learning_rate: 0.001
train_batch_size: 512
eval_batch_size: 512
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
num_epochs: 40

Training Loss	Epoch	Step	Validation Loss
3.4966	1.0	5	2.9381
2.5644	2.0	10	2.1833
1.9501	3.0	15	1.7482
1.6572	4.0	20	1.6286
1.5672	5.0	25	1.5984
1.5456	6.0	30	1.5522
1.5138	7.0	35	1.5623
1.5187	8.0	40	1.5271
1.4693	9.0	45	1.4929
1.4264	10.0	50	1.4364
1.3861	11.0	55	1.3868
1.3580	12.0	60	1.3506
1.3041	13.0	65	1.3024
1.2614	14.0	70	1.2353
1.1888	15.0	75	1.1557
1.1243	16.0	80	1.0897
1.0817	17.0	85	1.0597
1.0505	18.0	90	1.0613
1.0286	19.0	95	0.9825
0.9680	20.0	100	0.9919
0.9733	21.0	105	0.9443
0.9319	22.0	110	0.9685
0.9913	23.0	115	0.9770
0.9304	24.0	120	0.9019
0.8845	25.0	125	0.8840
0.8630	26.0	130	0.8568
0.8613	27.0	135	0.8521
0.8631	28.0	140	0.8390
0.8458	29.0	145	0.8094
0.8052	30.0	150	0.7904
0.7919	31.0	155	0.8096
0.7809	32.0	160	0.7677
0.7595	33.0	165	0.7542
0.7481	34.0	170	0.7461
0.7381	35.0	175	0.7301
0.7230	36.0	180	0.7188
0.7149	37.0	185	0.7115
0.7137	38.0	190	0.7115
0.7057	39.0	195	0.7009
0.7026	40.0	200	0.7002

zajacx/calculator_model_test

calculator_model_test