learning_rate: 1e-05
train_batch_size: 2
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 8
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: cosine
lr_scheduler_warmup_steps: 500
training_steps: 2000

Training Loss	Epoch	Step	Validation Loss
2.1095	0.7207	100	1.5263
1.0019	1.4396	200	0.6586
0.2729	2.1586	300	0.1942
0.1914	2.8793	400	0.1489
0.1573	3.5982	500	0.1290
0.1375	4.3171	600	0.1182
0.1299	5.0360	700	0.1134
0.1157	5.7568	800	0.1098
0.1307	6.4757	900	0.1066
0.1011	7.1946	1000	0.1033
0.1046	7.9153	1100	0.1026
0.1009	8.6342	1200	0.1006
0.1007	9.3532	1300	0.0996
0.0917	10.0721	1400	0.0991
0.0997	10.7928	1500	0.0978
0.1066	11.5117	1600	0.0973
0.1021	12.2306	1700	0.0970
0.0948	12.9514	1800	0.0967
0.0846	13.6703	1900	0.0966
0.0962	14.3892	2000	0.0966

mlundberg/mlue_amt_run02

mlue_amt_run02