learning_rate: 0.0001
train_batch_size: 8
eval_batch_size: 8
seed: 42
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: linear
lr_scheduler_warmup_steps: 0.1
num_epochs: 4
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Exact Match	F1
2.3572	0.2818	2500	2.1112	40.1578	53.4230
1.6572	0.5636	5000	1.4769	55.1184	69.5452
1.4454	0.8455	7500	1.3624	58.1608	72.8710
1.3369	1.1273	10000	1.3156	59.9680	74.2630
1.3073	1.4091	12500	1.2375	61.4091	75.5437
1.2606	1.6909	15000	1.2085	62.5872	76.4412
1.2184	1.9727	17500	1.1566	62.6444	76.7167
1.1433	2.2545	20000	1.1639	62.7016	77.0749
1.1860	2.5364	22500	1.1396	63.1820	77.5246
1.1538	2.8182	25000	1.1430	63.3879	77.4303
1.0322	3.1000	27500	1.1489	63.6166	77.7949
1.0920	3.3818	30000	1.1070	63.8339	77.9774
1.0786	3.6636	32500	1.1126	63.9369	78.2527
1.1140	3.9454	35000	1.1090	63.9826	78.1693
1.1140	4.0	35484	1.1098	63.9369	78.1840

NastasiaM/mbert-loraxs-qa-rM-r64-qkvd-iscale-0.7

mbert-loraxs-qa-rM-r64-qkvd-iscale-0.7