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: 7
num_epochs: 3.0
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Wer
1.1745	0.2174	5	1.3261	44.5865
0.9645	0.4348	10	1.2003	43.2518
0.7911	0.6522	15	1.0692	42.3931
0.7656	0.8696	20	0.9521	38.9864
0.7238	1.0870	25	0.8509	37.5770
0.6382	1.3043	30	0.7706	31.0155
0.6145	1.5217	35	0.7130	26.0314
0.6849	1.7391	40	0.6744	21.6072
0.4682	1.9565	45	0.6454	20.8326
0.6015	2.1739	50	0.6239	19.8712
0.5553	2.3913	55	0.6087	19.8805
0.5006	2.6087	60	0.5987	19.6005
0.403	2.8261	65	0.5932	19.6005

greenw0lf/whisper-nonnat-5h-maxcos-xlsr

whisper-nonnat-5h-maxcos-xlsr