learning_rate: 5e-06
train_batch_size: 16
eval_batch_size: 16
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: constant_with_warmup
lr_scheduler_warmup_steps: 250
training_steps: 2500

Training Loss	Epoch	Step	Validation Loss	Global Wer
0.5979	0.8197	250	0.4591	41.4638
0.1985	1.6393	500	0.2546	35.2767
0.1273	2.4590	750	0.2472	33.9095
0.0723	3.2787	1000	0.2472	32.2985
0.0637	4.0984	1250	0.2595	32.1113
0.0366	4.9180	1500	0.2670	31.7847
0.0219	5.7377	1750	0.2938	32.1156
0.0110	6.5574	2000	0.3168	33.1040
0.0085	7.3770	2250	0.3258	31.7760
0.0065	8.1967	2500	0.3433	31.6715

dianavdavidson/wh_small_cv_common_voices_52731__val_finetune_trial

wh_small_cv_common_voices_52731__val_finetune_trial