learning_rate: 2.2e-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: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
1.2387	0.0065	1	1.2545	98.1567	40.8419
1.2196	0.0130	2	0.9953	96.2212	38.3242
1.0997	0.0195	3	0.9733	95.4839	37.7781
0.9178	0.0260	4	0.9377	91.6129	36.0863
0.9721	0.0325	5	0.9072	87.4654	34.1681
0.9532	0.0390	6	0.8430	85.4378	29.9587
1.0501	0.0455	7	0.8121	83.5023	28.4401
0.9397	0.0519	8	0.7631	97.7880	32.0234
0.8483	0.0584	9	0.7368	88.2028	32.3165
0.791	0.0649	10	0.7102	81.0138	30.2651
0.7428	0.0714	11	0.6773	102.3041	35.5402
0.8248	0.0779	12	0.6520	98.5253	31.9036
0.6883	0.0844	13	0.6314	81.1060	29.1062
0.6443	0.0909	14	0.6146	84.4240	28.3735
0.6296	0.0974	15	0.6003	79.9078	26.8283
0.6931	0.1039	16	0.5896	84.0553	26.7350
0.6001	0.1104	17	0.5811	83.9631	26.8683
0.4982	0.1169	18	0.5728	85.2535	26.4820
0.6864	0.1234	19	0.5733	78.8018	25.0966
0.568	0.1299	20	0.5656	76.4977	24.5371
0.5281	0.1364	21	0.5546	74.0092	23.8711
0.5937	0.1429	22	0.5504	79.8157	26.0424
0.6851	0.1494	23	0.5529	85.4378	30.6381
0.5729	0.1558	24	0.5480	91.7051	32.9692
0.6369	0.1623	25	0.5458	93.0876	31.0111
0.5485	0.1688	26	0.5456	85.4378	27.6142
0.5796	0.1753	27	0.5409	84.7926	27.5476
0.5651	0.1818	28	0.5395	75.9447	22.7654
0.6338	0.1883	29	0.5328	75.9447	23.9909
0.5372	0.1948	30	0.5289	76.4977	22.6056
0.5036	0.2013	31	0.5330	77.5115	24.5504
0.6579	0.2078	32	0.5295	76.1290	25.1499
0.5241	0.2143	33	0.5217	71.8894	22.7121
0.5133	0.2208	34	0.5156	74.1935	22.1926
0.4856	0.2273	35	0.5171	76.8664	22.6056
0.6538	0.2338	36	0.5166	75.3917	22.6988
0.5058	0.2403	37	0.5142	72.1659	22.1926
0.4749	0.2468	38	0.5148	70.8756	21.8729
0.5914	0.2532	39	0.5117	70.7834	21.6997
0.5161	0.2597	40	0.5085	70.9677	21.7930
0.5804	0.2662	41	0.5091	73.5484	22.0594
0.64	0.2727	42	0.5107	77.5115	22.4058
0.5374	0.2792	43	0.5034	74.3779	21.9662
0.5593	0.2857	44	0.5062	73.2719	21.8862
0.4657	0.2922	45	0.5181	73.3641	22.2326
0.4997	0.2987	46	0.5099	73.4562	22.0727
0.6306	0.3052	47	0.5019	72.8111	22.4058
0.5612	0.3117	48	0.5027	74.1014	22.2725
0.5751	0.3182	49	0.5013	75.5760	22.6455
0.5521	0.3247	50	0.4981	83.5945	25.8692
0.4115	0.3312	51	0.4940	85.5300	28.2936
0.4882	0.3377	52	0.4903	85.7143	28.5200
0.473	0.3442	53	0.4922	74.2857	25.1632
0.6014	0.3506	54	0.4911	77.6037	27.2412
0.5046	0.3571	55	0.4860	73.7327	24.7103
0.4508	0.3636	56	0.4829	77.2350	25.2564
0.4719	0.3701	57	0.4832	78.9862	25.6294
0.5942	0.3766	58	0.4822	81.2903	25.1232
0.4674	0.3831	59	0.4847	80.1843	24.4971
0.4823	0.3896	60	0.4852	72.4424	21.4200
0.4015	0.3961	61	0.4861	69.4931	20.5808
0.4773	0.4026	62	0.4836	67.4654	21.2468
0.4752	0.4091	63	0.4794	68.3871	21.2602
0.5606	0.4156	64	0.4730	68.0184	20.9538
0.4165	0.4221	65	0.4731	72.3502	21.8596
0.4667	0.4286	66	0.4827	85.0691	24.2307
0.544	0.4351	67	0.4796	84.7005	24.3506
0.587	0.4416	68	0.4689	76.4977	23.6180
0.4802	0.4481	69	0.4647	74.1014	23.2716
0.5787	0.4545	70	0.4667	70.6912	21.8196
0.4201	0.4610	71	0.4676	70.6912	22.0194
0.4908	0.4675	72	0.4653	71.9816	21.3800
0.4464	0.4740	73	0.4694	74.7465	21.9662
0.4895	0.4805	74	0.4666	74.1935	20.9405
0.4652	0.4870	75	0.4592	72.6267	20.6208
0.4538	0.4935	76	0.4574	75.0230	22.7121
0.5679	0.5	77	0.4571	72.9032	23.7911
0.4652	0.5065	78	0.4519	70.8756	20.7540
0.3902	0.5130	79	0.4511	72.7189	20.4209
0.4752	0.5195	80	0.4523	73.4562	20.5541
0.4174	0.5260	81	0.4520	70.4147	20.5808
0.4411	0.5325	82	0.4476	69.7696	21.9662
0.4656	0.5390	83	0.4469	68.8479	22.6056
0.4624	0.5455	84	0.4509	69.4931	20.5941
0.4579	0.5519	85	0.4549	72.4424	21.9262
0.5003	0.5584	86	0.4558	75.4839	23.4315
0.4901	0.5649	87	0.4551	82.4885	24.3240
0.5715	0.5714	88	0.4512	75.2074	22.0461
0.4613	0.5779	89	0.4468	71.1521	20.6208
0.5258	0.5844	90	0.4460	70.1382	21.0870
0.5251	0.5909	91	0.4450	68.5714	20.9804
0.585	0.5974	92	0.4420	67.1889	20.3011
0.5782	0.6039	93	0.4383	67.6498	19.7283
0.4054	0.6104	94	0.4385	69.0323	20.3543
0.4843	0.6169	95	0.4460	71.4286	20.3677
0.4768	0.6234	96	0.4527	74.1014	20.7140
0.5296	0.6299	97	0.4482	76.0369	20.8339
0.5168	0.6364	98	0.4457	76.0369	21.9662
0.4001	0.6429	99	0.4470	75.8525	22.0194
0.3331	0.6494	100	0.4452	69.3088	21.0071
0.5514	0.6558	101	0.4441	66.6359	20.5142
0.4749	0.6623	102	0.4419	67.0968	20.8339
0.4486	0.6688	103	0.4411	69.6774	21.0204
0.424	0.6753	104	0.4408	69.3088	20.7673
0.5302	0.6818	105	0.4408	72.3502	21.1669
0.5155	0.6883	106	0.4427	72.5346	21.3134
0.4792	0.6948	107	0.4415	70.2304	20.9804
0.4364	0.7013	108	0.4389	72.1659	22.2725
0.4848	0.7078	109	0.4357	70.7834	20.9005
0.4912	0.7143	110	0.4323	68.1106	19.9414
0.4015	0.7208	111	0.4313	70.0461	20.7140
0.4851	0.7273	112	0.4324	72.3502	19.8215
0.4781	0.7338	113	0.4331	73.4562	19.8348
0.6363	0.7403	114	0.4331	75.8525	20.0613
0.4136	0.7468	115	0.4302	71.9816	19.8615
0.3998	0.7532	116	0.4255	68.4793	19.7149
0.4568	0.7597	117	0.4268	69.3088	20.8072
0.3839	0.7662	118	0.4285	69.1244	20.7540
0.5634	0.7727	119	0.4276	66.5438	19.4086
0.4726	0.7792	120	0.4262	68.4793	19.6217
0.4532	0.7857	121	0.4279	67.3733	19.8748
0.4535	0.7922	122	0.4314	70.4147	19.7416
0.3747	0.7987	123	0.4319	72.7189	20.5541
0.4561	0.8052	124	0.4298	68.0184	19.2354
0.4191	0.8117	125	0.4288	72.5346	20.8872
0.4094	0.8182	126	0.4321	76.4977	21.0337
0.438	0.8247	127	0.4296	77.0507	21.0870
0.6168	0.8312	128	0.4240	75.6682	20.7806
0.455	0.8377	129	0.4190	74.4700	21.2468
0.4381	0.8442	130	0.4176	74.0092	20.6740
0.4561	0.8506	131	0.4171	75.8525	21.7264
0.4084	0.8571	132	0.4159	71.6129	20.2344
0.4567	0.8636	133	0.4173	66.7281	19.3952
0.4511	0.8701	134	0.4202	65.9908	18.9690
0.4872	0.8766	135	0.4205	67.3733	19.2753
0.5269	0.8831	136	0.4199	72.3502	20.8472
0.3945	0.8896	137	0.4193	72.8111	20.9937
0.4285	0.8961	138	0.4180	71.3364	20.7406
0.4399	0.9026	139	0.4170	68.4793	19.2354
0.3326	0.9091	140	0.4175	72.9954	21.5932
0.4036	0.9156	141	0.4176	71.2442	21.1936
0.4123	0.9221	142	0.4165	66.1751	18.5827
0.3858	0.9286	143	0.4178	66.4516	18.7159
0.4965	0.9351	144	0.4200	67.7419	18.8491
0.4524	0.9416	145	0.4215	68.1106	18.8757
0.3694	0.9481	146	0.4214	67.8341	19.0089
0.4985	0.9545	147	0.4189	71.8894	20.9405
0.4025	0.9610	148	0.4175	68.6636	20.0080
0.3589	0.9675	149	0.4173	68.7558	19.9947
0.4562	0.9740	150	0.4178	70.2304	20.4076
0.3679	0.9805	151	0.4197	72.7189	20.8339
0.3805	0.9870	152	0.4191	73.2719	21.1003
0.4967	0.9935	153	0.4196	69.6774	19.7949
0.4382	1.0	154	0.4198	70.2304	19.9814
0.2325	1.0065	155	0.4178	70.1382	20.0480
0.3524	1.0130	156	0.4161	68.8479	19.8348
0.2725	1.0195	157	0.4153	67.4654	19.3020
0.2413	1.0260	158	0.4155	66.8203	19.0755
0.3114	1.0325	159	0.4163	66.1751	18.6226
0.2559	1.0390	160	0.4180	66.3594	18.5161
0.2572	1.0455	161	0.4209	68.2949	18.8624
0.2583	1.0519	162	0.4263	71.0599	19.3153
0.2926	1.0584	163	0.4298	72.3502	19.1555
0.2772	1.0649	164	0.4282	72.4424	19.1421
0.3135	1.0714	165	0.4220	68.1106	18.8757
0.2704	1.0779	166	0.4196	65.2535	18.7025
0.2479	1.0844	167	0.4187	64.9770	18.8091
0.2063	1.0909	168	0.4169	63.5945	18.5294
0.2563	1.0974	169	0.4151	64.5161	18.3962
0.2294	1.1039	170	0.4168	68.4793	18.9423
0.2862	1.1104	171	0.4225	72.5346	19.3553
0.2319	1.1169	172	0.4239	72.9954	19.4618
0.27	1.1234	173	0.4225	71.7051	19.3952
0.2993	1.1299	174	0.4191	69.8618	19.4352
0.2681	1.1364	175	0.4171	66.3594	19.1155
0.2655	1.1429	176	0.4168	64.4240	18.7292
0.2654	1.1494	177	0.4148	64.3318	18.7691
0.2787	1.1558	178	0.4124	69.4009	20.4343
0.2525	1.1623	179	0.4110	72.8111	20.7140
0.2678	1.1688	180	0.4135	76.2212	21.0737
0.2449	1.1753	181	0.4190	72.9954	19.3153
0.2328	1.1818	182	0.4206	73.1797	19.4618
0.267	1.1883	183	0.4173	71.7972	19.4086
0.2566	1.1948	184	0.4133	69.5853	18.9556
0.3085	1.2013	185	0.4094	71.8894	20.5541
0.2483	1.2078	186	0.4073	71.2442	20.3677
0.2356	1.2143	187	0.4077	69.8618	20.0346
0.1994	1.2208	188	0.4092	71.0599	20.4209
0.1976	1.2273	189	0.4130	69.4009	18.8358
0.2179	1.2338	190	0.4169	71.9816	19.3153
0.2384	1.2403	191	0.4198	75.2074	19.5684
0.2273	1.2468	192	0.4207	75.2995	19.6483
0.2735	1.2532	193	0.4185	72.7189	19.4219
0.2307	1.2597	194	0.4163	70.5069	19.2487
0.2786	1.2662	195	0.4145	67.0968	18.8757
0.2382	1.2727	196	0.4142	65.6221	18.7025
0.267	1.2792	197	0.4128	65.7143	18.5693
0.2544	1.2857	198	0.4129	66.3594	18.6093
0.2382	1.2922	199	0.4149	69.9539	19.0889
0.2328	1.2987	200	0.4188	73.1797	19.5684
0.2633	1.3052	201	0.4209	73.7327	19.5551
0.2298	1.3117	202	0.4171	72.3502	19.3419
0.2535	1.3182	203	0.4130	68.2028	18.8491
0.2861	1.3247	204	0.4107	71.0599	20.6740
0.2535	1.3312	205	0.4111	69.8618	20.5941
0.27	1.3377	206	0.4118	70.5069	20.6208
0.3346	1.3442	207	0.4106	71.1521	20.6341
0.2653	1.3506	208	0.4107	73.3641	20.9804
0.2144	1.3571	209	0.4119	76.3134	21.2202
0.2678	1.3636	210	0.4133	79.6313	21.6864
0.2197	1.3701	211	0.4132	74.8387	19.5950
0.2402	1.3766	212	0.4108	73.7327	19.4885
0.2726	1.3831	213	0.4081	72.1659	19.4485
0.245	1.3896	214	0.4053	69.7696	19.1288
0.2348	1.3961	215	0.4030	67.2811	18.7958
0.2291	1.4026	216	0.4026	66.0829	18.6493
0.2075	1.4091	217	0.4024	66.1751	18.6359
0.2469	1.4156	218	0.4028	66.1751	18.7159
0.273	1.4221	219	0.4040	66.0829	18.5161
0.2214	1.4286	220	0.4057	66.7281	18.6892
0.2726	1.4351	221	0.4071	68.1106	18.8757
0.2687	1.4416	222	0.4078	68.2949	18.5693
0.231	1.4481	223	0.4086	70.0461	18.7825
0.2646	1.4545	224	0.4080	69.8618	18.6226
0.2777	1.4610	225	0.4070	69.4931	18.5027
0.2863	1.4675	226	0.4067	69.5853	18.4628
0.2003	1.4740	227	0.4068	69.2166	18.3962
0.3105	1.4805	228	0.4062	69.3088	18.4361
0.2364	1.4870	229	0.4046	69.3088	18.5161
0.2222	1.4935	230	0.4027	68.4793	18.5027
0.2237	1.5	231	0.4012	68.3871	18.4628
0.2386	1.5065	232	0.4001	67.9263	18.5161
0.2043	1.5130	233	0.3997	67.9263	18.5693
0.2588	1.5195	234	0.3999	67.9263	18.6626
0.1928	1.5260	235	0.3996	67.7419	18.6759
0.2247	1.5325	236	0.3988	68.1106	18.6226
0.2903	1.5390	237	0.3978	68.2028	18.4628
0.2216	1.5455	238	0.3968	67.0046	18.1164
0.2584	1.5519	239	0.3963	66.9124	18.1297
0.244	1.5584	240	0.3959	67.0968	17.8633
0.2353	1.5649	241	0.3957	67.0046	17.9433
0.2445	1.5714	242	0.3952	66.9124	17.8766
0.1748	1.5779	243	0.3947	67.0046	17.9299
0.2242	1.5844	244	0.3940	66.5438	17.9566
0.2231	1.5909	245	0.3929	66.1751	18.0631
0.2675	1.5974	246	0.3923	67.1889	18.2763
0.31	1.6039	247	0.3920	66.2673	18.1164
0.2517	1.6104	248	0.3917	66.4516	18.3296
0.2825	1.6169	249	0.3923	66.9124	18.2763
0.3018	1.6234	250	0.3926	66.4516	18.2896
0.2569	1.6299	251	0.3925	66.8203	18.3562
0.2173	1.6364	252	0.3929	66.6359	18.2097
0.2598	1.6429	253	0.3929	66.9124	18.0631
0.202	1.6494	254	0.3928	66.7281	18.0365
0.2168	1.6558	255	0.3937	67.4654	18.1964
0.2233	1.6623	256	0.3943	68.2949	18.3029
0.2647	1.6688	257	0.3952	68.7558	18.2097
0.2529	1.6753	258	0.3957	70.1382	18.2630
0.2106	1.6818	259	0.3966	70.5991	18.2230
0.2538	1.6883	260	0.3972	70.6912	18.2097
0.2781	1.6948	261	0.3973	71.0599	18.2630
0.1988	1.7013	262	0.3967	70.9677	18.2896
0.1784	1.7078	263	0.3963	71.1521	18.3162
0.2529	1.7143	264	0.3951	69.7696	18.2496
0.2315	1.7208	265	0.3946	68.9401	18.1697
0.202	1.7273	266	0.3940	68.1106	18.2630
0.3104	1.7338	267	0.3930	67.2811	18.3296
0.1801	1.7403	268	0.3928	67.1889	18.2763
0.2066	1.7468	269	0.3923	67.0968	18.2097
0.2075	1.7532	270	0.3919	66.7281	18.3029
0.228	1.7597	271	0.3919	67.1889	18.4494
0.2282	1.7662	272	0.3918	66.5438	18.2097
0.2287	1.7727	273	0.3919	66.5438	18.3296
0.2514	1.7792	274	0.3919	66.9124	18.3429
0.2874	1.7857	275	0.3922	67.0968	18.2363
0.2283	1.7922	276	0.3923	67.0046	18.2230
0.2041	1.7987	277	0.3924	67.2811	18.3695
0.2625	1.8052	278	0.3927	67.2811	18.1830
0.2938	1.8117	279	0.3926	68.2028	18.4628
0.2374	1.8182	280	0.3926	68.7558	18.6226
0.2848	1.8247	281	0.3927	68.7558	18.5161
0.1775	1.8312	282	0.3927	68.6636	18.4494
0.2224	1.8377	283	0.3929	69.1244	18.3828
0.2532	1.8442	284	0.3928	69.1244	18.4228
0.1904	1.8506	285	0.3927	69.5853	18.4628
0.2004	1.8571	286	0.3928	69.8618	18.5027
0.2069	1.8636	287	0.3930	69.7696	18.5294
0.2279	1.8701	288	0.3927	69.6774	18.4628
0.2839	1.8766	289	0.3930	69.7696	18.5560
0.2258	1.8831	290	0.3929	69.5853	18.4628
0.3162	1.8896	291	0.3932	69.7696	18.5027
0.2574	1.8961	292	0.3929	69.8618	18.4095
0.2575	1.9026	293	0.3932	70.4147	18.4761
0.1884	1.9091	294	0.3931	69.6774	18.3695
0.2561	1.9156	295	0.3932	70.5069	18.5427
0.2088	1.9221	296	0.3929	70.0461	18.4494
0.1983	1.9286	297	0.3928	70.0461	18.5161
0.2823	1.9351	298	0.3931	69.8618	18.4628
0.2579	1.9416	299	0.3928	70.3226	18.3962
0.2234	1.9481	300	0.3928	69.3088	18.3962
0.2227	1.9545	301	0.3929	69.8618	18.4095
0.2583	1.9610	302	0.3931	69.4931	18.4761
0.2862	1.9675	303	0.3929	69.4931	18.4228
0.29	1.9740	304	0.3928	69.7696	18.4494
0.2115	1.9805	305	0.3928	69.6774	18.3695
0.2109	1.9870	306	0.3929	69.5853	18.4494
0.229	1.9935	307	0.3928	69.4931	18.3429
0.2898	2.0	308	0.3930	70.0461	18.5560

Framework versions

Transformers 4.57.3
Pytorch 2.9.1+cu128
Datasets 4.4.1
Tokenizers 0.22.1

nkkbr/whisper-large-v3-zatoichi-ja-JDG_ver_20260217_lr_2.2e-5

Whisper Large v3 - Japanese Zatoichi ASR

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

nkkbr/
whisper-large-v3-zatoichi-ja-JDG_ver_20260217_lr_2.2e-5