learning_rate: 1e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Use adamw_torch 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: 1000
num_epochs: 1
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Wer
0.3195	0.0969	3000	0.5062	37.2976
0.2551	0.1938	6000	0.4393	35.2086
0.2327	0.2907	9000	0.4074	29.3827
0.2171	0.3876	12000	0.3905	29.3637
0.2114	0.4845	15000	0.3819	27.7804
0.2	0.5814	18000	0.3728	31.6371
0.1943	0.6783	21000	0.3657	27.8206
0.1918	0.7751	24000	0.3608	27.6068
0.1871	0.8720	27000	0.3540	29.6109
0.1865	0.9689	30000	0.3521	28.5190

mondhs/l3-whisper-small-l3c3n_e1_v4

l3-whisper-small-l3c3n_e1_v4