learning_rate: 5e-05
train_batch_size: 32
eval_batch_size: 128
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 128
optimizer: Use adamw_hf with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 10

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
2.622	1.0	26	0.6508	0.6349	0.5900	0.5758	0.5758
2.5166	2.0	52	0.6293	0.6703	0.7476	0.5500	0.4956
2.5527	3.0	78	0.6549	0.6022	0.6064	0.6150	0.5961
2.3672	4.0	104	0.5995	0.6975	0.7001	0.6087	0.6017
1.9234	5.0	130	0.6055	0.6839	0.6574	0.6564	0.6569
0.9818	6.0	156	0.8319	0.6676	0.6434	0.6468	0.6448
0.3056	7.0	182	0.9884	0.6730	0.6484	0.6511	0.6495
0.0518	8.0	208	1.2367	0.6730	0.6492	0.6527	0.6506
0.0083	9.0	234	1.2961	0.6839	0.6586	0.6596	0.6591
0.0023	9.6214	250	1.3402	0.6948	0.6664	0.6471	0.6518

callmesan/ModernBERT-large-hinglish-binary

ModernBERT-large-hinglish-binary