learning_rate: 1e-05
train_batch_size: 16
eval_batch_size: 1
seed: 42
optimizer: Use OptimizerNames.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: 500
num_epochs: 1
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Wer
0.3468	0.0933	1000	0.3951	30.6669
0.2608	0.1866	2000	0.3316	25.5966
0.2264	0.2799	3000	0.3077	23.4182
0.2041	0.3731	4000	0.2867	22.0167
0.1891	0.4664	5000	0.2720	20.9015
0.1789	0.5597	6000	0.2669	20.0772
0.1708	0.6530	7000	0.2585	19.6288
0.1644	0.7463	8000	0.2524	18.8911
0.1604	0.8396	9000	0.2478	18.6108
0.155	0.9328	10000	0.2449	18.3818

mondhs/l3-whisper-medium-l2c_v6

l3-whisper-medium-l2c_v6