learning_rate: 2e-05
train_batch_size: 4
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 16
optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 17

Training Loss	Epoch	Step	Validation Loss
2.0744	1.0	37	1.7766
1.3055	2.0	74	0.8003
0.4971	3.0	111	0.3517
0.3507	4.0	148	0.3053
0.3101	5.0	185	0.2685
0.2721	6.0	222	0.2193
0.2233	7.0	259	0.1834
0.1958	8.0	296	0.1557
0.1741	9.0	333	0.1406
0.1608	10.0	370	0.1282
0.1483	11.0	407	0.1175
0.1368	12.0	444	0.1080
0.1293	13.0	481	0.1028
0.1238	14.0	518	0.0991
0.1208	15.0	555	0.0972
0.119	16.0	592	0.0963
0.1179	17.0	629	0.0962

RodrigoR07/FinetuneSmolVLMINSTRCUT

FinetuneSmolVLMINSTRCUT