learning_rate: 1e-05
train_batch_size: 32
eval_batch_size: 8
seed: 42
optimizer: Use 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_ratio: 0.1
num_epochs: 4

Training Loss	Epoch	Step	Validation Loss	Wer
1.2888	0.0862	5	1.1765	132.3735
1.0325	0.1724	10	0.9613	153.6187
0.861	0.2586	15	0.7908	128.9494
0.6462	0.3448	20	0.7104	96.4202
0.9435	0.4310	25	0.6531	113.2296
0.6052	0.5172	30	0.6207	97.1984
0.7387	0.6034	35	0.5958	91.2062
0.6503	0.6897	40	0.5807	80.6226
0.6283	0.7759	45	0.5656	100.1556
0.5325	0.8621	50	0.5547	93.3852
0.4403	0.9483	55	0.5594	84.5914
0.3437	1.0345	60	0.5479	90.8171
0.3554	1.1207	65	0.5415	80.5447
0.3463	1.2069	70	0.5401	100.4669
0.3314	1.2931	75	0.5450	100.2335
0.4401	1.3793	80	0.5370	80.7782
0.302	1.4655	85	0.5337	88.4825
0.2989	1.5517	90	0.5324	86.7704
0.3945	1.6379	95	0.5313	95.2529
0.4678	1.7241	100	0.5305	85.2918
0.3573	1.8103	105	0.5211	71.9066
0.2756	1.8966	110	0.5188	72.0623
0.2811	1.9828	115	0.5208	71.7510
0.2331	2.0690	120	0.5258	68.4825
0.1991	2.1552	125	0.5427	71.2062
0.1762	2.2414	130	0.5538	83.5798
0.2704	2.3276	135	0.5542	89.6498
0.1987	2.4138	140	0.5499	88.5603
0.2566	2.5	145	0.5573	84.6693
0.1981	2.5862	150	0.5609	74.3969
0.1835	2.6724	155	0.5524	83.7354
0.2074	2.7586	160	0.5462	80.0778
0.2059	2.8448	165	0.5472	69.4942
0.201	2.9310	170	0.5448	61.2451
0.1231	3.0172	175	0.5441	72.9961
0.0972	3.1034	180	0.5578	78.9105
0.1052	3.1897	185	0.5708	69.4163
0.1072	3.2759	190	0.5834	67.1595
0.0987	3.3621	195	0.5906	70.7393
0.1138	3.4483	200	0.5963	63.2685
0.0979	3.5345	205	0.5950	62.4125
0.1296	3.6207	210	0.5966	63.4241
0.0799	3.7069	215	0.5946	63.0350
0.1338	3.7931	220	0.5946	62.4125
0.089	3.8793	225	0.5939	63.1907
0.1048	3.9655	230	0.5930	62.8016

nkkbr/whisper-large-v3-zatoichi-ja-EX-2

Whisper Large v3 - Japanese Zatoichi ASR