learning_rate: 0.0005705665024630542
train_batch_size: 16
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
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: 1
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Wer
2.3476	0.2436	100	0.4135	0.3730
0.4631	0.4872	200	0.3958	0.3640
0.4441	0.7308	300	0.3826	0.3493
0.4367	0.9744	400	0.3784	0.3522
0.424	1.2168	500	0.3734	0.3468
0.4256	1.4604	600	0.3760	0.3516
0.4189	1.7040	700	0.3739	0.3474
0.4293	1.9476	800	0.3678	0.3448
0.4105	2.1900	900	0.3663	0.3466
0.3981	2.4336	1000	0.3736	0.3441
0.4161	2.6772	1100	0.3725	0.3487
0.4188	2.9208	1200	0.3628	0.3434
0.3984	3.1632	1300	0.3671	0.3426
0.4108	3.4068	1400	0.3674	0.3410
0.3905	3.6504	1500	0.3593	0.3388
0.3977	3.8940	1600	0.3626	0.3387
0.3961	4.1364	1700	0.3610	0.3322
0.3844	4.3800	1800	0.3644	0.3423
0.3938	4.6236	1900	0.3554	0.3355
0.3808	4.8672	2000	0.3579	0.3349
0.3822	5.1096	2100	0.3562	0.3330
0.3755	5.3532	2200	0.3556	0.3307
0.3789	5.5968	2300	0.3514	0.3303
0.3742	5.8404	2400	0.3472	0.3328
0.3608	6.0828	2500	0.3470	0.3276
0.3647	6.3264	2600	0.3468	0.3295
0.3719	6.5700	2700	0.3457	0.3260
0.3678	6.8136	2800	0.3423	0.3179
0.3575	7.0560	2900	0.3422	0.3201
0.3427	7.2996	3000	0.3516	0.3232
0.3661	7.5432	3100	0.3420	0.3216
0.3502	7.7868	3200	0.3430	0.3238
0.3681	8.0292	3300	0.3388	0.3201
0.3454	8.2728	3400	0.3397	0.3218