learning_rate: 2e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Use OptimizerNames.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
0.3435	1.0	215	0.3388	0.2084	0.5895	0.3783
0.2792	2.0	430	0.2883	0.3809	0.6536	0.5041
0.2542	3.0	645	0.2818	0.4632	0.7008	0.5378
0.1826	4.0	860	0.3152	0.4773	0.6780	0.5308
0.1609	5.0	1075	0.3442	0.4930	0.6867	0.5041
0.1225	6.0	1290	0.3350	0.5364	0.7165	0.5041
0.1052	7.0	1505	0.3624	0.5266	0.7139	0.5111
0.0882	8.0	1720	0.3806	0.5441	0.7213	0.5367
0.0513	9.0	1935	0.4068	0.5399	0.7237	0.4831
0.0446	10.0	2150	0.4116	0.5559	0.7336	0.5355
0.036	11.0	2365	0.4638	0.5580	0.7343	0.5239
0.0229	12.0	2580	0.4976	0.5289	0.7171	0.4866
0.0254	13.0	2795	0.5105	0.5422	0.7340	0.5064
0.0143	14.0	3010	0.5074	0.5591	0.7351	0.5099
0.0083	15.0	3225	0.5311	0.5509	0.7346	0.5052
0.0073	16.0	3440	0.5451	0.5411	0.7236	0.5052
0.0099	17.0	3655	0.5425	0.5363	0.7192	0.5064
0.0088	18.0	3870	0.5537	0.5423	0.7238	0.4971

sercetexam9/afro-xlmr-large-76L-tir-finetuned-augmentation-LUNAR

afro-xlmr-large-76L-tir-finetuned-augmentation-LUNAR