-05,
488
+ "loss": 0.6106186389923096,
489
+ "step": 650
490
+ },
491
+ {
492
+ "epoch": 1.8032786885245902,
493
+ "grad_norm": 13.546385765075684,
494
+ "learning_rate": 2.7331511839708563e-05,
495
+ "loss": 1.4015119552612305,
496
+ "step": 660
497
+ },
498
+ {
499
+ "epoch": 1.830601092896175,
500
+ "grad_norm": 8.754522323608398,
501
+ "learning_rate": 2.7240437158469946e-05,
502
+ "loss": 0.9790468215942383,
503
+ "step": 670
504
+ },
505
+ {
506
+ "epoch": 1.8579234972677594,
507
+ "grad_norm": 13.214091300964355,
508
+ "learning_rate": 2.7149362477231333e-05,
509
+ "loss": 0.6752419948577881,
510
+ "step": 680
511
+ },
512
+ {
513
+ "epoch": 1.8852459016393444,
514
+ "grad_norm": 2.506131887435913,
515
+ "learning_rate": 2.7058287795992716e-05,
516
+ "loss": 0.2787914752960205,
517
+ "step": 690
518
+ },
519
+ {
520
+ "epoch": 1.9125683060109289,
521
+ "grad_norm": 0.4683574438095093,
522
+ "learning_rate": 2.69672131147541e-05,
523
+ "loss": 0.8134347915649414,
524
+ "step": 700
525
+ },
526
+ {
527
+ "epoch": 1.9398907103825138,
528
+ "grad_norm": 0.8958243131637573,
529
+ "learning_rate": 2.6876138433515483e-05,
530
+ "loss": 0.4489922046661377,
531
+ "step": 710
532
+ },
533
+ {
534
+ "epoch": 1.9672131147540983,
535
+ "grad_norm": 95.41824340820312,
536
+ "learning_rate": 2.678506375227687e-05,
537
+ "loss": 0.9731103897094726,
538
+ "step": 720
539
+ },
540
+ {
541
+ "epoch": 1.994535519125683,
542
+ "grad_norm": 0.7653670907020569,
543
+ "learning_rate": 2.6693989071038253e-05,
544
+ "loss": 1.0658687591552733,
545
+ "step": 730
546
+ },
547
+ {
548
+ "epoch": 2.0,
549
+ "eval_accuracy": 0.8659,
550
+ "eval_confusion_matrix": [
551
+ [
552
+ 281,
553
+ 1
554
+ ],
555
+ [
556
+ 47,
557
+ 29
558
+ ]
559
+ ],
560
+ "eval_f1": 0.5472,
561
+ "eval_loss": 0.45224741101264954,
562
+ "eval_pr_auc": 0.6674,
563
+ "eval_precision": 0.9667,
564
+ "eval_recall": 0.3816,
565
+ "eval_roc_auc": 0.8086,
566
+ "eval_runtime": 137.9939,
567
+ "eval_samples_per_second": 2.594,
568
+ "eval_steps_per_second": 0.652,
569
+ "step": 732
570
+ },
571
+ {
572
+ "epoch": 2.021857923497268,
573
+ "grad_norm": 32.09031677246094,
574
+ "learning_rate": 2.6602914389799637e-05,
575
+ "loss": 0.7306198120117188,
576
+ "step": 740
577
+ },
578
+ {
579
+ "epoch": 2.0491803278688523,
580
+ "grad_norm": 51.15188217163086,
581
+ "learning_rate": 2.651183970856102e-05,
582
+ "loss": 0.49246606826782224,
583
+ "step": 750
584
+ },
585
+ {
586
+ "epoch": 2.0765027322404372,
587
+ "grad_norm": 0.5293377041816711,
588
+ "learning_rate": 2.6420765027322407e-05,
589
+ "loss": 0.30229816436767576,
590
+ "step": 760
591
+ },
592
+ {
593
+ "epoch": 2.1038251366120218,
594
+ "grad_norm": 0.5600886344909668,
595
+ "learning_rate": 2.632969034608379e-05,
596
+ "loss": 1.428311824798584,
597
+ "step": 770
598
+ },
599
+ {
600
+ "epoch": 2.1311475409836067,
601
+ "grad_norm": 8.111888885498047,
602
+ "learning_rate": 2.6238615664845174e-05,
603
+ "loss": 0.7316344738006592,
604
+ "step": 780
605
+ },
606
+ {
607
+ "epoch": 2.158469945355191,
608
+ "grad_norm": 9.91689682006836,
609
+ "learning_rate": 2.6147540983606557e-05,
610
+ "loss": 0.7728833198547364,
611
+ "step": 790
612
+ },
613
+ {
614
+ "epoch": 2.185792349726776,
615
+ "grad_norm": 24.687915802001953,
616
+ "learning_rate": 2.6056466302367944e-05,
617
+ "loss": 0.6872700691223145,
618
+ "step": 800
619
+ },
620
+ {
621
+ "epoch": 2.2131147540983607,
622
+ "grad_norm": 15.530792236328125,
623
+ "learning_rate": 2.5965391621129327e-05,
624
+ "loss": 0.8627532005310059,
625
+ "step": 810
626
+ },
627
+ {
628
+ "epoch": 2.240437158469945,
629
+ "grad_norm": 0.7304887175559998,
630
+ "learning_rate": 2.587431693989071e-05,
631
+ "loss": 0.5328950405120849,
632
+ "step": 820
633
+ },
634
+ {
635
+ "epoch": 2.26775956284153,
636
+ "grad_norm": 52.83394241333008,
637
+ "learning_rate": 2.5783242258652094e-05,
638
+ "loss": 0.45772466659545896,
639
+ "step": 830
640
+ },
641
+ {
642
+ "epoch": 2.2950819672131146,
643
+ "grad_norm": 23.795230865478516,
644
+ "learning_rate": 2.569216757741348e-05,
645
+ "loss": 0.6016922950744629,
646
+ "step": 840
647
+ },
648
+ {
649
+ "epoch": 2.3224043715846996,
650
+ "grad_norm": 22.11771011352539,
651
+ "learning_rate": 2.5601092896174864e-05,
652
+ "loss": 0.5151076316833496,
653
+ "step": 850
654
+ },
655
+ {
656
+ "epoch": 2.349726775956284,
657
+ "grad_norm": 0.44267675280570984,
658
+ "learning_rate": 2.5510018214936247e-05,
659
+ "loss": 0.9358211517333984,
660
+ "step": 860
661
+ },
662
+ {
663
+ "epoch": 2.3770491803278686,
664
+ "grad_norm": 0.3188877999782562,
665
+ "learning_rate": 2.541894353369763e-05,
666
+ "loss": 0.46982393264770506,
667
+ "step": 870
668
+ },
669
+ {
670
+ "epoch": 2.4043715846994536,
671
+ "grad_norm": 0.409311980009079,
672
+ "learning_rate": 2.5327868852459018e-05,
673
+ "loss": 0.5571242332458496,
674
+ "step": 880
675
+ },
676
+ {
677
+ "epoch": 2.431693989071038,
678
+ "grad_norm": 0.22160018980503082,
679
+ "learning_rate": 2.52367941712204e-05,
680
+ "loss": 0.686149263381958,
681
+ "step": 890
682
+ },
683
+ {
684
+ "epoch": 2.459016393442623,
685
+ "grad_norm": 56.0511360168457,
686
+ "learning_rate": 2.5145719489981784e-05,
687
+ "loss": 0.4332759857177734,
688
+ "step": 900
689
+ },
690
+ {
691
+ "epoch": 2.4863387978142075,
692
+ "grad_norm": 14.211305618286133,
693
+ "learning_rate": 2.5054644808743168e-05,
694
+ "loss": 0.46988306045532224,
695
+ "step": 910
696
+ },
697
+ {
698
+ "epoch": 2.5136612021857925,
699
+ "grad_norm": 0.394004762172699,
700
+ "learning_rate": 2.4963570127504555e-05,
701
+ "loss": 0.3961897611618042,
702
+ "step": 920
703
+ },
704
+ {
705
+ "epoch": 2.540983606557377,
706
+ "grad_norm": 56.96977233886719,
707
+ "learning_rate": 2.4872495446265938e-05,
708
+ "loss": 1.283346939086914,
709
+ "step": 930
710
+ },
711
+ {
712
+ "epoch": 2.5683060109289615,
713
+ "grad_norm": 24.957056045532227,
714
+ "learning_rate": 2.478142076502732e-05,
715
+ "loss": 0.8271417617797852,
716
+ "step": 940
717
+ },
718
+ {
719
+ "epoch": 2.5956284153005464,
720
+ "grad_norm": 1.057257056236267,
721
+ "learning_rate": 2.4690346083788705e-05,
722
+ "loss": 1.5506049156188966,
723
+ "step": 950
724
+ },
725
+ {
726
+ "epoch": 2.6229508196721314,
727
+ "grad_norm": 2.8221116065979004,
728
+ "learning_rate": 2.459927140255009e-05,
729
+ "loss": 0.5170619487762451,
730
+ "step": 960
731
+ },
732
+ {
733
+ "epoch": 2.650273224043716,
734
+ "grad_norm": 9.166945457458496,
735
+ "learning_rate": 2.4508196721311478e-05,
736
+ "loss": 1.0213055610656738,
737
+ "step": 970
738
+ },
739
+ {
740
+ "epoch": 2.6775956284153004,
741
+ "grad_norm": 13.581059455871582,
742
+ "learning_rate": 2.441712204007286e-05,
743
+ "loss": 0.604024600982666,
744
+ "step": 980
745
+ },
746
+ {
747
+ "epoch": 2.7049180327868854,
748
+ "grad_norm": 0.4070906639099121,
749
+ "learning_rate": 2.4326047358834245e-05,
750
+ "loss": 0.7995193481445313,
751
+ "step": 990
752
+ },
753
+ {
754
+ "epoch": 2.73224043715847,
755
+ "grad_norm": 0.3183801472187042,
756
+ "learning_rate": 2.4234972677595632e-05,
757
+ "loss": 0.7718863487243652,
758
+ "step": 1000
759
+ },
760
+ {
761
+ "epoch": 2.7595628415300544,
762
+ "grad_norm": 15.769966125488281,
763
+ "learning_rate": 2.4143897996357015e-05,
764
+ "loss": 0.7727096080780029,
765
+ "step": 1010
766
+ },
767
+ {
768
+ "epoch": 2.7868852459016393,
769
+ "grad_norm": 13.82215690612793,
770
+ "learning_rate": 2.40528233151184e-05,
771
+ "loss": 0.6281228542327881,
772
+ "step": 1020
773
+ },
774
+ {
775
+ "epoch": 2.8142076502732243,
776
+ "grad_norm": 0.24273018538951874,
777
+ "learning_rate": 2.3961748633879782e-05,
778
+ "loss": 0.7842752933502197,
779
+ "step": 1030
780
+ },
781
+ {
782
+ "epoch": 2.841530054644809,
783
+ "grad_norm": 0.21391047537326813,
784
+ "learning_rate": 2.387067395264117e-05,
785
+ "loss": 0.8222448348999023,
786
+ "step": 1040
787
+ },
788
+ {
789
+ "epoch": 2.8688524590163933,
790
+ "grad_norm": 0.5637283325195312,
791
+ "learning_rate": 2.3779599271402552e-05,
792
+ "loss": 1.0477510452270509,
793
+ "step": 1050
794
+ },
795
+ {
796
+ "epoch": 2.8961748633879782,
797
+ "grad_norm": 0.2845185101032257,
798
+ "learning_rate": 2.3688524590163936e-05,
799
+ "loss": 0.6434628009796143,
800
+ "step": 1060
801
+ },
802
+ {
803
+ "epoch": 2.9234972677595628,
804
+ "grad_norm": 1.1965957880020142,
805
+ "learning_rate": 2.359744990892532e-05,
806
+ "loss": 0.47115397453308105,
807
+ "step": 1070
808
+ },
809
+ {
810
+ "epoch": 2.9508196721311473,
811
+ "grad_norm": 0.9190018773078918,
812
+ "learning_rate": 2.3506375227686706e-05,
813
+ "loss": 0.7713288784027099,
814
+ "step": 1080
815
+ },
816
+ {
817
+ "epoch": 2.978142076502732,
818
+ "grad_norm": 1.0127384662628174,
819
+ "learning_rate": 2.341530054644809e-05,
820
+ "loss": 0.6314740657806397,
821
+ "step": 1090
822
+ },
823
+ {
824
+ "epoch": 3.0,
825
+ "eval_accuracy": 0.8547,
826
+ "eval_confusion_matrix": [
827
+ [
828
+ 277,
829
+ 5
830
+ ],
831
+ [
832
+ 47,
833
+ 29
834
+ ]
835
+ ],
836
+ "eval_f1": 0.5273,
837
+ "eval_loss": 0.498869389295578,
838
+ "eval_pr_auc": 0.6438,
839
+ "eval_precision": 0.8529,
840
+ "eval_recall": 0.3816,
841
+ "eval_roc_auc": 0.7867,
842
+ "eval_runtime": 137.6925,
843
+ "eval_samples_per_second": 2.6,
844
+ "eval_steps_per_second": 0.654,
845
+ "step": 1098
846
+ }
847
+ ],
848
+ "logging_steps": 10,
849
+ "max_steps": 3660,
850
+ "num_input_tokens_seen": 0,
851
+ "num_train_epochs": 10,
852
+ "save_steps": 500,
853
+ "stateful_callbacks": {
854
+ "EarlyStoppingCallback": {
855
+ "args": {
856
+ "early_stopping_patience": 3,
857
+ "early_stopping_threshold": 0.0
858
+ },
859
+ "attributes": {
860
+ "early_stopping_patience_counter": 2
861
+ }
862
+ },
863
+ "TrainerControl": {
864
+ "args": {
865
+ "should_epoch_stop": false,
866
+ "should_evaluate": false,
867
+ "should_log": false,
868
+ "should_save": true,
869
+ "should_training_stop": false
870
+ },
871
+ "attributes": {}
872
+ }
873
+ },
874
+ "total_flos": 1155583755141120.0,
875
+ "train_batch_size": 4,
876
+ "trial_name": null,
877
+ "trial_params": null
878
+ }
checkpoint-1098/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:637685e84eea04f2c0ca07978bad2436247d122cfd9ea34a4a652aece0212809
3
+ size 5265
checkpoint-1464/config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "RobertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "dtype": "float32",
10
+ "eos_token_id": 2,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "safe",
16
+ "1": "vulnerable"
17
+ },
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 3072,
20
+ "is_decoder": false,
21
+ "label2id": {
22
+ "safe": 0,
23
+ "vulnerable": 1
24
+ },
25
+ "layer_norm_eps": 1e-05,
26
+ "max_position_embeddings": 514,
27
+ "model_type": "roberta",
28
+ "num_attention_heads": 12,
29
+ "num_hidden_layers": 12,
30
+ "output_past": true,
31
+ "pad_token_id": 1,
32
+ "problem_type": "single_label_classification",
33
+ "tie_word_embeddings": true,
34
+ "transformers_version": "5.13.1",
35
+ "type_vocab_size": 1,
36
+ "use_cache": false,
37
+ "vocab_size": 50265
38
+ }
checkpoint-1464/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f62f271401ab127d950d2a0b1f3d04ce2e2ec14bbe14003f0b407920c3350a5e
3
+ size 498612824
checkpoint-1464/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67dedb18e38592408092c629a3e9a50945aba3bb6e6214be0629f6f9df356c1f
3
+ size 997340299
checkpoint-1464/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c585cc600d90695117b7d63884b3f6c865e2c9be941de1c6441d2613fac41f5d
3
+ size 14455
checkpoint-1464/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d77d3ae716dedb8dcc0aa236aa37084c265f16a6901863a9217535901ef4a01
3
+ size 1465
checkpoint-1464/trainer_state.json ADDED
@@ -0,0 +1,1161 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 366,
3
+ "best_metric": 0.41911664605140686,
4
+ "best_model_checkpoint": "training_output/run_20260713_203733/model/checkpoint-366",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1464,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0273224043715847,
14
+ "grad_norm": 4.927176475524902,
15
+ "learning_rate": 7.377049180327868e-07,
16
+ "loss": 0.6899909496307373,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.0546448087431694,
21
+ "grad_norm": 6.0230512619018555,
22
+ "learning_rate": 1.557377049180328e-06,
23
+ "loss": 0.7346426486968994,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.08196721311475409,
28
+ "grad_norm": 6.2535271644592285,
29
+ "learning_rate": 2.377049180327869e-06,
30
+ "loss": 0.6693419456481934,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.1092896174863388,
35
+ "grad_norm": 10.038297653198242,
36
+ "learning_rate": 3.1967213114754097e-06,
37
+ "loss": 0.6888828277587891,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.1366120218579235,
42
+ "grad_norm": 10.041961669921875,
43
+ "learning_rate": 4.016393442622951e-06,
44
+ "loss": 0.623580026626587,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.16393442622950818,
49
+ "grad_norm": 9.107032775878906,
50
+ "learning_rate": 4.836065573770492e-06,
51
+ "loss": 0.692430830001831,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.1912568306010929,
56
+ "grad_norm": 9.090417861938477,
57
+ "learning_rate": 5.655737704918032e-06,
58
+ "loss": 0.6110644817352295,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.2185792349726776,
63
+ "grad_norm": 7.914913177490234,
64
+ "learning_rate": 6.4754098360655735e-06,
65
+ "loss": 0.6268288612365722,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.2459016393442623,
70
+ "grad_norm": 8.18041706085205,
71
+ "learning_rate": 7.2950819672131145e-06,
72
+ "loss": 0.6859075546264648,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.273224043715847,
77
+ "grad_norm": 5.7157392501831055,
78
+ "learning_rate": 8.114754098360657e-06,
79
+ "loss": 0.7717578887939454,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.3005464480874317,
84
+ "grad_norm": 11.109936714172363,
85
+ "learning_rate": 8.934426229508197e-06,
86
+ "loss": 0.6476659774780273,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.32786885245901637,
91
+ "grad_norm": 10.369423866271973,
92
+ "learning_rate": 9.754098360655738e-06,
93
+ "loss": 0.6707509994506836,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.3551912568306011,
98
+ "grad_norm": 10.730874061584473,
99
+ "learning_rate": 1.0573770491803279e-05,
100
+ "loss": 0.6787842750549317,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.3825136612021858,
105
+ "grad_norm": 8.120079040527344,
106
+ "learning_rate": 1.139344262295082e-05,
107
+ "loss": 0.6765446186065673,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.4098360655737705,
112
+ "grad_norm": 14.429889678955078,
113
+ "learning_rate": 1.221311475409836e-05,
114
+ "loss": 0.623370361328125,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.4371584699453552,
119
+ "grad_norm": 9.193517684936523,
120
+ "learning_rate": 1.3032786885245902e-05,
121
+ "loss": 0.6362186908721924,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.4644808743169399,
126
+ "grad_norm": 8.7772855758667,
127
+ "learning_rate": 1.3852459016393443e-05,
128
+ "loss": 0.6641538619995118,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.4918032786885246,
133
+ "grad_norm": 5.2151780128479,
134
+ "learning_rate": 1.4672131147540984e-05,
135
+ "loss": 0.6653701782226562,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.5191256830601093,
140
+ "grad_norm": 17.197834014892578,
141
+ "learning_rate": 1.5491803278688525e-05,
142
+ "loss": 0.6582591533660889,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.546448087431694,
147
+ "grad_norm": 7.26473331451416,
148
+ "learning_rate": 1.6311475409836068e-05,
149
+ "loss": 0.765175724029541,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.5737704918032787,
154
+ "grad_norm": 5.47290563583374,
155
+ "learning_rate": 1.7131147540983607e-05,
156
+ "loss": 0.676247501373291,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.6010928961748634,
161
+ "grad_norm": 4.55684232711792,
162
+ "learning_rate": 1.7950819672131146e-05,
163
+ "loss": 0.6748649597167968,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.6284153005464481,
168
+ "grad_norm": 9.52070140838623,
169
+ "learning_rate": 1.877049180327869e-05,
170
+ "loss": 0.7023877143859864,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.6557377049180327,
175
+ "grad_norm": 10.813623428344727,
176
+ "learning_rate": 1.959016393442623e-05,
177
+ "loss": 0.5075790405273437,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.6830601092896175,
182
+ "grad_norm": 15.77322769165039,
183
+ "learning_rate": 2.040983606557377e-05,
184
+ "loss": 0.5947454929351806,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.7103825136612022,
189
+ "grad_norm": 14.141736030578613,
190
+ "learning_rate": 2.122950819672131e-05,
191
+ "loss": 0.7629669666290283,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.7377049180327869,
196
+ "grad_norm": 0.5761239528656006,
197
+ "learning_rate": 2.2049180327868853e-05,
198
+ "loss": 0.717600679397583,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.7650273224043715,
203
+ "grad_norm": 13.026739120483398,
204
+ "learning_rate": 2.2868852459016393e-05,
205
+ "loss": 1.5919431686401366,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.7923497267759563,
210
+ "grad_norm": 12.15133285522461,
211
+ "learning_rate": 2.3688524590163936e-05,
212
+ "loss": 0.6558570861816406,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.819672131147541,
217
+ "grad_norm": 20.35292625427246,
218
+ "learning_rate": 2.4508196721311478e-05,
219
+ "loss": 0.6991987705230713,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.8469945355191257,
224
+ "grad_norm": 9.59134292602539,
225
+ "learning_rate": 2.5327868852459018e-05,
226
+ "loss": 1.1239334106445313,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.8743169398907104,
231
+ "grad_norm": 4.265232086181641,
232
+ "learning_rate": 2.6147540983606557e-05,
233
+ "loss": 0.4866987705230713,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.9016393442622951,
238
+ "grad_norm": 9.87460708618164,
239
+ "learning_rate": 2.69672131147541e-05,
240
+ "loss": 0.7435293674468995,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.9289617486338798,
245
+ "grad_norm": 12.707701683044434,
246
+ "learning_rate": 2.7786885245901642e-05,
247
+ "loss": 0.9141542434692382,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.9562841530054644,
252
+ "grad_norm": 2.9138967990875244,
253
+ "learning_rate": 2.860655737704918e-05,
254
+ "loss": 0.5319347858428956,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.9836065573770492,
259
+ "grad_norm": 6.917504787445068,
260
+ "learning_rate": 2.942622950819672e-05,
261
+ "loss": 0.5438683509826661,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 1.0,
266
+ "eval_accuracy": 0.8492,
267
+ "eval_confusion_matrix": [
268
+ [
269
+ 282,
270
+ 0
271
+ ],
272
+ [
273
+ 54,
274
+ 22
275
+ ]
276
+ ],
277
+ "eval_f1": 0.449,
278
+ "eval_loss": 0.41911664605140686,
279
+ "eval_pr_auc": 0.6187,
280
+ "eval_precision": 1.0,
281
+ "eval_recall": 0.2895,
282
+ "eval_roc_auc": 0.7485,
283
+ "eval_runtime": 138.7598,
284
+ "eval_samples_per_second": 2.58,
285
+ "eval_steps_per_second": 0.649,
286
+ "step": 366
287
+ },
288
+ {
289
+ "epoch": 1.010928961748634,
290
+ "grad_norm": 6.519931316375732,
291
+ "learning_rate": 2.9972677595628417e-05,
292
+ "loss": 0.796769380569458,
293
+ "step": 370
294
+ },
295
+ {
296
+ "epoch": 1.0382513661202186,
297
+ "grad_norm": 4.874457359313965,
298
+ "learning_rate": 2.98816029143898e-05,
299
+ "loss": 0.398636531829834,
300
+ "step": 380
301
+ },
302
+ {
303
+ "epoch": 1.0655737704918034,
304
+ "grad_norm": 15.488590240478516,
305
+ "learning_rate": 2.9790528233151184e-05,
306
+ "loss": 0.7061968803405761,
307
+ "step": 390
308
+ },
309
+ {
310
+ "epoch": 1.092896174863388,
311
+ "grad_norm": 19.74693489074707,
312
+ "learning_rate": 2.9699453551912568e-05,
313
+ "loss": 0.5211454391479492,
314
+ "step": 400
315
+ },
316
+ {
317
+ "epoch": 1.1202185792349726,
318
+ "grad_norm": 2.513188362121582,
319
+ "learning_rate": 2.9608378870673954e-05,
320
+ "loss": 0.7898036003112793,
321
+ "step": 410
322
+ },
323
+ {
324
+ "epoch": 1.1475409836065573,
325
+ "grad_norm": 25.31527328491211,
326
+ "learning_rate": 2.9517304189435338e-05,
327
+ "loss": 0.8195189476013184,
328
+ "step": 420
329
+ },
330
+ {
331
+ "epoch": 1.174863387978142,
332
+ "grad_norm": 26.276199340820312,
333
+ "learning_rate": 2.942622950819672e-05,
334
+ "loss": 0.8042145729064941,
335
+ "step": 430
336
+ },
337
+ {
338
+ "epoch": 1.2021857923497268,
339
+ "grad_norm": 19.838821411132812,
340
+ "learning_rate": 2.9335154826958105e-05,
341
+ "loss": 0.5532180786132812,
342
+ "step": 440
343
+ },
344
+ {
345
+ "epoch": 1.2295081967213115,
346
+ "grad_norm": 1.5704028606414795,
347
+ "learning_rate": 2.924408014571949e-05,
348
+ "loss": 0.46767144203186034,
349
+ "step": 450
350
+ },
351
+ {
352
+ "epoch": 1.2568306010928962,
353
+ "grad_norm": 0.24576333165168762,
354
+ "learning_rate": 2.9153005464480875e-05,
355
+ "loss": 0.7125775337219238,
356
+ "step": 460
357
+ },
358
+ {
359
+ "epoch": 1.2841530054644807,
360
+ "grad_norm": 2.8051435947418213,
361
+ "learning_rate": 2.9061930783242258e-05,
362
+ "loss": 1.4706084251403808,
363
+ "step": 470
364
+ },
365
+ {
366
+ "epoch": 1.3114754098360657,
367
+ "grad_norm": 2.1515614986419678,
368
+ "learning_rate": 2.897085610200364e-05,
369
+ "loss": 0.740934181213379,
370
+ "step": 480
371
+ },
372
+ {
373
+ "epoch": 1.3387978142076502,
374
+ "grad_norm": 5.335806846618652,
375
+ "learning_rate": 2.8879781420765028e-05,
376
+ "loss": 0.5567354679107666,
377
+ "step": 490
378
+ },
379
+ {
380
+ "epoch": 1.366120218579235,
381
+ "grad_norm": 9.499934196472168,
382
+ "learning_rate": 2.878870673952641e-05,
383
+ "loss": 0.5102160930633545,
384
+ "step": 500
385
+ },
386
+ {
387
+ "epoch": 1.3934426229508197,
388
+ "grad_norm": 0.30213791131973267,
389
+ "learning_rate": 2.8697632058287795e-05,
390
+ "loss": 0.8127809524536133,
391
+ "step": 510
392
+ },
393
+ {
394
+ "epoch": 1.4207650273224044,
395
+ "grad_norm": 17.827165603637695,
396
+ "learning_rate": 2.860655737704918e-05,
397
+ "loss": 1.3400749206542968,
398
+ "step": 520
399
+ },
400
+ {
401
+ "epoch": 1.4480874316939891,
402
+ "grad_norm": 2.58740234375,
403
+ "learning_rate": 2.8515482695810565e-05,
404
+ "loss": 0.5003597736358643,
405
+ "step": 530
406
+ },
407
+ {
408
+ "epoch": 1.4754098360655736,
409
+ "grad_norm": 22.007869720458984,
410
+ "learning_rate": 2.842440801457195e-05,
411
+ "loss": 1.6859323501586914,
412
+ "step": 540
413
+ },
414
+ {
415
+ "epoch": 1.5027322404371586,
416
+ "grad_norm": 14.658230781555176,
417
+ "learning_rate": 2.8333333333333332e-05,
418
+ "loss": 0.9195033073425293,
419
+ "step": 550
420
+ },
421
+ {
422
+ "epoch": 1.530054644808743,
423
+ "grad_norm": 1.4521483182907104,
424
+ "learning_rate": 2.8242258652094715e-05,
425
+ "loss": 0.4108741283416748,
426
+ "step": 560
427
+ },
428
+ {
429
+ "epoch": 1.5573770491803278,
430
+ "grad_norm": 10.377533912658691,
431
+ "learning_rate": 2.8151183970856102e-05,
432
+ "loss": 0.655418586730957,
433
+ "step": 570
434
+ },
435
+ {
436
+ "epoch": 1.5846994535519126,
437
+ "grad_norm": 0.5859184265136719,
438
+ "learning_rate": 2.8060109289617486e-05,
439
+ "loss": 1.2600435256958007,
440
+ "step": 580
441
+ },
442
+ {
443
+ "epoch": 1.6120218579234973,
444
+ "grad_norm": 0.8876230120658875,
445
+ "learning_rate": 2.796903460837887e-05,
446
+ "loss": 1.0982278823852538,
447
+ "step": 590
448
+ },
449
+ {
450
+ "epoch": 1.639344262295082,
451
+ "grad_norm": 13.285502433776855,
452
+ "learning_rate": 2.7877959927140256e-05,
453
+ "loss": 0.6685476779937745,
454
+ "step": 600
455
+ },
456
+ {
457
+ "epoch": 1.6666666666666665,
458
+ "grad_norm": 0.6555774211883545,
459
+ "learning_rate": 2.7786885245901642e-05,
460
+ "loss": 0.19812407493591308,
461
+ "step": 610
462
+ },
463
+ {
464
+ "epoch": 1.6939890710382515,
465
+ "grad_norm": 2.7470834255218506,
466
+ "learning_rate": 2.7695810564663026e-05,
467
+ "loss": 1.1334525108337403,
468
+ "step": 620
469
+ },
470
+ {
471
+ "epoch": 1.721311475409836,
472
+ "grad_norm": 14.42418098449707,
473
+ "learning_rate": 2.760473588342441e-05,
474
+ "loss": 0.9498401641845703,
475
+ "step": 630
476
+ },
477
+ {
478
+ "epoch": 1.748633879781421,
479
+ "grad_norm": 57.35378646850586,
480
+ "learning_rate": 2.7513661202185793e-05,
481
+ "loss": 1.2678983688354493,
482
+ "step": 640
483
+ },
484
+ {
485
+ "epoch": 1.7759562841530054,
486
+ "grad_norm": 0.7189310789108276,
487
+ "learning_rate": 2.742258652094718e-05,
488
+ "loss": 0.6106186389923096,
489
+ "step": 650
490
+ },
491
+ {
492
+ "epoch": 1.8032786885245902,
493
+ "grad_norm": 13.546385765075684,
494
+ "learning_rate": 2.7331511839708563e-05,
495
+ "loss": 1.4015119552612305,
496
+ "step": 660
497
+ },
498
+ {
499
+ "epoch": 1.830601092896175,
500
+ "grad_norm": 8.754522323608398,
501
+ "learning_rate": 2.7240437158469946e-05,
502
+ "loss": 0.9790468215942383,
503
+ "step": 670
504
+ },
505
+ {
506
+ "epoch": 1.8579234972677594,
507
+ "grad_norm": 13.214091300964355,
508
+ "learning_rate": 2.7149362477231333e-05,
509
+ "loss": 0.6752419948577881,
510
+ "step": 680
511
+ },
512
+ {
513
+ "epoch": 1.8852459016393444,
514
+ "grad_norm": 2.506131887435913,
515
+ "learning_rate": 2.7058287795992716e-05,
516
+ "loss": 0.2787914752960205,
517
+ "step": 690
518
+ },
519
+ {
520
+ "epoch": 1.9125683060109289,
521
+ "grad_norm": 0.4683574438095093,
522
+ "learning_rate": 2.69672131147541e-05,
523
+ "loss": 0.8134347915649414,
524
+ "step": 700
525
+ },
526
+ {
527
+ "epoch": 1.9398907103825138,
528
+ "grad_norm": 0.8958243131637573,
529
+ "learning_rate": 2.6876138433515483e-05,
530
+ "loss": 0.4489922046661377,
531
+ "step": 710
532
+ },
533
+ {
534
+ "epoch": 1.9672131147540983,
535
+ "grad_norm": 95.41824340820312,
536
+ "learning_rate": 2.678506375227687e-05,
537
+ "loss": 0.9731103897094726,
538
+ "step": 720
539
+ },
540
+ {
541
+ "epoch": 1.994535519125683,
542
+ "grad_norm": 0.7653670907020569,
543
+ "learning_rate": 2.6693989071038253e-05,
544
+ "loss": 1.0658687591552733,
545
+ "step": 730
546
+ },
547
+ {
548
+ "epoch": 2.0,
549
+ "eval_accuracy": 0.8659,
550
+ "eval_confusion_matrix": [
551
+ [
552
+ 281,
553
+ 1
554
+ ],
555
+ [
556
+ 47,
557
+ 29
558
+ ]
559
+ ],
560
+ "eval_f1": 0.5472,
561
+ "eval_loss": 0.45224741101264954,
562
+ "eval_pr_auc": 0.6674,
563
+ "eval_precision": 0.9667,
564
+ "eval_recall": 0.3816,
565
+ "eval_roc_auc": 0.8086,
566
+ "eval_runtime": 137.9939,
567
+ "eval_samples_per_second": 2.594,
568
+ "eval_steps_per_second": 0.652,
569
+ "step": 732
570
+ },
571
+ {
572
+ "epoch": 2.021857923497268,
573
+ "grad_norm": 32.09031677246094,
574
+ "learning_rate": 2.6602914389799637e-05,
575
+ "loss": 0.7306198120117188,
576
+ "step": 740
577
+ },
578
+ {
579
+ "epoch": 2.0491803278688523,
580
+ "grad_norm": 51.15188217163086,
581
+ "learning_rate": 2.651183970856102e-05,
582
+ "loss": 0.49246606826782224,
583
+ "step": 750
584
+ },
585
+ {
586
+ "epoch": 2.0765027322404372,
587
+ "grad_norm": 0.5293377041816711,
588
+ "learning_rate": 2.6420765027322407e-05,
589
+ "loss": 0.30229816436767576,
590
+ "step": 760
591
+ },
592
+ {
593
+ "epoch": 2.1038251366120218,
594
+ "grad_norm": 0.5600886344909668,
595
+ "learning_rate": 2.632969034608379e-05,
596
+ "loss": 1.428311824798584,
597
+ "step": 770
598
+ },
599
+ {
600
+ "epoch": 2.1311475409836067,
601
+ "grad_norm": 8.111888885498047,
602
+ "learning_rate": 2.6238615664845174e-05,
603
+ "loss": 0.7316344738006592,
604
+ "step": 780
605
+ },
606
+ {
607
+ "epoch": 2.158469945355191,
608
+ "grad_norm": 9.91689682006836,
609
+ "learning_rate": 2.6147540983606557e-05,
610
+ "loss": 0.7728833198547364,
611
+ "step": 790
612
+ },
613
+ {
614
+ "epoch": 2.185792349726776,
615
+ "grad_norm": 24.687915802001953,
616
+ "learning_rate": 2.6056466302367944e-05,
617
+ "loss": 0.6872700691223145,
618
+ "step": 800
619
+ },
620
+ {
621
+ "epoch": 2.2131147540983607,
622
+ "grad_norm": 15.530792236328125,
623
+ "learning_rate": 2.5965391621129327e-05,
624
+ "loss": 0.8627532005310059,
625
+ "step": 810
626
+ },
627
+ {
628
+ "epoch": 2.240437158469945,
629
+ "grad_norm": 0.7304887175559998,
630
+ "learning_rate": 2.587431693989071e-05,
631
+ "loss": 0.5328950405120849,
632
+ "step": 820
633
+ },
634
+ {
635
+ "epoch": 2.26775956284153,
636
+ "grad_norm": 52.83394241333008,
637
+ "learning_rate": 2.5783242258652094e-05,
638
+ "loss": 0.45772466659545896,
639
+ "step": 830
640
+ },
641
+ {
642
+ "epoch": 2.2950819672131146,
643
+ "grad_norm": 23.795230865478516,
644
+ "learning_rate": 2.569216757741348e-05,
645
+ "loss": 0.6016922950744629,
646
+ "step": 840
647
+ },
648
+ {
649
+ "epoch": 2.3224043715846996,
650
+ "grad_norm": 22.11771011352539,
651
+ "learning_rate": 2.5601092896174864e-05,
652
+ "loss": 0.5151076316833496,
653
+ "step": 850
654
+ },
655
+ {
656
+ "epoch": 2.349726775956284,
657
+ "grad_norm": 0.44267675280570984,
658
+ "learning_rate": 2.5510018214936247e-05,
659
+ "loss": 0.9358211517333984,
660
+ "step": 860
661
+ },
662
+ {
663
+ "epoch": 2.3770491803278686,
664
+ "grad_norm": 0.3188877999782562,
665
+ "learning_rate": 2.541894353369763e-05,
666
+ "loss": 0.46982393264770506,
667
+ "step": 870
668
+ },
669
+ {
670
+ "epoch": 2.4043715846994536,
671
+ "grad_norm": 0.409311980009079,
672
+ "learning_rate": 2.5327868852459018e-05,
673
+ "loss": 0.5571242332458496,
674
+ "step": 880
675
+ },
676
+ {
677
+ "epoch": 2.431693989071038,
678
+ "grad_norm": 0.22160018980503082,
679
+ "learning_rate": 2.52367941712204e-05,
680
+ "loss": 0.686149263381958,
681
+ "step": 890
682
+ },
683
+ {
684
+ "epoch": 2.459016393442623,
685
+ "grad_norm": 56.0511360168457,
686
+ "learning_rate": 2.5145719489981784e-05,
687
+ "loss": 0.4332759857177734,
688
+ "step": 900
689
+ },
690
+ {
691
+ "epoch": 2.4863387978142075,
692
+ "grad_norm": 14.211305618286133,
693
+ "learning_rate": 2.5054644808743168e-05,
694
+ "loss": 0.46988306045532224,
695
+ "step": 910
696
+ },
697
+ {
698
+ "epoch": 2.5136612021857925,
699
+ "grad_norm": 0.394004762172699,
700
+ "learning_rate": 2.4963570127504555e-05,
701
+ "loss": 0.3961897611618042,
702
+ "step": 920
703
+ },
704
+ {
705
+ "epoch": 2.540983606557377,
706
+ "grad_norm": 56.96977233886719,
707
+ "learning_rate": 2.4872495446265938e-05,
708
+ "loss": 1.283346939086914,
709
+ "step": 930
710
+ },
711
+ {
712
+ "epoch": 2.5683060109289615,
713
+ "grad_norm": 24.957056045532227,
714
+ "learning_rate": 2.478142076502732e-05,
715
+ "loss": 0.8271417617797852,
716
+ "step": 940
717
+ },
718
+ {
719
+ "epoch": 2.5956284153005464,
720
+ "grad_norm": 1.057257056236267,
721
+ "learning_rate": 2.4690346083788705e-05,
722
+ "loss": 1.5506049156188966,
723
+ "step": 950
724
+ },
725
+ {
726
+ "epoch": 2.6229508196721314,
727
+ "grad_norm": 2.8221116065979004,
728
+ "learning_rate": 2.459927140255009e-05,
729
+ "loss": 0.5170619487762451,
730
+ "step": 960
731
+ },
732
+ {
733
+ "epoch": 2.650273224043716,
734
+ "grad_norm": 9.166945457458496,
735
+ "learning_rate": 2.4508196721311478e-05,
736
+ "loss": 1.0213055610656738,
737
+ "step": 970
738
+ },
739
+ {
740
+ "epoch": 2.6775956284153004,
741
+ "grad_norm": 13.581059455871582,
742
+ "learning_rate": 2.441712204007286e-05,
743
+ "loss": 0.604024600982666,
744
+ "step": 980
745
+ },
746
+ {
747
+ "epoch": 2.7049180327868854,
748
+ "grad_norm": 0.4070906639099121,
749
+ "learning_rate": 2.4326047358834245e-05,
750
+ "loss": 0.7995193481445313,
751
+ "step": 990
752
+ },
753
+ {
754
+ "epoch": 2.73224043715847,
755
+ "grad_norm": 0.3183801472187042,
756
+ "learning_rate": 2.4234972677595632e-05,
757
+ "loss": 0.7718863487243652,
758
+ "step": 1000
759
+ },
760
+ {
761
+ "epoch": 2.7595628415300544,
762
+ "grad_norm": 15.769966125488281,
763
+ "learning_rate": 2.4143897996357015e-05,
764
+ "loss": 0.7727096080780029,
765
+ "step": 1010
766
+ },
767
+ {
768
+ "epoch": 2.7868852459016393,
769
+ "grad_norm": 13.82215690612793,
770
+ "learning_rate": 2.40528233151184e-05,
771
+ "loss": 0.6281228542327881,
772
+ "step": 1020
773
+ },
774
+ {
775
+ "epoch": 2.8142076502732243,
776
+ "grad_norm": 0.24273018538951874,
777
+ "learning_rate": 2.3961748633879782e-05,
778
+ "loss": 0.7842752933502197,
779
+ "step": 1030
780
+ },
781
+ {
782
+ "epoch": 2.841530054644809,
783
+ "grad_norm": 0.21391047537326813,
784
+ "learning_rate": 2.387067395264117e-05,
785
+ "loss": 0.8222448348999023,
786
+ "step": 1040
787
+ },
788
+ {
789
+ "epoch": 2.8688524590163933,
790
+ "grad_norm": 0.5637283325195312,
791
+ "learning_rate": 2.3779599271402552e-05,
792
+ "loss": 1.0477510452270509,
793
+ "step": 1050
794
+ },
795
+ {
796
+ "epoch": 2.8961748633879782,
797
+ "grad_norm": 0.2845185101032257,
798
+ "learning_rate": 2.3688524590163936e-05,
799
+ "loss": 0.6434628009796143,
800
+ "step": 1060
801
+ },
802
+ {
803
+ "epoch": 2.9234972677595628,
804
+ "grad_norm": 1.1965957880020142,
805
+ "learning_rate": 2.359744990892532e-05,
806
+ "loss": 0.47115397453308105,
807
+ "step": 1070
808
+ },
809
+ {
810
+ "epoch": 2.9508196721311473,
811
+ "grad_norm": 0.9190018773078918,
812
+ "learning_rate": 2.3506375227686706e-05,
813
+ "loss": 0.7713288784027099,
814
+ "step": 1080
815
+ },
816
+ {
817
+ "epoch": 2.978142076502732,
818
+ "grad_norm": 1.0127384662628174,
819
+ "learning_rate": 2.341530054644809e-05,
820
+ "loss": 0.6314740657806397,
821
+ "step": 1090
822
+ },
823
+ {
824
+ "epoch": 3.0,
825
+ "eval_accuracy": 0.8547,
826
+ "eval_confusion_matrix": [
827
+ [
828
+ 277,
829
+ 5
830
+ ],
831
+ [
832
+ 47,
833
+ 29
834
+ ]
835
+ ],
836
+ "eval_f1": 0.5273,
837
+ "eval_loss": 0.498869389295578,
838
+ "eval_pr_auc": 0.6438,
839
+ "eval_precision": 0.8529,
840
+ "eval_recall": 0.3816,
841
+ "eval_roc_auc": 0.7867,
842
+ "eval_runtime": 137.6925,
843
+ "eval_samples_per_second": 2.6,
844
+ "eval_steps_per_second": 0.654,
845
+ "step": 1098
846
+ },
847
+ {
848
+ "epoch": 3.0054644808743167,
849
+ "grad_norm": 59.913047790527344,
850
+ "learning_rate": 2.3324225865209472e-05,
851
+ "loss": 1.0216832160949707,
852
+ "step": 1100
853
+ },
854
+ {
855
+ "epoch": 3.0327868852459017,
856
+ "grad_norm": 13.406026840209961,
857
+ "learning_rate": 2.3233151183970856e-05,
858
+ "loss": 0.3810760974884033,
859
+ "step": 1110
860
+ },
861
+ {
862
+ "epoch": 3.060109289617486,
863
+ "grad_norm": 15.36068058013916,
864
+ "learning_rate": 2.3142076502732243e-05,
865
+ "loss": 0.607788372039795,
866
+ "step": 1120
867
+ },
868
+ {
869
+ "epoch": 3.087431693989071,
870
+ "grad_norm": 0.6151846051216125,
871
+ "learning_rate": 2.3051001821493626e-05,
872
+ "loss": 0.4322453498840332,
873
+ "step": 1130
874
+ },
875
+ {
876
+ "epoch": 3.1147540983606556,
877
+ "grad_norm": 81.44578552246094,
878
+ "learning_rate": 2.295992714025501e-05,
879
+ "loss": 1.3470041275024414,
880
+ "step": 1140
881
+ },
882
+ {
883
+ "epoch": 3.1420765027322406,
884
+ "grad_norm": 3.3564834594726562,
885
+ "learning_rate": 2.2868852459016393e-05,
886
+ "loss": 0.7912285327911377,
887
+ "step": 1150
888
+ },
889
+ {
890
+ "epoch": 3.169398907103825,
891
+ "grad_norm": 15.688278198242188,
892
+ "learning_rate": 2.277777777777778e-05,
893
+ "loss": 1.0597174644470215,
894
+ "step": 1160
895
+ },
896
+ {
897
+ "epoch": 3.19672131147541,
898
+ "grad_norm": 17.383037567138672,
899
+ "learning_rate": 2.2686703096539163e-05,
900
+ "loss": 0.5524578094482422,
901
+ "step": 1170
902
+ },
903
+ {
904
+ "epoch": 3.2240437158469946,
905
+ "grad_norm": 75.40896606445312,
906
+ "learning_rate": 2.2595628415300546e-05,
907
+ "loss": 0.6506473541259765,
908
+ "step": 1180
909
+ },
910
+ {
911
+ "epoch": 3.251366120218579,
912
+ "grad_norm": 31.155624389648438,
913
+ "learning_rate": 2.250455373406193e-05,
914
+ "loss": 0.8582865715026855,
915
+ "step": 1190
916
+ },
917
+ {
918
+ "epoch": 3.278688524590164,
919
+ "grad_norm": 0.39654970169067383,
920
+ "learning_rate": 2.2413479052823316e-05,
921
+ "loss": 0.24634058475494386,
922
+ "step": 1200
923
+ },
924
+ {
925
+ "epoch": 3.3060109289617485,
926
+ "grad_norm": 0.2892335057258606,
927
+ "learning_rate": 2.23224043715847e-05,
928
+ "loss": 0.7392459869384765,
929
+ "step": 1210
930
+ },
931
+ {
932
+ "epoch": 3.3333333333333335,
933
+ "grad_norm": 0.13577429950237274,
934
+ "learning_rate": 2.2231329690346083e-05,
935
+ "loss": 0.8051422119140625,
936
+ "step": 1220
937
+ },
938
+ {
939
+ "epoch": 3.360655737704918,
940
+ "grad_norm": 36.66856384277344,
941
+ "learning_rate": 2.2140255009107467e-05,
942
+ "loss": 0.20498306751251222,
943
+ "step": 1230
944
+ },
945
+ {
946
+ "epoch": 3.387978142076503,
947
+ "grad_norm": 0.3154204785823822,
948
+ "learning_rate": 2.2049180327868853e-05,
949
+ "loss": 0.3335949182510376,
950
+ "step": 1240
951
+ },
952
+ {
953
+ "epoch": 3.4153005464480874,
954
+ "grad_norm": 17.92582130432129,
955
+ "learning_rate": 2.1958105646630237e-05,
956
+ "loss": 0.8984568595886231,
957
+ "step": 1250
958
+ },
959
+ {
960
+ "epoch": 3.442622950819672,
961
+ "grad_norm": 17.423526763916016,
962
+ "learning_rate": 2.186703096539162e-05,
963
+ "loss": 0.7581507205963135,
964
+ "step": 1260
965
+ },
966
+ {
967
+ "epoch": 3.469945355191257,
968
+ "grad_norm": 16.00027847290039,
969
+ "learning_rate": 2.1775956284153004e-05,
970
+ "loss": 0.710300064086914,
971
+ "step": 1270
972
+ },
973
+ {
974
+ "epoch": 3.4972677595628414,
975
+ "grad_norm": 64.70391845703125,
976
+ "learning_rate": 2.168488160291439e-05,
977
+ "loss": 0.7110927104949951,
978
+ "step": 1280
979
+ },
980
+ {
981
+ "epoch": 3.5245901639344264,
982
+ "grad_norm": 21.78537368774414,
983
+ "learning_rate": 2.1593806921675774e-05,
984
+ "loss": 0.4351015090942383,
985
+ "step": 1290
986
+ },
987
+ {
988
+ "epoch": 3.551912568306011,
989
+ "grad_norm": 34.40507888793945,
990
+ "learning_rate": 2.1502732240437157e-05,
991
+ "loss": 0.4454930782318115,
992
+ "step": 1300
993
+ },
994
+ {
995
+ "epoch": 3.579234972677596,
996
+ "grad_norm": 19.285673141479492,
997
+ "learning_rate": 2.141165755919854e-05,
998
+ "loss": 1.04069185256958,
999
+ "step": 1310
1000
+ },
1001
+ {
1002
+ "epoch": 3.6065573770491803,
1003
+ "grad_norm": 42.792259216308594,
1004
+ "learning_rate": 2.1320582877959927e-05,
1005
+ "loss": 0.6689809799194336,
1006
+ "step": 1320
1007
+ },
1008
+ {
1009
+ "epoch": 3.633879781420765,
1010
+ "grad_norm": 0.1760418862104416,
1011
+ "learning_rate": 2.122950819672131e-05,
1012
+ "loss": 0.36717102527618406,
1013
+ "step": 1330
1014
+ },
1015
+ {
1016
+ "epoch": 3.66120218579235,
1017
+ "grad_norm": 0.2453140765428543,
1018
+ "learning_rate": 2.1138433515482697e-05,
1019
+ "loss": 0.2981537103652954,
1020
+ "step": 1340
1021
+ },
1022
+ {
1023
+ "epoch": 3.6885245901639343,
1024
+ "grad_norm": 0.12099751085042953,
1025
+ "learning_rate": 2.104735883424408e-05,
1026
+ "loss": 0.7060315608978271,
1027
+ "step": 1350
1028
+ },
1029
+ {
1030
+ "epoch": 3.7158469945355193,
1031
+ "grad_norm": 0.1845744252204895,
1032
+ "learning_rate": 2.0956284153005468e-05,
1033
+ "loss": 0.6227611541748047,
1034
+ "step": 1360
1035
+ },
1036
+ {
1037
+ "epoch": 3.7431693989071038,
1038
+ "grad_norm": 0.11914532631635666,
1039
+ "learning_rate": 2.086520947176685e-05,
1040
+ "loss": 0.051645058393478396,
1041
+ "step": 1370
1042
+ },
1043
+ {
1044
+ "epoch": 3.7704918032786887,
1045
+ "grad_norm": 0.1372411698102951,
1046
+ "learning_rate": 2.0774134790528234e-05,
1047
+ "loss": 0.756002140045166,
1048
+ "step": 1380
1049
+ },
1050
+ {
1051
+ "epoch": 3.797814207650273,
1052
+ "grad_norm": 30.7832088470459,
1053
+ "learning_rate": 2.0683060109289618e-05,
1054
+ "loss": 0.16540950536727905,
1055
+ "step": 1390
1056
+ },
1057
+ {
1058
+ "epoch": 3.8251366120218577,
1059
+ "grad_norm": 0.09204358607530594,
1060
+ "learning_rate": 2.0591985428051005e-05,
1061
+ "loss": 0.39240407943725586,
1062
+ "step": 1400
1063
+ },
1064
+ {
1065
+ "epoch": 3.8524590163934427,
1066
+ "grad_norm": 0.13517121970653534,
1067
+ "learning_rate": 2.0500910746812388e-05,
1068
+ "loss": 0.7909457683563232,
1069
+ "step": 1410
1070
+ },
1071
+ {
1072
+ "epoch": 3.879781420765027,
1073
+ "grad_norm": 0.40855443477630615,
1074
+ "learning_rate": 2.040983606557377e-05,
1075
+ "loss": 0.8180214881896972,
1076
+ "step": 1420
1077
+ },
1078
+ {
1079
+ "epoch": 3.907103825136612,
1080
+ "grad_norm": 0.7466303110122681,
1081
+ "learning_rate": 2.0318761384335155e-05,
1082
+ "loss": 0.42211570739746096,
1083
+ "step": 1430
1084
+ },
1085
+ {
1086
+ "epoch": 3.9344262295081966,
1087
+ "grad_norm": 0.7725684642791748,
1088
+ "learning_rate": 2.022768670309654e-05,
1089
+ "loss": 0.43922696113586424,
1090
+ "step": 1440
1091
+ },
1092
+ {
1093
+ "epoch": 3.9617486338797816,
1094
+ "grad_norm": 0.3130258619785309,
1095
+ "learning_rate": 2.0136612021857925e-05,
1096
+ "loss": 0.5256992816925049,
1097
+ "step": 1450
1098
+ },
1099
+ {
1100
+ "epoch": 3.989071038251366,
1101
+ "grad_norm": 0.023413795977830887,
1102
+ "learning_rate": 2.0045537340619308e-05,
1103
+ "loss": 0.5202861785888672,
1104
+ "step": 1460
1105
+ },
1106
+ {
1107
+ "epoch": 4.0,
1108
+ "eval_accuracy": 0.8575,
1109
+ "eval_confusion_matrix": [
1110
+ [
1111
+ 278,
1112
+ 4
1113
+ ],
1114
+ [
1115
+ 47,
1116
+ 29
1117
+ ]
1118
+ ],
1119
+ "eval_f1": 0.5321,
1120
+ "eval_loss": 0.7507914304733276,
1121
+ "eval_pr_auc": 0.6323,
1122
+ "eval_precision": 0.8788,
1123
+ "eval_recall": 0.3816,
1124
+ "eval_roc_auc": 0.7597,
1125
+ "eval_runtime": 140.186,
1126
+ "eval_samples_per_second": 2.554,
1127
+ "eval_steps_per_second": 0.642,
1128
+ "step": 1464
1129
+ }
1130
+ ],
1131
+ "logging_steps": 10,
1132
+ "max_steps": 3660,
1133
+ "num_input_tokens_seen": 0,
1134
+ "num_train_epochs": 10,
1135
+ "save_steps": 500,
1136
+ "stateful_callbacks": {
1137
+ "EarlyStoppingCallback": {
1138
+ "args": {
1139
+ "early_stopping_patience": 3,
1140
+ "early_stopping_threshold": 0.0
1141
+ },
1142
+ "attributes": {
1143
+ "early_stopping_patience_counter": 3
1144
+ }
1145
+ },
1146
+ "TrainerControl": {
1147
+ "args": {
1148
+ "should_epoch_stop": false,
1149
+ "should_evaluate": false,
1150
+ "should_log": false,
1151
+ "should_save": true,
1152
+ "should_training_stop": true
1153
+ },
1154
+ "attributes": {}
1155
+ }
1156
+ },
1157
+ "total_flos": 1540778340188160.0,
1158
+ "train_batch_size": 4,
1159
+ "trial_name": null,
1160
+ "trial_params": null
1161
+ }
checkpoint-1464/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:637685e84eea04f2c0ca07978bad2436247d122cfd9ea34a4a652aece0212809
3
+ size 5265
checkpoint-366/config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "RobertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "dtype": "float32",
10
+ "eos_token_id": 2,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "safe",
16
+ "1": "vulnerable"
17
+ },
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 3072,
20
+ "is_decoder": false,
21
+ "label2id": {
22
+ "safe": 0,
23
+ "vulnerable": 1
24
+ },
25
+ "layer_norm_eps": 1e-05,
26
+ "max_position_embeddings": 514,
27
+ "model_type": "roberta",
28
+ "num_attention_heads": 12,
29
+ "num_hidden_layers": 12,
30
+ "output_past": true,
31
+ "pad_token_id": 1,
32
+ "problem_type": "single_label_classification",
33
+ "tie_word_embeddings": true,
34
+ "transformers_version": "5.13.1",
35
+ "type_vocab_size": 1,
36
+ "use_cache": false,
37
+ "vocab_size": 50265
38
+ }
checkpoint-366/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59da1e0b742c0b9f9ad0e6078e2195023ce00a1cffd33a1f5766d20a13aca4a5
3
+ size 498612824
checkpoint-366/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0aa384c26896412aba7a458be89969ee11318e6bed78b1258d12746b0d37d43
3
+ size 997340299
checkpoint-366/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b79e49b0815d6739c78f1d25bedea67ba4c28d55869d7e6b5acd38bf06a9a9be
3
+ size 14455
checkpoint-366/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35115f2606ef316bf8b6dde13017d01e34afac594a5d9164e4cf8021b4292e3b
3
+ size 1465
checkpoint-366/trainer_state.json ADDED
@@ -0,0 +1,319 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 366,
3
+ "best_metric": 0.41911664605140686,
4
+ "best_model_checkpoint": "training_output/run_20260713_203733/model/checkpoint-366",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 366,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0273224043715847,
14
+ "grad_norm": 4.927176475524902,
15
+ "learning_rate": 7.377049180327868e-07,
16
+ "loss": 0.6899909496307373,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.0546448087431694,
21
+ "grad_norm": 6.0230512619018555,
22
+ "learning_rate": 1.557377049180328e-06,
23
+ "loss": 0.7346426486968994,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.08196721311475409,
28
+ "grad_norm": 6.2535271644592285,
29
+ "learning_rate": 2.377049180327869e-06,
30
+ "loss": 0.6693419456481934,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.1092896174863388,
35
+ "grad_norm": 10.038297653198242,
36
+ "learning_rate": 3.1967213114754097e-06,
37
+ "loss": 0.6888828277587891,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.1366120218579235,
42
+ "grad_norm": 10.041961669921875,
43
+ "learning_rate": 4.016393442622951e-06,
44
+ "loss": 0.623580026626587,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.16393442622950818,
49
+ "grad_norm": 9.107032775878906,
50
+ "learning_rate": 4.836065573770492e-06,
51
+ "loss": 0.692430830001831,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.1912568306010929,
56
+ "grad_norm": 9.090417861938477,
57
+ "learning_rate": 5.655737704918032e-06,
58
+ "loss": 0.6110644817352295,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.2185792349726776,
63
+ "grad_norm": 7.914913177490234,
64
+ "learning_rate": 6.4754098360655735e-06,
65
+ "loss": 0.6268288612365722,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.2459016393442623,
70
+ "grad_norm": 8.18041706085205,
71
+ "learning_rate": 7.2950819672131145e-06,
72
+ "loss": 0.6859075546264648,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.273224043715847,
77
+ "grad_norm": 5.7157392501831055,
78
+ "learning_rate": 8.114754098360657e-06,
79
+ "loss": 0.7717578887939454,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.3005464480874317,
84
+ "grad_norm": 11.109936714172363,
85
+ "learning_rate": 8.934426229508197e-06,
86
+ "loss": 0.6476659774780273,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.32786885245901637,
91
+ "grad_norm": 10.369423866271973,
92
+ "learning_rate": 9.754098360655738e-06,
93
+ "loss": 0.6707509994506836,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.3551912568306011,
98
+ "grad_norm": 10.730874061584473,
99
+ "learning_rate": 1.0573770491803279e-05,
100
+ "loss": 0.6787842750549317,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.3825136612021858,
105
+ "grad_norm": 8.120079040527344,
106
+ "learning_rate": 1.139344262295082e-05,
107
+ "loss": 0.6765446186065673,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.4098360655737705,
112
+ "grad_norm": 14.429889678955078,
113
+ "learning_rate": 1.221311475409836e-05,
114
+ "loss": 0.623370361328125,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.4371584699453552,
119
+ "grad_norm": 9.193517684936523,
120
+ "learning_rate": 1.3032786885245902e-05,
121
+ "loss": 0.6362186908721924,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.4644808743169399,
126
+ "grad_norm": 8.7772855758667,
127
+ "learning_rate": 1.3852459016393443e-05,
128
+ "loss": 0.6641538619995118,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.4918032786885246,
133
+ "grad_norm": 5.2151780128479,
134
+ "learning_rate": 1.4672131147540984e-05,
135
+ "loss": 0.6653701782226562,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.5191256830601093,
140
+ "grad_norm": 17.197834014892578,
141
+ "learning_rate": 1.5491803278688525e-05,
142
+ "loss": 0.6582591533660889,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.546448087431694,
147
+ "grad_norm": 7.26473331451416,
148
+ "learning_rate": 1.6311475409836068e-05,
149
+ "loss": 0.765175724029541,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.5737704918032787,
154
+ "grad_norm": 5.47290563583374,
155
+ "learning_rate": 1.7131147540983607e-05,
156
+ "loss": 0.676247501373291,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.6010928961748634,
161
+ "grad_norm": 4.55684232711792,
162
+ "learning_rate": 1.7950819672131146e-05,
163
+ "loss": 0.6748649597167968,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.6284153005464481,
168
+ "grad_norm": 9.52070140838623,
169
+ "learning_rate": 1.877049180327869e-05,
170
+ "loss": 0.7023877143859864,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.6557377049180327,
175
+ "grad_norm": 10.813623428344727,
176
+ "learning_rate": 1.959016393442623e-05,
177
+ "loss": 0.5075790405273437,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.6830601092896175,
182
+ "grad_norm": 15.77322769165039,
183
+ "learning_rate": 2.040983606557377e-05,
184
+ "loss": 0.5947454929351806,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.7103825136612022,
189
+ "grad_norm": 14.141736030578613,
190
+ "learning_rate": 2.122950819672131e-05,
191
+ "loss": 0.7629669666290283,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.7377049180327869,
196
+ "grad_norm": 0.5761239528656006,
197
+ "learning_rate": 2.2049180327868853e-05,
198
+ "loss": 0.717600679397583,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.7650273224043715,
203
+ "grad_norm": 13.026739120483398,
204
+ "learning_rate": 2.2868852459016393e-05,
205
+ "loss": 1.5919431686401366,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.7923497267759563,
210
+ "grad_norm": 12.15133285522461,
211
+ "learning_rate": 2.3688524590163936e-05,
212
+ "loss": 0.6558570861816406,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.819672131147541,
217
+ "grad_norm": 20.35292625427246,
218
+ "learning_rate": 2.4508196721311478e-05,
219
+ "loss": 0.6991987705230713,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.8469945355191257,
224
+ "grad_norm": 9.59134292602539,
225
+ "learning_rate": 2.5327868852459018e-05,
226
+ "loss": 1.1239334106445313,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.8743169398907104,
231
+ "grad_norm": 4.265232086181641,
232
+ "learning_rate": 2.6147540983606557e-05,
233
+ "loss": 0.4866987705230713,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.9016393442622951,
238
+ "grad_norm": 9.87460708618164,
239
+ "learning_rate": 2.69672131147541e-05,
240
+ "loss": 0.7435293674468995,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.9289617486338798,
245
+ "grad_norm": 12.707701683044434,
246
+ "learning_rate": 2.7786885245901642e-05,
247
+ "loss": 0.9141542434692382,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.9562841530054644,
252
+ "grad_norm": 2.9138967990875244,
253
+ "learning_rate": 2.860655737704918e-05,
254
+ "loss": 0.5319347858428956,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.9836065573770492,
259
+ "grad_norm": 6.917504787445068,
260
+ "learning_rate": 2.942622950819672e-05,
261
+ "loss": 0.5438683509826661,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 1.0,
266
+ "eval_accuracy": 0.8492,
267
+ "eval_confusion_matrix": [
268
+ [
269
+ 282,
270
+ 0
271
+ ],
272
+ [
273
+ 54,
274
+ 22
275
+ ]
276
+ ],
277
+ "eval_f1": 0.449,
278
+ "eval_loss": 0.41911664605140686,
279
+ "eval_pr_auc": 0.6187,
280
+ "eval_precision": 1.0,
281
+ "eval_recall": 0.2895,
282
+ "eval_roc_auc": 0.7485,
283
+ "eval_runtime": 138.7598,
284
+ "eval_samples_per_second": 2.58,
285
+ "eval_steps_per_second": 0.649,
286
+ "step": 366
287
+ }
288
+ ],
289
+ "logging_steps": 10,
290
+ "max_steps": 3660,
291
+ "num_input_tokens_seen": 0,
292
+ "num_train_epochs": 10,
293
+ "save_steps": 500,
294
+ "stateful_callbacks": {
295
+ "EarlyStoppingCallback": {
296
+ "args": {
297
+ "early_stopping_patience": 3,
298
+ "early_stopping_threshold": 0.0
299
+ },
300
+ "attributes": {
301
+ "early_stopping_patience_counter": 0
302
+ }
303
+ },
304
+ "TrainerControl": {
305
+ "args": {
306
+ "should_epoch_stop": false,
307
+ "should_evaluate": false,
308
+ "should_log": false,
309
+ "should_save": true,
310
+ "should_training_stop": false
311
+ },
312
+ "attributes": {}
313
+ }
314
+ },
315
+ "total_flos": 385194585047040.0,
316
+ "train_batch_size": 4,
317
+ "trial_name": null,
318
+ "trial_params": null
319
+ }
checkpoint-366/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:637685e84eea04f2c0ca07978bad2436247d122cfd9ea34a4a652aece0212809
3
+ size 5265
checkpoint-732/config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "RobertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "dtype": "float32",
10
+ "eos_token_id": 2,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "safe",
16
+ "1": "vulnerable"
17
+ },
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 3072,
20
+ "is_decoder": false,
21
+ "label2id": {
22
+ "safe": 0,
23
+ "vulnerable": 1
24
+ },
25
+ "layer_norm_eps": 1e-05,
26
+ "max_position_embeddings": 514,
27
+ "model_type": "roberta",
28
+ "num_attention_heads": 12,
29
+ "num_hidden_layers": 12,
30
+ "output_past": true,
31
+ "pad_token_id": 1,
32
+ "problem_type": "single_label_classification",
33
+ "tie_word_embeddings": true,
34
+ "transformers_version": "5.13.1",
35
+ "type_vocab_size": 1,
36
+ "use_cache": false,
37
+ "vocab_size": 50265
38
+ }
checkpoint-732/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5da7fc1cc3d2a44314e892efd438e575e36c9377ef8b8d64a5ef5e6540aa42b4
3
+ size 498612824
checkpoint-732/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0cc8d44f861aad7c561403c74e55a9e98661a7fe3d41efcdcab042f390dfccf
3
+ size 997340299
checkpoint-732/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2d49a356abaefde16bbcb66416bbb7851dc5f5c7c2b233e78d8544edf065935
3
+ size 14455
checkpoint-732/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cba395fc7db4460dbf029c42ce692fa36e3561fb1ee48ee3b97e55b98b167463
3
+ size 1465
checkpoint-732/trainer_state.json ADDED
@@ -0,0 +1,602 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 366,
3
+ "best_metric": 0.41911664605140686,
4
+ "best_model_checkpoint": "training_output/run_20260713_203733/model/checkpoint-366",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 732,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0273224043715847,
14
+ "grad_norm": 4.927176475524902,
15
+ "learning_rate": 7.377049180327868e-07,
16
+ "loss": 0.6899909496307373,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.0546448087431694,
21
+ "grad_norm": 6.0230512619018555,
22
+ "learning_rate": 1.557377049180328e-06,
23
+ "loss": 0.7346426486968994,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.08196721311475409,
28
+ "grad_norm": 6.2535271644592285,
29
+ "learning_rate": 2.377049180327869e-06,
30
+ "loss": 0.6693419456481934,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.1092896174863388,
35
+ "grad_norm": 10.038297653198242,
36
+ "learning_rate": 3.1967213114754097e-06,
37
+ "loss": 0.6888828277587891,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.1366120218579235,
42
+ "grad_norm": 10.041961669921875,
43
+ "learning_rate": 4.016393442622951e-06,
44
+ "loss": 0.623580026626587,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.16393442622950818,
49
+ "grad_norm": 9.107032775878906,
50
+ "learning_rate": 4.836065573770492e-06,
51
+ "loss": 0.692430830001831,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.1912568306010929,
56
+ "grad_norm": 9.090417861938477,
57
+ "learning_rate": 5.655737704918032e-06,
58
+ "loss": 0.6110644817352295,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.2185792349726776,
63
+ "grad_norm": 7.914913177490234,
64
+ "learning_rate": 6.4754098360655735e-06,
65
+ "loss": 0.6268288612365722,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.2459016393442623,
70
+ "grad_norm": 8.18041706085205,
71
+ "learning_rate": 7.2950819672131145e-06,
72
+ "loss": 0.6859075546264648,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.273224043715847,
77
+ "grad_norm": 5.7157392501831055,
78
+ "learning_rate": 8.114754098360657e-06,
79
+ "loss": 0.7717578887939454,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.3005464480874317,
84
+ "grad_norm": 11.109936714172363,
85
+ "learning_rate": 8.934426229508197e-06,
86
+ "loss": 0.6476659774780273,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.32786885245901637,
91
+ "grad_norm": 10.369423866271973,
92
+ "learning_rate": 9.754098360655738e-06,
93
+ "loss": 0.6707509994506836,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.3551912568306011,
98
+ "grad_norm": 10.730874061584473,
99
+ "learning_rate": 1.0573770491803279e-05,
100
+ "loss": 0.6787842750549317,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.3825136612021858,
105
+ "grad_norm": 8.120079040527344,
106
+ "learning_rate": 1.139344262295082e-05,
107
+ "loss": 0.6765446186065673,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.4098360655737705,
112
+ "grad_norm": 14.429889678955078,
113
+ "learning_rate": 1.221311475409836e-05,
114
+ "loss": 0.623370361328125,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.4371584699453552,
119
+ "grad_norm": 9.193517684936523,
120
+ "learning_rate": 1.3032786885245902e-05,
121
+ "loss": 0.6362186908721924,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.4644808743169399,
126
+ "grad_norm": 8.7772855758667,
127
+ "learning_rate": 1.3852459016393443e-05,
128
+ "loss": 0.6641538619995118,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.4918032786885246,
133
+ "grad_norm": 5.2151780128479,
134
+ "learning_rate": 1.4672131147540984e-05,
135
+ "loss": 0.6653701782226562,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.5191256830601093,
140
+ "grad_norm": 17.197834014892578,
141
+ "learning_rate": 1.5491803278688525e-05,
142
+ "loss": 0.6582591533660889,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.546448087431694,
147
+ "grad_norm": 7.26473331451416,
148
+ "learning_rate": 1.6311475409836068e-05,
149
+ "loss": 0.765175724029541,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.5737704918032787,
154
+ "grad_norm": 5.47290563583374,
155
+ "learning_rate": 1.7131147540983607e-05,
156
+ "loss": 0.676247501373291,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.6010928961748634,
161
+ "grad_norm": 4.55684232711792,
162
+ "learning_rate": 1.7950819672131146e-05,
163
+ "loss": 0.6748649597167968,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.6284153005464481,
168
+ "grad_norm": 9.52070140838623,
169
+ "learning_rate": 1.877049180327869e-05,
170
+ "loss": 0.7023877143859864,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.6557377049180327,
175
+ "grad_norm": 10.813623428344727,
176
+ "learning_rate": 1.959016393442623e-05,
177
+ "loss": 0.5075790405273437,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.6830601092896175,
182
+ "grad_norm": 15.77322769165039,
183
+ "learning_rate": 2.040983606557377e-05,
184
+ "loss": 0.5947454929351806,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.7103825136612022,
189
+ "grad_norm": 14.141736030578613,
190
+ "learning_rate": 2.122950819672131e-05,
191
+ "loss": 0.7629669666290283,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.7377049180327869,
196
+ "grad_norm": 0.5761239528656006,
197
+ "learning_rate": 2.2049180327868853e-05,
198
+ "loss": 0.717600679397583,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.7650273224043715,
203
+ "grad_norm": 13.026739120483398,
204
+ "learning_rate": 2.2868852459016393e-05,
205
+ "loss": 1.5919431686401366,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.7923497267759563,
210
+ "grad_norm": 12.15133285522461,
211
+ "learning_rate": 2.3688524590163936e-05,
212
+ "loss": 0.6558570861816406,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.819672131147541,
217
+ "grad_norm": 20.35292625427246,
218
+ "learning_rate": 2.4508196721311478e-05,
219
+ "loss": 0.6991987705230713,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.8469945355191257,
224
+ "grad_norm": 9.59134292602539,
225
+ "learning_rate": 2.5327868852459018e-05,
226
+ "loss": 1.1239334106445313,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.8743169398907104,
231
+ "grad_norm": 4.265232086181641,
232
+ "learning_rate": 2.6147540983606557e-05,
233
+ "loss": 0.4866987705230713,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.9016393442622951,
238
+ "grad_norm": 9.87460708618164,
239
+ "learning_rate": 2.69672131147541e-05,
240
+ "loss": 0.7435293674468995,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.9289617486338798,
245
+ "grad_norm": 12.707701683044434,
246
+ "learning_rate": 2.7786885245901642e-05,
247
+ "loss": 0.9141542434692382,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.9562841530054644,
252
+ "grad_norm": 2.9138967990875244,
253
+ "learning_rate": 2.860655737704918e-05,
254
+ "loss": 0.5319347858428956,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.9836065573770492,
259
+ "grad_norm": 6.917504787445068,
260
+ "learning_rate": 2.942622950819672e-05,
261
+ "loss": 0.5438683509826661,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 1.0,
266
+ "eval_accuracy": 0.8492,
267
+ "eval_confusion_matrix": [
268
+ [
269
+ 282,
270
+ 0
271
+ ],
272
+ [
273
+ 54,
274
+ 22
275
+ ]
276
+ ],
277
+ "eval_f1": 0.449,
278
+ "eval_loss": 0.41911664605140686,
279
+ "eval_pr_auc": 0.6187,
280
+ "eval_precision": 1.0,
281
+ "eval_recall": 0.2895,
282
+ "eval_roc_auc": 0.7485,
283
+ "eval_runtime": 138.7598,
284
+ "eval_samples_per_second": 2.58,
285
+ "eval_steps_per_second": 0.649,
286
+ "step": 366
287
+ },
288
+ {
289
+ "epoch": 1.010928961748634,
290
+ "grad_norm": 6.519931316375732,
291
+ "learning_rate": 2.9972677595628417e-05,
292
+ "loss": 0.796769380569458,
293
+ "step": 370
294
+ },
295
+ {
296
+ "epoch": 1.0382513661202186,
297
+ "grad_norm": 4.874457359313965,
298
+ "learning_rate": 2.98816029143898e-05,
299
+ "loss": 0.398636531829834,
300
+ "step": 380
301
+ },
302
+ {
303
+ "epoch": 1.0655737704918034,
304
+ "grad_norm": 15.488590240478516,
305
+ "learning_rate": 2.9790528233151184e-05,
306
+ "loss": 0.7061968803405761,
307
+ "step": 390
308
+ },
309
+ {
310
+ "epoch": 1.092896174863388,
311
+ "grad_norm": 19.74693489074707,
312
+ "learning_rate": 2.9699453551912568e-05,
313
+ "loss": 0.5211454391479492,
314
+ "step": 400
315
+ },
316
+ {
317
+ "epoch": 1.1202185792349726,
318
+ "grad_norm": 2.513188362121582,
319
+ "learning_rate": 2.9608378870673954e-05,
320
+ "loss": 0.7898036003112793,
321
+ "step": 410
322
+ },
323
+ {
324
+ "epoch": 1.1475409836065573,
325
+ "grad_norm": 25.31527328491211,
326
+ "learning_rate": 2.9517304189435338e-05,
327
+ "loss": 0.8195189476013184,
328
+ "step": 420
329
+ },
330
+ {
331
+ "epoch": 1.174863387978142,
332
+ "grad_norm": 26.276199340820312,
333
+ "learning_rate": 2.942622950819672e-05,
334
+ "loss": 0.8042145729064941,
335
+ "step": 430
336
+ },
337
+ {
338
+ "epoch": 1.2021857923497268,
339
+ "grad_norm": 19.838821411132812,
340
+ "learning_rate": 2.9335154826958105e-05,
341
+ "loss": 0.5532180786132812,
342
+ "step": 440
343
+ },
344
+ {
345
+ "epoch": 1.2295081967213115,
346
+ "grad_norm": 1.5704028606414795,
347
+ "learning_rate": 2.924408014571949e-05,
348
+ "loss": 0.46767144203186034,
349
+ "step": 450
350
+ },
351
+ {
352
+ "epoch": 1.2568306010928962,
353
+ "grad_norm": 0.24576333165168762,
354
+ "learning_rate": 2.9153005464480875e-05,
355
+ "loss": 0.7125775337219238,
356
+ "step": 460
357
+ },
358
+ {
359
+ "epoch": 1.2841530054644807,
360
+ "grad_norm": 2.8051435947418213,
361
+ "learning_rate": 2.9061930783242258e-05,
362
+ "loss": 1.4706084251403808,
363
+ "step": 470
364
+ },
365
+ {
366
+ "epoch": 1.3114754098360657,
367
+ "grad_norm": 2.1515614986419678,
368
+ "learning_rate": 2.897085610200364e-05,
369
+ "loss": 0.740934181213379,
370
+ "step": 480
371
+ },
372
+ {
373
+ "epoch": 1.3387978142076502,
374
+ "grad_norm": 5.335806846618652,
375
+ "learning_rate": 2.8879781420765028e-05,
376
+ "loss": 0.5567354679107666,
377
+ "step": 490
378
+ },
379
+ {
380
+ "epoch": 1.366120218579235,
381
+ "grad_norm": 9.499934196472168,
382
+ "learning_rate": 2.878870673952641e-05,
383
+ "loss": 0.5102160930633545,
384
+ "step": 500
385
+ },
386
+ {
387
+ "epoch": 1.3934426229508197,
388
+ "grad_norm": 0.30213791131973267,
389
+ "learning_rate": 2.8697632058287795e-05,
390
+ "loss": 0.8127809524536133,
391
+ "step": 510
392
+ },
393
+ {
394
+ "epoch": 1.4207650273224044,
395
+ "grad_norm": 17.827165603637695,
396
+ "learning_rate": 2.860655737704918e-05,
397
+ "loss": 1.3400749206542968,
398
+ "step": 520
399
+ },
400
+ {
401
+ "epoch": 1.4480874316939891,
402
+ "grad_norm": 2.58740234375,
403
+ "learning_rate": 2.8515482695810565e-05,
404
+ "loss": 0.5003597736358643,
405
+ "step": 530
406
+ },
407
+ {
408
+ "epoch": 1.4754098360655736,
409
+ "grad_norm": 22.007869720458984,
410
+ "learning_rate": 2.842440801457195e-05,
411
+ "loss": 1.6859323501586914,
412
+ "step": 540
413
+ },
414
+ {
415
+ "epoch": 1.5027322404371586,
416
+ "grad_norm": 14.658230781555176,
417
+ "learning_rate": 2.8333333333333332e-05,
418
+ "loss": 0.9195033073425293,
419
+ "step": 550
420
+ },
421
+ {
422
+ "epoch": 1.530054644808743,
423
+ "grad_norm": 1.4521483182907104,
424
+ "learning_rate": 2.8242258652094715e-05,
425
+ "loss": 0.4108741283416748,
426
+ "step": 560
427
+ },
428
+ {
429
+ "epoch": 1.5573770491803278,
430
+ "grad_norm": 10.377533912658691,
431
+ "learning_rate": 2.8151183970856102e-05,
432
+ "loss": 0.655418586730957,
433
+ "step": 570
434
+ },
435
+ {
436
+ "epoch": 1.5846994535519126,
437
+ "grad_norm": 0.5859184265136719,
438
+ "learning_rate": 2.8060109289617486e-05,
439
+ "loss": 1.2600435256958007,
440
+ "step": 580
441
+ },
442
+ {
443
+ "epoch": 1.6120218579234973,
444
+ "grad_norm": 0.8876230120658875,
445
+ "learning_rate": 2.796903460837887e-05,
446
+ "loss": 1.0982278823852538,
447
+ "step": 590
448
+ },
449
+ {
450
+ "epoch": 1.639344262295082,
451
+ "grad_norm": 13.285502433776855,
452
+ "learning_rate": 2.7877959927140256e-05,
453
+ "loss": 0.6685476779937745,
454
+ "step": 600
455
+ },
456
+ {
457
+ "epoch": 1.6666666666666665,
458
+ "grad_norm": 0.6555774211883545,
459
+ "learning_rate": 2.7786885245901642e-05,
460
+ "loss": 0.19812407493591308,
461
+ "step": 610
462
+ },
463
+ {
464
+ "epoch": 1.6939890710382515,
465
+ "grad_norm": 2.7470834255218506,
466
+ "learning_rate": 2.7695810564663026e-05,
467
+ "loss": 1.1334525108337403,
468
+ "step": 620
469
+ },
470
+ {
471
+ "epoch": 1.721311475409836,
472
+ "grad_norm": 14.42418098449707,
473
+ "learning_rate": 2.760473588342441e-05,
474
+ "loss": 0.9498401641845703,
475
+ "step": 630
476
+ },
477
+ {
478
+ "epoch": 1.748633879781421,
479
+ "grad_norm": 57.35378646850586,
480
+ "learning_rate": 2.7513661202185793e-05,
481
+ "loss": 1.2678983688354493,
482
+ "step": 640
483
+ },
484
+ {
485
+ "epoch": 1.7759562841530054,
486
+ "grad_norm": 0.7189310789108276,
487
+ "learning_rate": 2.742258652094718e-05,
488
+ "loss": 0.6106186389923096,
489
+ "step": 650
490
+ },
491
+ {
492
+ "epoch": 1.8032786885245902,
493
+ "grad_norm": 13.546385765075684,
494
+ "learning_rate": 2.7331511839708563e-05,
495
+ "loss": 1.4015119552612305,
496
+ "step": 660
497
+ },
498
+ {
499
+ "epoch": 1.830601092896175,
500
+ "grad_norm": 8.754522323608398,
501
+ "learning_rate": 2.7240437158469946e-05,
502
+ "loss": 0.9790468215942383,
503
+ "step": 670
504
+ },
505
+ {
506
+ "epoch": 1.8579234972677594,
507
+ "grad_norm": 13.214091300964355,
508
+ "learning_rate": 2.7149362477231333e-05,
509
+ "loss": 0.6752419948577881,
510
+ "step": 680
511
+ },
512
+ {
513
+ "epoch": 1.8852459016393444,
514
+ "grad_norm": 2.506131887435913,
515
+ "learning_rate": 2.7058287795992716e-05,
516
+ "loss": 0.2787914752960205,
517
+ "step": 690
518
+ },
519
+ {
520
+ "epoch": 1.9125683060109289,
521
+ "grad_norm": 0.4683574438095093,
522
+ "learning_rate": 2.69672131147541e-05,
523
+ "loss": 0.8134347915649414,
524
+ "step": 700
525
+ },
526
+ {
527
+ "epoch": 1.9398907103825138,
528
+ "grad_norm": 0.8958243131637573,
529
+ "learning_rate": 2.6876138433515483e-05,
530
+ "loss": 0.4489922046661377,
531
+ "step": 710
532
+ },
533
+ {
534
+ "epoch": 1.9672131147540983,
535
+ "grad_norm": 95.41824340820312,
536
+ "learning_rate": 2.678506375227687e-05,
537
+ "loss": 0.9731103897094726,
538
+ "step": 720
539
+ },
540
+ {
541
+ "epoch": 1.994535519125683,
542
+ "grad_norm": 0.7653670907020569,
543
+ "learning_rate": 2.6693989071038253e-05,
544
+ "loss": 1.0658687591552733,
545
+ "step": 730
546
+ },
547
+ {
548
+ "epoch": 2.0,
549
+ "eval_accuracy": 0.8659,
550
+ "eval_confusion_matrix": [
551
+ [
552
+ 281,
553
+ 1
554
+ ],
555
+ [
556
+ 47,
557
+ 29
558
+ ]
559
+ ],
560
+ "eval_f1": 0.5472,
561
+ "eval_loss": 0.45224741101264954,
562
+ "eval_pr_auc": 0.6674,
563
+ "eval_precision": 0.9667,
564
+ "eval_recall": 0.3816,
565
+ "eval_roc_auc": 0.8086,
566
+ "eval_runtime": 137.9939,
567
+ "eval_samples_per_second": 2.594,
568
+ "eval_steps_per_second": 0.652,
569
+ "step": 732
570
+ }
571
+ ],
572
+ "logging_steps": 10,
573
+ "max_steps": 3660,
574
+ "num_input_tokens_seen": 0,
575
+ "num_train_epochs": 10,
576
+ "save_steps": 500,
577
+ "stateful_callbacks": {
578
+ "EarlyStoppingCallback": {
579
+ "args": {
580
+ "early_stopping_patience": 3,
581
+ "early_stopping_threshold": 0.0
582
+ },
583
+ "attributes": {
584
+ "early_stopping_patience_counter": 1
585
+ }
586
+ },
587
+ "TrainerControl": {
588
+ "args": {
589
+ "should_epoch_stop": false,
590
+ "should_evaluate": false,
591
+ "should_log": false,
592
+ "should_save": true,
593
+ "should_training_stop": false
594
+ },
595
+ "attributes": {}
596
+ }
597
+ },
598
+ "total_flos": 770389170094080.0,
599
+ "train_batch_size": 4,
600
+ "trial_name": null,
601
+ "trial_params": null
602
+ }
checkpoint-732/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:637685e84eea04f2c0ca07978bad2436247d122cfd9ea34a4a652aece0212809
3
+ size 5265
config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "RobertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "dtype": "float32",
10
+ "eos_token_id": 2,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "safe",
16
+ "1": "vulnerable"
17
+ },
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 3072,
20
+ "is_decoder": false,
21
+ "label2id": {
22
+ "safe": 0,
23
+ "vulnerable": 1
24
+ },
25
+ "layer_norm_eps": 1e-05,
26
+ "max_position_embeddings": 514,
27
+ "model_type": "roberta",
28
+ "num_attention_heads": 12,
29
+ "num_hidden_layers": 12,
30
+ "output_past": true,
31
+ "pad_token_id": 1,
32
+ "problem_type": "single_label_classification",
33
+ "tie_word_embeddings": true,
34
+ "transformers_version": "5.13.1",
35
+ "type_vocab_size": 1,
36
+ "use_cache": false,
37
+ "vocab_size": 50265
38
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59da1e0b742c0b9f9ad0e6078e2195023ce00a1cffd33a1f5766d20a13aca4a5
3
+ size 498612824
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "errors": "replace",
8
+ "is_local": false,
9
+ "local_files_only": false,
10
+ "mask_token": "<mask>",
11
+ "model_max_length": 512,
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "tokenizer_class": "RobertaTokenizer",
15
+ "trim_offsets": true,
16
+ "unk_token": "<unk>"
17
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:637685e84eea04f2c0ca07978bad2436247d122cfd9ea34a4a652aece0212809
3
+ size 5265