learning_rate: 5e-05
train_batch_size: 16
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine_with_restarts
lr_scheduler_warmup_steps: 0.1
num_epochs: 5
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0677	0.4087	500	0.0285	0.8815	0.9724	0.9247	0.9913
0.0359	0.8173	1000	0.0150	0.9371	0.9710	0.9538	0.9957
0.0200	1.2256	1500	0.0127	0.9476	0.9846	0.9658	0.9965
0.0168	1.6342	2000	0.0109	0.9422	0.9741	0.9579	0.9965
0.0135	2.0425	2500	0.0096	0.9560	0.9695	0.9627	0.9971
0.0067	2.4512	3000	0.0128	0.9482	0.9583	0.9532	0.9965

jefftherover/modernbert-pii-mapped-v8

modernbert-pii-mapped-v8