v1_mistral_lora_real

This model is a fine-tuned version of peiyi9979/math-shepherd-mistral-7b-prm on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0002
• Accuracy: 1.0
• Precision: 1.0
• Recall: 1.0
• F1: 1.0

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• total_eval_batch_size: 32
• optimizer: Use 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_ratio: 0.1
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.7125	0.0071	10	0.5957	0.6805	0.5472	0.6915	0.6110
0.7473	0.0143	20	0.5921	0.6931	0.5622	0.6965	0.6222
0.6843	0.0214	30	0.5800	0.7094	0.5855	0.6816	0.6299
0.7083	0.0285	40	0.5597	0.7401	0.6432	0.6368	0.64
0.6862	0.0357	50	0.5293	0.7780	0.7216	0.6318	0.6737
0.626	0.0428	60	0.4788	0.8267	0.8107	0.6816	0.7405
0.4406	0.0499	70	0.4027	0.8917	0.8653	0.8308	0.8477
0.46	0.0571	80	0.2929	0.9386	0.9154	0.9154	0.9154
0.3254	0.0642	90	0.1629	0.9819	0.9848	0.9652	0.9749
0.2359	0.0714	100	0.0554	0.9982	0.9950	1.0	0.9975
0.263	0.0785	110	0.0200	1.0	1.0	1.0	1.0
0.228	0.0856	120	0.0094	1.0	1.0	1.0	1.0
0.2553	0.0928	130	0.0114	1.0	1.0	1.0	1.0
0.1633	0.0999	140	0.0083	1.0	1.0	1.0	1.0
0.145	0.1070	150	0.0087	1.0	1.0	1.0	1.0
0.1409	0.1142	160	0.0041	1.0	1.0	1.0	1.0
0.1955	0.1213	170	0.0042	1.0	1.0	1.0	1.0
0.1628	0.1284	180	0.0036	1.0	1.0	1.0	1.0
0.1454	0.1356	190	0.0019	1.0	1.0	1.0	1.0
0.1311	0.1427	200	0.0044	1.0	1.0	1.0	1.0
0.1937	0.1498	210	0.0035	1.0	1.0	1.0	1.0
0.1059	0.1570	220	0.0020	1.0	1.0	1.0	1.0
0.1352	0.1641	230	0.0023	1.0	1.0	1.0	1.0
0.1491	0.1712	240	0.0019	1.0	1.0	1.0	1.0
0.1245	0.1784	250	0.0012	1.0	1.0	1.0	1.0
0.1354	0.1855	260	0.0012	1.0	1.0	1.0	1.0
0.1177	0.1927	270	0.0012	1.0	1.0	1.0	1.0
0.1424	0.1998	280	0.0008	1.0	1.0	1.0	1.0
0.1343	0.2069	290	0.0008	1.0	1.0	1.0	1.0
0.1567	0.2141	300	0.0010	1.0	1.0	1.0	1.0
0.1094	0.2212	310	0.0009	1.0	1.0	1.0	1.0
0.1537	0.2283	320	0.0006	1.0	1.0	1.0	1.0
0.1344	0.2355	330	0.0006	1.0	1.0	1.0	1.0
0.1286	0.2426	340	0.0006	1.0	1.0	1.0	1.0
0.142	0.2497	350	0.0006	1.0	1.0	1.0	1.0
0.1177	0.2569	360	0.0009	1.0	1.0	1.0	1.0
0.1383	0.2640	370	0.0009	1.0	1.0	1.0	1.0
0.1647	0.2711	380	0.0004	1.0	1.0	1.0	1.0
0.0803	0.2783	390	0.0005	1.0	1.0	1.0	1.0
0.1476	0.2854	400	0.0004	1.0	1.0	1.0	1.0
0.1003	0.2925	410	0.0005	1.0	1.0	1.0	1.0
0.1122	0.2997	420	0.0004	1.0	1.0	1.0	1.0
0.1867	0.3068	430	0.0003	1.0	1.0	1.0	1.0
0.1216	0.3139	440	0.0005	1.0	1.0	1.0	1.0
0.1288	0.3211	450	0.0006	1.0	1.0	1.0	1.0
0.1243	0.3282	460	0.0005	1.0	1.0	1.0	1.0
0.1127	0.3354	470	0.0003	1.0	1.0	1.0	1.0
0.0775	0.3425	480	0.0003	1.0	1.0	1.0	1.0
0.1246	0.3496	490	0.0004	1.0	1.0	1.0	1.0
0.0864	0.3568	500	0.0002	1.0	1.0	1.0	1.0
0.1241	0.3639	510	0.0004	1.0	1.0	1.0	1.0
0.109	0.3710	520	0.0003	1.0	1.0	1.0	1.0
0.1117	0.3782	530	0.0004	1.0	1.0	1.0	1.0
0.1137	0.3853	540	0.0003	1.0	1.0	1.0	1.0
0.1193	0.3924	550	0.0006	1.0	1.0	1.0	1.0
0.1209	0.3996	560	0.0007	1.0	1.0	1.0	1.0
0.0934	0.4067	570	0.0007	1.0	1.0	1.0	1.0
0.1276	0.4138	580	0.0005	1.0	1.0	1.0	1.0
0.0851	0.4210	590	0.0004	1.0	1.0	1.0	1.0
0.1056	0.4281	600	0.0005	1.0	1.0	1.0	1.0
0.0951	0.4352	610	0.0004	1.0	1.0	1.0	1.0
0.1308	0.4424	620	0.0004	1.0	1.0	1.0	1.0
0.0814	0.4495	630	0.0004	1.0	1.0	1.0	1.0
0.0696	0.4567	640	0.0004	1.0	1.0	1.0	1.0
0.0721	0.4638	650	0.0004	1.0	1.0	1.0	1.0
0.0962	0.4709	660	0.0003	1.0	1.0	1.0	1.0
0.0829	0.4781	670	0.0002	1.0	1.0	1.0	1.0
0.1158	0.4852	680	0.0002	1.0	1.0	1.0	1.0
0.0949	0.4923	690	0.0002	1.0	1.0	1.0	1.0
0.1287	0.4995	700	0.0003	1.0	1.0	1.0	1.0
0.0834	0.5066	710	0.0003	1.0	1.0	1.0	1.0
0.099	0.5137	720	0.0003	1.0	1.0	1.0	1.0
0.12	0.5209	730	0.0004	1.0	1.0	1.0	1.0
0.0571	0.5280	740	0.0003	1.0	1.0	1.0	1.0
0.1133	0.5351	750	0.0004	1.0	1.0	1.0	1.0
0.1178	0.5423	760	0.0003	1.0	1.0	1.0	1.0
0.0866	0.5494	770	0.0004	1.0	1.0	1.0	1.0
0.0964	0.5565	780	0.0003	1.0	1.0	1.0	1.0
0.1165	0.5637	790	0.0004	1.0	1.0	1.0	1.0
0.1174	0.5708	800	0.0003	1.0	1.0	1.0	1.0
0.1468	0.5780	810	0.0002	1.0	1.0	1.0	1.0
0.1128	0.5851	820	0.0004	1.0	1.0	1.0	1.0
0.1446	0.5922	830	0.0003	1.0	1.0	1.0	1.0
0.0961	0.5994	840	0.0003	1.0	1.0	1.0	1.0
0.0736	0.6065	850	0.0002	1.0	1.0	1.0	1.0
0.0847	0.6136	860	0.0002	1.0	1.0	1.0	1.0
0.139	0.6208	870	0.0003	1.0	1.0	1.0	1.0
0.0775	0.6279	880	0.0003	1.0	1.0	1.0	1.0
0.0916	0.6350	890	0.0003	1.0	1.0	1.0	1.0
0.0944	0.6422	900	0.0002	1.0	1.0	1.0	1.0
0.1242	0.6493	910	0.0002	1.0	1.0	1.0	1.0
0.0975	0.6564	920	0.0002	1.0	1.0	1.0	1.0
0.0896	0.6636	930	0.0002	1.0	1.0	1.0	1.0
0.1359	0.6707	940	0.0002	1.0	1.0	1.0	1.0
0.0905	0.6778	950	0.0003	1.0	1.0	1.0	1.0
0.1045	0.6850	960	0.0002	1.0	1.0	1.0	1.0
0.0806	0.6921	970	0.0002	1.0	1.0	1.0	1.0
0.1121	0.6993	980	0.0002	1.0	1.0	1.0	1.0
0.1184	0.7064	990	0.0002	1.0	1.0	1.0	1.0
0.0945	0.7135	1000	0.0002	1.0	1.0	1.0	1.0
0.1041	0.7207	1010	0.0002	1.0	1.0	1.0	1.0
0.0912	0.7278	1020	0.0002	1.0	1.0	1.0	1.0
0.1167	0.7349	1030	0.0002	1.0	1.0	1.0	1.0
0.0952	0.7421	1040	0.0002	1.0	1.0	1.0	1.0
0.1048	0.7492	1050	0.0002	1.0	1.0	1.0	1.0
0.0877	0.7563	1060	0.0002	1.0	1.0	1.0	1.0
0.1051	0.7635	1070	0.0002	1.0	1.0	1.0	1.0
0.1027	0.7706	1080	0.0002	1.0	1.0	1.0	1.0
0.0802	0.7777	1090	0.0002	1.0	1.0	1.0	1.0
0.1118	0.7849	1100	0.0002	1.0	1.0	1.0	1.0
0.109	0.7920	1110	0.0002	1.0	1.0	1.0	1.0
0.097	0.7991	1120	0.0002	1.0	1.0	1.0	1.0
0.1045	0.8063	1130	0.0002	1.0	1.0	1.0	1.0
0.0872	0.8134	1140	0.0002	1.0	1.0	1.0	1.0
0.1075	0.8205	1150	0.0002	1.0	1.0	1.0	1.0
0.1322	0.8277	1160	0.0002	1.0	1.0	1.0	1.0
0.1056	0.8348	1170	0.0002	1.0	1.0	1.0	1.0
0.0884	0.8420	1180	0.0002	1.0	1.0	1.0	1.0
0.1284	0.8491	1190	0.0002	1.0	1.0	1.0	1.0
0.1099	0.8562	1200	0.0002	1.0	1.0	1.0	1.0
0.1023	0.8634	1210	0.0002	1.0	1.0	1.0	1.0
0.086	0.8705	1220	0.0002	1.0	1.0	1.0	1.0
0.0877	0.8776	1230	0.0002	1.0	1.0	1.0	1.0
0.1032	0.8848	1240	0.0002	1.0	1.0	1.0	1.0
0.1446	0.8919	1250	0.0002	1.0	1.0	1.0	1.0
0.1079	0.8990	1260	0.0002	1.0	1.0	1.0	1.0
0.0716	0.9062	1270	0.0002	1.0	1.0	1.0	1.0
0.1181	0.9133	1280	0.0002	1.0	1.0	1.0	1.0
0.1087	0.9204	1290	0.0002	1.0	1.0	1.0	1.0
0.086	0.9276	1300	0.0002	1.0	1.0	1.0	1.0
0.071	0.9347	1310	0.0002	1.0	1.0	1.0	1.0
0.0858	0.9418	1320	0.0002	1.0	1.0	1.0	1.0
0.0859	0.9490	1330	0.0002	1.0	1.0	1.0	1.0
0.1165	0.9561	1340	0.0002	1.0	1.0	1.0	1.0
0.1189	0.9633	1350	0.0002	1.0	1.0	1.0	1.0
0.142	0.9704	1360	0.0002	1.0	1.0	1.0	1.0
0.1336	0.9775	1370	0.0002	1.0	1.0	1.0	1.0
0.1183	0.9847	1380	0.0002	1.0	1.0	1.0	1.0
0.0961	0.9918	1390	0.0002	1.0	1.0	1.0	1.0
0.076	0.9989	1400	0.0002	1.0	1.0	1.0	1.0

Framework versions

• PEFT 0.12.0
• Transformers 4.46.0
• Pytorch 2.4.0+cu118
• Datasets 3.0.0
• Tokenizers 0.20.1