learning_rate: 3e-05
train_batch_size: 16
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 30
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Per Hf Metric	Per Avg
0.3623	1.0	102	0.5230	0.0628	0.0859
0.298	2.0	204	0.7039	0.0646	0.0897
0.2867	3.0	306	0.6019	0.0641	0.0856
0.2837	4.0	408	0.6328	0.0638	0.0892
0.2509	5.0	510	0.6487	0.0610	0.0838
0.2483	6.0	612	0.5484	0.0606	0.0818
0.2405	7.0	714	0.5911	0.0577	0.0786
0.2452	8.0	816	0.5757	0.0620	0.0840
0.2282	9.0	918	0.5540	0.0591	0.0798
0.2066	10.0	1020	0.6506	0.0622	0.0861
0.2168	11.0	1122	0.6155	0.0598	0.0814

alinerodrigues/wav2vec2-large-xlsr-grosman-53-aug-words-phoneme-exp-1-v7

wav2vec2-large-xlsr-grosman-53-aug-words-phoneme-exp-1-v7