learning_rate: 2e-05
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 100
num_epochs: 20

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.3622	1.0	108	0.3550	0.0	0.5	0.2369
0.3379	2.0	216	0.3358	0.0942	0.5309	0.3502
0.3103	3.0	324	0.3205	0.1286	0.5500	0.4146
0.2932	4.0	432	0.3268	0.1689	0.5622	0.4303
0.2732	5.0	540	0.3086	0.1758	0.5631	0.4355
0.2489	6.0	648	0.3074	0.2167	0.5734	0.4408
0.2428	7.0	756	0.2987	0.2273	0.5803	0.4634
0.22	8.0	864	0.2971	0.2467	0.5888	0.4808
0.1866	9.0	972	0.3061	0.3799	0.6492	0.5105
0.1667	10.0	1080	0.3213	0.3645	0.6397	0.5035
0.1628	11.0	1188	0.3132	0.3835	0.6539	0.5105
0.1509	12.0	1296	0.3225	0.3915	0.6547	0.5052
0.135	13.0	1404	0.3189	0.4136	0.6634	0.5261
0.1434	14.0	1512	0.3221	0.3949	0.6541	0.5157
0.1253	15.0	1620	0.3331	0.4210	0.6658	0.5105
0.13	16.0	1728	0.3298	0.4164	0.6610	0.5139
0.1178	17.0	1836	0.3305	0.4163	0.6621	0.5174
0.1148	18.0	1944	0.3340	0.4468	0.6765	0.5226
0.1087	19.0	2052	0.3357	0.4378	0.6725	0.5209
0.1136	20.0	2160	0.3354	0.4399	0.6736	0.5226

Kuongan/xlm-roberta-base-orm-noaug

xlm-roberta-base-orm-noaug