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: 22
num_epochs: 3.0
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Wer
0.9396	0.2895	22	1.2268	43.7081
0.6061	0.5789	44	0.6882	27.0782
0.4696	0.8684	66	0.5846	26.1275
0.4267	1.1579	88	0.5511	23.4768
0.4024	1.4474	110	0.5330	23.7812
0.4006	1.7368	132	0.5215	24.5258
0.4201	2.0263	154	0.5162	22.2030
0.3841	2.3158	176	0.5107	22.0765
0.3876	2.6053	198	0.5083	21.9922
0.3926	2.8947	220	0.5069	21.9641

greenw0lf/whisper-new-nnat-10h-random-seed42

whisper-new-nnat-10h-random-seed42