learning_rate: 5e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 16
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: 22
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Accuracy	Dementia Recall
No log	1.7273	34	0.1023	0.2727	1.0
No log	3.4156	68	0.1008	0.2727	1.0
No log	5.1039	102	0.1023	0.2727	1.0
0.1091	6.8312	136	0.1023	0.2727	1.0
0.1091	8.5195	170	0.1017	0.2727	1.0
0.1091	10.2078	204	0.1017	0.2727	1.0
0.1010	11.9351	238	0.1022	0.2727	1.0
0.1010	13.6234	272	0.1004	0.2727	1.0
0.1010	15.3117	306	0.1011	0.2727	1.0
0.0960	17.0	340	0.1011	0.2727	1.0
0.0960	18.7273	374	0.1007	0.2727	1.0
0.0960	20.4156	408	0.1006	0.2727	1.0

m0nkspade/wav2vec2-large-xls-r-300m-dm32

wav2vec2-large-xls-r-300m-dm32