learning_rate: 0.0001
train_batch_size: 48
eval_batch_size: 32
seed: 42
optimizer: Use 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: 62
num_epochs: 3.0
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Wer
0.7477	0.2404	50	0.7460	32.1635
0.4284	0.4808	100	0.4838	16.5204
0.4038	0.7212	150	0.4521	15.9418
0.435	0.9615	200	0.4336	15.1764
0.3785	1.2019	250	0.4256	15.2884
0.3626	1.4423	300	0.4186	14.8777
0.3541	1.6827	350	0.4185	15.0084
0.3557	1.9231	400	0.4103	14.4111
0.3163	2.1635	450	0.4090	14.2990
0.35	2.4038	500	0.4060	14.5604
0.3243	2.6442	550	0.4041	14.3457
0.3429	2.8846	600	0.4041	14.3924

greenw0lf/whisper-n-nat-exp-all-adults

whisper-n-nat-exp-all-adults