learning_rate: 0.001
train_batch_size: 2
eval_batch_size: 8
seed: 42
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
lr_scheduler_warmup_steps: 100
num_epochs: 5
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Wer
8.0676	0.1172	100	4.4018	1.0002
7.3239	0.2345	200	2.7893	1.0
1.0618	0.3517	300	0.8008	0.6929
0.7754	0.4689	400	0.7478	0.6571
0.7946	0.5862	500	0.7324	0.6474
0.7816	0.7034	600	0.7188	0.6480
0.7853	0.8206	700	0.6898	0.6192
0.7036	0.9379	800	0.6756	0.6323
0.7144	1.0551	900	0.6748	0.6228
0.7345	1.1723	1000	0.6361	0.6048
0.6324	1.2896	1100	0.6530	0.6159
0.7395	1.4068	1200	0.6167	0.5889
0.6685	1.5240	1300	0.6394	0.6229
0.8858	1.6413	1400	0.6168	0.5972
0.6431	1.7585	1500	0.6291	0.5875
0.7026	1.8757	1600	0.6552	0.6050
0.643	1.9930	1700	0.6156	0.5863
0.8062	2.1102	1800	0.6047	0.5817
0.6369	2.2274	1900	0.6448	0.6177
0.653	2.3447	2000	0.6089	0.5718
0.6353	2.4619	2100	0.6123	0.5774
0.6512	2.5791	2200	0.5982	0.5718
0.6253	2.6964	2300	0.5930	0.5699
0.6256	2.8136	2400	0.6017	0.5652
0.6276	2.9308	2500	0.5955	0.5626
0.6138	3.0481	2600	0.5832	0.5626
0.5827	3.1653	2700	0.5919	0.5612
0.7225	3.2825	2800	0.5834	0.5695
0.6179	3.3998	2900	0.5768	0.5615
0.6587	3.5170	3000	0.5895	0.5616
0.5442	3.6342	3100	0.5840	0.5632
0.6419	3.7515	3200	0.5817	0.5649
0.5755	3.8687	3300	0.5753	0.5601
0.5922	3.9859	3400	0.5740	0.5590
0.5895	4.1032	3500	0.5774	0.5526
0.5844	4.2204	3600	0.5775	0.5546
0.5867	4.3376	3700	0.5771	0.5624
0.5972	4.4549	3800	0.5742	0.5561
0.5205	4.5721	3900	0.5741	0.5543
0.7986	4.6893	4000	0.5721	0.5535
0.5652	4.8066	4100	0.5752	0.5532
0.5528	4.9238	4200	0.5712	0.5541

robertp408/wav2vec2-large-mms-1b-aft-lke

wav2vec2-large-mms-1b-aft-lke