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
7.1189	0.1103	100	3.9365	0.9997
3.7311	0.2205	200	3.4926	0.9992
3.183	0.3308	300	3.2031	0.9983
3.1554	0.4410	400	3.1032	0.9984
2.83	0.5513	500	2.6496	1.1285
2.1028	0.6615	600	1.2437	0.8693
1.1593	0.7718	700	0.9672	0.7664
0.9267	0.8820	800	0.8151	0.6238
0.8494	0.9923	900	0.7620	0.5697
0.8212	1.1025	1000	0.7095	0.5130
0.8276	1.2128	1100	0.6966	0.5023
0.7277	1.3230	1200	0.6713	0.4840
0.7475	1.4333	1300	0.6426	0.4679
0.7154	1.5436	1400	0.6239	0.4445
0.6515	1.6538	1500	0.6159	0.4420
0.6596	1.7641	1600	0.6075	0.4401
0.6902	1.8743	1700	0.6162	0.4396
0.6291	1.9846	1800	0.6043	0.4320
0.5982	2.0948	1900	0.5982	0.4204
0.604	2.2051	2000	0.5894	0.4126
0.6337	2.3153	2100	0.5890	0.4133
0.6308	2.4256	2200	0.5793	0.4054
0.6905	2.5358	2300	0.5754	0.4057
0.6395	2.6461	2400	0.5744	0.4023
0.5888	2.7563	2500	0.5620	0.3997
0.581	2.8666	2600	0.5660	0.3954
0.585	2.9768	2700	0.5641	0.3986
0.5605	3.0871	2800	0.5609	0.3927
0.6057	3.1974	2900	0.5695	0.4123
0.5301	3.3076	3000	0.5537	0.3775
0.576	3.4179	3100	0.5476	0.3836
0.5543	3.5281	3200	0.5473	0.3769
0.5143	3.6384	3300	0.5394	0.3760
0.5844	3.7486	3400	0.5377	0.3697
0.5905	3.8589	3500	0.5384	0.3718
0.5353	3.9691	3600	0.5320	0.3673
0.5245	4.0794	3700	0.5293	0.3667
0.4384	4.1896	3800	0.5388	0.3620
0.5966	4.2999	3900	0.5272	0.3704
0.5916	4.4101	4000	0.5238	0.3644
0.4985	4.5204	4100	0.5224	0.3613
0.5135	4.6307	4200	0.5241	0.3588
0.53	4.7409	4300	0.5187	0.3594
0.5106	4.8512	4400	0.5188	0.3595
0.4574	4.9614	4500	0.5188	0.3580

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

wav2vec2-large-mms-1b-aft-pne