learning_rate: 0.0002
train_batch_size: 32
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 256
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 0.05
num_epochs: 8
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss	Top5 Accuracy
51.2238	0.0674	2000	0.0442	6.3411	0.1203
41.3822	0.1347	4000	0.0790	4.9888	0.2743
36.6208	0.2021	6000	0.0978	4.3841	0.3289
34.1621	0.2695	8000	0.1300	4.0865	0.3881
32.5538	0.3368	10000	0.1516	3.9081	0.4229
31.4295	0.4042	12000	0.173	3.7777	0.4589
30.6596	0.4716	14000	0.1900	3.6560	0.4899
29.9171	0.5389	16000	0.205	3.5772	0.5101
29.3951	0.6063	18000	0.2210	3.5026	0.5327
28.8304	0.6737	20000	0.2341	3.4387	0.5535
28.4445	0.7411	22000	0.2465	3.3696	0.5693
27.8682	0.8084	24000	0.2544	3.3363	0.5802
27.7150	0.8758	26000	0.2633	3.2787	0.5974
27.2758	0.9432	28000	0.2722	3.2303	0.6119
26.6353	1.0105	30000	0.2793	3.2009	0.6202
26.6178	1.0779	32000	0.2846	3.1678	0.6291
26.2986	1.1452	34000	0.2933	3.1411	0.6367
26.1541	1.2126	36000	0.2990	3.1127	0.638
25.9314	1.2800	38000	0.3054	3.0818	0.6509
25.7470	1.3474	40000	0.3060	3.0629	0.6562
25.5577	1.4147	42000	0.3150	3.0334	0.6666
25.3361	1.4821	44000	0.3235	3.0070	0.6729
25.2382	1.5495	46000	0.332	2.9805	0.6769
25.0611	1.6168	48000	0.3352	2.9525	0.6838
24.8286	1.6842	50000	0.3399	2.9375	0.6879
24.6659	1.7516	52000	0.3457	2.9056	0.6962
24.5298	1.8189	54000	0.3489	2.8906	0.7036
24.2796	1.8863	56000	0.3487	2.8683	0.7122
24.1826	1.9537	58000	0.3579	2.8424	0.7151
23.6416	2.0210	60000	0.3582	2.8303	0.7207
23.5601	2.0884	62000	0.3681	2.8129	0.7257
23.4484	2.1558	64000	0.3678	2.7928	0.7295
23.3427	2.2231	66000	0.3758	2.7726	0.7349
23.1908	2.2905	68000	0.3707	2.7648	0.739
23.2677	2.3579	70000	0.3759	2.7461	0.7433
23.0687	2.4252	72000	0.381	2.7335	0.7481
22.9007	2.4926	74000	0.3874	2.7178	0.7491
22.8318	2.5600	76000	0.3880	2.6960	0.7552
22.7372	2.6273	78000	0.3906	2.6794	0.7616
22.6547	2.6947	80000	0.3887	2.6676	0.7634
22.5251	2.7621	82000	0.3987	2.6484	0.7651
22.4387	2.8294	84000	0.4015	2.6389	0.7685
22.2422	2.8968	86000	0.4033	2.6314	0.7684
22.2428	2.9642	88000	0.4073	2.6147	0.7753
21.6002	3.0315	90000	0.4108	2.6059	0.7766
22.3475	3.0989	92000	2.6969	0.3895	0.7555
22.4535	3.1663	94000	2.6879	0.3946	0.7572
22.3754	3.2336	96000	2.6955	0.3931	0.7586
22.3138	3.3010	98000	2.6783	0.3934	0.756
22.3112	3.3684	100000	2.6752	0.3945	0.7605
22.2399	3.4357	102000	2.6553	0.3963	0.7692
22.1860	3.5031	104000	2.6482	0.4018	0.7688
22.1966	3.5705	106000	2.6313	0.3974	0.7714
22.1117	3.6378	108000	2.6443	0.3982	0.7692
22.0962	3.7052	110000	2.6165	0.406	0.7763
21.9232	3.7726	112000	2.6135	0.4081	0.776
21.9250	3.8399	114000	2.6073	0.4099	0.7832
21.8086	3.9073	116000	2.6034	0.4056	0.7791
21.8148	3.9747	118000	2.5902	0.4088	0.778
21.1899	4.0420	120000	2.5835	0.4101	0.7834
21.2872	4.1094	122000	2.5744	0.413	0.7842
21.2121	4.1768	124000	2.5634	0.4115	0.7883
21.1537	4.2441	126000	2.5567	0.4207	0.7898
21.1143	4.3115	128000	2.5476	0.4237	0.7928
21.1197	4.3789	130000	2.5450	0.4199	0.7907
21.0866	4.4462	132000	2.5373	0.4245	0.7927
21.0555	4.5136	134000	2.5237	0.4268	0.7954
20.9838	4.5810	136000	2.5196	0.4269	0.7989
20.9613	4.6484	138000	2.5114	0.4286	0.8029
20.8328	4.7157	140000	2.5009	0.4317	0.7982
20.8385	4.7831	142000	2.4929	0.4336	0.8071
20.7476	4.8505	144000	2.4884	0.4335	0.8049
20.7141	4.9178	146000	2.4841	0.4333	0.8073
20.6954	4.9852	148000	2.4720	0.4348	0.8089
20.1964	5.0525	150000	2.4757	0.4417	0.8084
20.1919	5.1199	152000	2.4668	0.4421	0.8105
20.2029	5.1873	154000	2.4591	0.4435	0.8104
20.1102	5.2547	156000	2.4589	0.4404	0.8128
20.0693	5.3220	158000	2.4492	0.4432	0.8167
20.1529	5.3894	160000	2.4448	0.4465	0.8172
20.0076	5.4568	162000	2.4388	0.4483	0.8172
20.0621	5.5241	164000	2.4324	0.4468	0.8187
20.0232	5.5915	166000	2.4288	0.4476	0.822
19.9641	5.6589	168000	2.4208	0.4516	0.8217
20.0099	5.7262	170000	2.4136	0.4518	0.8183
19.9398	5.7936	172000	2.4098	0.4502	0.8216
19.9230	5.8610	174000	2.4090	0.4525	0.8206
19.8980	5.9283	176000	2.4040	0.4551	0.8255
19.8680	5.9957	178000	2.4050	0.4561	0.8256
19.3353	6.0631	180000	2.3990	0.4552	0.8251
19.4134	6.1304	182000	2.3977	0.4575	0.8259
19.3965	6.1978	184000	2.3921	0.4562	0.8238
19.4687	6.2652	186000	2.3889	0.4543	0.8305
19.3202	6.3325	188000	2.3877	0.4579	0.8285
19.3963	6.3999	190000	2.3822	0.4603	0.828
19.2950	6.4673	192000	2.3776	0.4597	0.8312
19.2901	6.5346	194000	2.3750	0.4644	0.8288
19.2375	6.6020	196000	2.3715	0.4649	0.8306
19.3044	6.6694	198000	2.3713	0.4652	0.8321
19.3410	6.7367	200000	2.3715	0.4638	0.8314
19.3258	6.8041	202000	2.3674	0.4674	0.8314
19.2042	6.8715	204000	2.3629	0.4656	0.8326
19.2207	6.9388	206000	2.3630	0.4671	0.8334
18.9808	7.0062	208000	2.3619	0.4644	0.8322
19.0327	7.0736	210000	2.3632	0.4662	0.8336
18.9676	7.1409	212000	2.3636	0.4685	0.8312
18.9669	7.2083	214000	2.3614	0.4678	0.8323
19.0012	7.2757	216000	2.3578	0.4671	0.8338
18.9087	7.3430	218000	2.3599	0.469	0.8337
18.9079	7.4104	220000	2.3582	0.4691	0.8348
18.8884	7.4778	222000	2.3571	0.4692	0.832
18.9994	7.5451	224000	2.3570	0.4697	0.8333
19.0156	7.6125	226000	2.3567	0.4696	0.8336
18.9452	7.6799	228000	2.3561	0.4705	0.8328
18.9182	7.7473	230000	2.3554	0.4701	0.8338
19.0061	7.8146	232000	2.3554	0.4697	0.8336
18.9271	7.8820	234000	2.3553	0.4691	0.8338
18.9573	7.9494	236000	2.3553	0.4691	0.8338

Framework versions

Transformers 5.2.0
Pytorch 2.7.1+cu118
Datasets 4.4.1
Tokenizers 0.22.1

benozen/chess-model-output

chess-model-output

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

benozen/
chess-model-output