learning_rate: 5e-06
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 32
optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 0.1
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1
0.3687	1.0	145	0.0521	0.9621	0.9328	0.9472
0.0851	2.0	290	0.0300	0.9728	0.9769	0.9748
0.1077	3.0	435	0.0265	0.9758	0.9758	0.9758
0.0657	4.0	580	0.0329	0.9747	0.9695	0.9721
0.0223	5.0	725	0.0382	0.9728	0.9769	0.9748
0.0415	6.0	870	0.0365	0.9840	0.9716	0.9778
0.0341	7.0	1015	0.0428	0.9840	0.9674	0.9756
0.0518	8.0	1160	0.0337	0.9779	0.9758	0.9769
0.0181	9.0	1305	0.0352	0.9820	0.9737	0.9778
0.0003	10.0	1450	0.0339	0.9831	0.9748	0.9789

buddhadeb33/outputs_Feb_2026

outputs_Feb_2026