learning_rate: 0.0003
train_batch_size: 8
eval_batch_size: 8
seed: 2025
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: linear
num_epochs: 15

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1
0.6258	1.0	447	0.8530	0.4362	0.5942	0.5031
0.6041	2.0	894	1.0887	0.4151	0.3188	0.3607
0.5717	3.0	1341	1.0252	0.4206	0.5761	0.4862
0.5455	4.0	1788	1.2918	0.4265	0.6304	0.5088
0.4904	5.0	2235	1.9671	0.4	0.4565	0.4264
0.4739	6.0	2682	1.7556	0.4039	0.5254	0.4567
0.4421	7.0	3129	2.4087	0.3928	0.5109	0.4441
0.4086	8.0	3576	2.2270	0.3970	0.4746	0.4323
0.427	9.0	4023	2.3055	0.3668	0.3841	0.3752
0.4233	10.0	4470	2.5657	0.3969	0.4674	0.4293
0.3891	11.0	4917	2.6900	0.3724	0.3913	0.3816
0.2841	12.0	5364	2.9719	0.3811	0.4239	0.4014
0.3323	13.0	5811	3.1814	0.3905	0.4457	0.4162
0.3462	14.0	6258	3.3791	0.3838	0.4130	0.3979
0.3005	15.0	6705	3.5377	0.3859	0.4167	0.4007

maud-dr/baseline_3-seed_2025

baseline_3-seed_2025