learning_rate: 0.0005
train_batch_size: 1
eval_batch_size: 1
seed: 12345
distributed_type: multi-GPU
num_devices: 8
total_train_batch_size: 8
total_eval_batch_size: 8
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: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Roc Auc	Best Threshold	Tpr At Best Threshold	Fpr At Best Threshold	Accuracy At Best Threshold	Avg Score True Positives	Avg Score True Negatives
No log	0	0	40.8364	0.6060	0.0100	0.4225	0.1364	0.5269	0.1403	0.0493
28.7586	0.0658	5	38.8030	0.6015	0.0222	0.4085	0.1591	0.5108	0.1569	0.0566
24.2448	0.1316	10	24.8508	0.6061	0.1944	0.4577	0.2045	0.5376	0.3293	0.1443
7.0614	0.1974	15	11.7330	0.6345	0.9955	0.4366	0.0909	0.5484	0.7320	0.6643
40.028	0.2632	20	13.3462	0.6418	1.0000	0.4155	0.0682	0.5376	0.9015	0.8906
21.0819	0.3289	25	9.7099	0.6663	0.9978	0.4155	0.0455	0.5430	0.7810	0.6951
12.7163	0.3947	30	9.8787	0.6726	0.9406	0.4789	0.1364	0.5699	0.6581	0.4570
14.1502	0.4605	35	8.1137	0.7021	0.8505	0.6408	0.2955	0.6559	0.7210	0.4873
6.374	0.5263	40	9.5062	0.7242	0.9985	0.6620	0.2727	0.6774	0.9449	0.8770
21.8456	0.5921	45	10.3835	0.7490	0.9998	0.5493	0.1136	0.6290	0.9719	0.9051
13.2588	0.6579	50	6.3049	0.7626	0.9919	0.6197	0.1591	0.6720	0.9147	0.7222
16.3367	0.7237	55	4.8704	0.8003	0.9415	0.6408	0.1364	0.6935	0.8146	0.4425
2.5361	0.7895	60	6.1344	0.8093	0.9863	0.7042	0.1591	0.7366	0.9221	0.7005
4.0751	0.8553	65	9.5643	0.7749	0.4688	0.5915	0.1591	0.6505	0.5490	0.1892
12.1795	0.9211	70	5.3396	0.7973	0.9641	0.6479	0.1591	0.6935	0.8787	0.6014
8.2354	0.9868	75	5.6834	0.8044	0.9930	0.6197	0.0909	0.6882	0.9154	0.6768
11.9897	1.0526	80	13.3635	0.7465	0.1871	0.6056	0.1818	0.6559	0.4373	0.1725
5.5301	1.1184	85	6.1608	0.7754	0.9771	0.7394	0.25	0.7419	0.9403	0.6768
7.8286	1.1842	90	9.2827	0.8239	0.9974	0.8451	0.2955	0.8118	0.9844	0.8395
2.0006	1.25	95	7.0314	0.8458	0.9998	0.5634	0.0227	0.6613	0.9716	0.7113
5.0788	1.3158	100	7.9317	0.8383	0.2422	0.7817	0.2045	0.7849	0.7012	0.2021
4.9874	1.3816	105	6.3856	0.8551	0.9907	0.8169	0.2273	0.8065	0.9542	0.5815
9.1875	1.4474	110	9.6971	0.8670	0.9998	0.7958	0.1818	0.8011	0.9953	0.7584
3.2955	1.5132	115	7.2262	0.8447	0.2975	0.7535	0.1818	0.7688	0.6750	0.1875
1.2063	1.5789	120	6.9988	0.8369	0.9898	0.8521	0.3409	0.8065	0.9757	0.6567
22.3711	1.6447	125	11.4332	0.8446	0.9999	0.7606	0.2273	0.7634	0.9978	0.8520
5.3319	1.7105	130	10.9098	0.8393	0.0043	0.9085	0.3636	0.8441	0.5549	0.1499
0.7761	1.7763	135	6.0298	0.8358	0.4688	0.7887	0.1818	0.7957	0.7428	0.2396
11.9151	1.8421	140	4.3366	0.8507	0.9281	0.7958	0.1591	0.8065	0.9075	0.4076
12.1123	1.9079	145	9.1366	0.8546	0.0359	0.8662	0.2273	0.8441	0.5397	0.1318
3.0131	1.9737	150	4.1858	0.8813	0.9877	0.7746	0.1364	0.7957	0.9608	0.5270
16.3813	2.0395	155	10.9843	0.8798	0.9999	0.7606	0.1364	0.7849	0.9993	0.9185
5.976	2.1053	160	4.0728	0.8480	0.9377	0.7183	0.1364	0.7527	0.8704	0.4236
0.1195	2.1711	165	4.6016	0.8680	0.6893	0.7817	0.1591	0.7957	0.7968	0.2738
5.0645	2.2368	170	6.5664	0.8827	0.9888	0.9014	0.2727	0.8602	0.9771	0.6194
9.5125	2.3026	175	5.7698	0.8694	0.7827	0.7817	0.1591	0.7957	0.8131	0.2740
8.5699	2.3684	180	8.7242	0.8552	0.6334	0.7183	0.1364	0.7527	0.7187	0.1793
0.0743	2.4342	185	6.8172	0.8955	0.9997	0.7817	0.1136	0.8065	0.9768	0.5694
0.717	2.5	190	7.2021	0.9039	0.9996	0.8239	0.1591	0.8280	0.9829	0.5844
0.59	2.5658	195	5.7756	0.8948	0.9999	0.6972	0.0455	0.7581	0.9540	0.4875
0.0308	2.6316	200	5.7955	0.8680	0.9775	0.7113	0.0909	0.7581	0.8333	0.2880
20.1623	2.6974	205	5.2183	0.8824	0.9585	0.8732	0.25	0.8441	0.9364	0.4051
0.0687	2.7632	210	5.5173	0.8712	0.9937	0.7887	0.1818	0.7957	0.9579	0.4748
0.4536	2.8289	215	5.6076	0.8126	0.8549	0.8310	0.2955	0.8011	0.8910	0.4075
13.8618	2.8947	220	9.7817	0.7738	0.0567	0.8592	0.3636	0.8065	0.6223	0.2844
0.0119	2.9605	225	7.7343	0.8123	0.9852	0.8803	0.3636	0.8226	0.9743	0.6635
0.2227	3.0263	230	8.9229	0.8509	0.9984	0.8521	0.2727	0.8226	0.9923	0.7379
0.0033	3.0921	235	5.5232	0.8592	0.9927	0.7958	0.2273	0.7903	0.9528	0.5273
0.0031	3.1579	240	5.6570	0.8608	0.9807	0.8451	0.2727	0.8172	0.9514	0.5105
3.345	3.2237	245	5.9922	0.8784	0.9987	0.7746	0.1591	0.7903	0.9560	0.5021
0.0002	3.2895	250	5.5904	0.8918	0.9756	0.8803	0.2273	0.8548	0.9384	0.4410
7.9272	3.3553	255	5.6920	0.8958	0.9641	0.7887	0.0909	0.8172	0.8714	0.2913
0.0038	3.4211	260	5.7780	0.8957	0.8970	0.8592	0.1818	0.8495	0.8743	0.2793
5.125	3.4868	265	5.7583	0.8954	0.9912	0.8169	0.1364	0.8280	0.9068	0.3718
2.3759	3.5526	270	6.3772	0.8788	0.9310	0.8521	0.2045	0.8387	0.8850	0.3230
0.026	3.6184	275	16.7049	0.8528	0.0046	0.8099	0.2045	0.8065	0.5950	0.1386
14.5264	3.6842	280	8.2086	0.8648	0.1422	0.8451	0.2045	0.8333	0.7696	0.2053
6.8855	3.75	285	5.2626	0.8892	0.8652	0.9014	0.2727	0.8602	0.9325	0.3585
0.0022	3.8158	290	4.9146	0.8949	0.8289	0.9085	0.25	0.8710	0.9355	0.3553
2.4643	3.8816	295	4.9935	0.8967	0.3923	0.9085	0.2273	0.8763	0.8801	0.2486
5.8053	3.9474	300	5.0847	0.8941	0.7879	0.8521	0.1818	0.8441	0.8710	0.2420
0.3474	4.0132	305	4.7603	0.8946	0.9149	0.8803	0.1818	0.8656	0.9228	0.2969
0.0246	4.0789	310	4.8641	0.9000	0.7154	0.8732	0.1818	0.8602	0.8787	0.2304
0.0001	4.1447	315	6.0055	0.9066	0.9995	0.8310	0.1136	0.8441	0.9672	0.4301
0.0002	4.2105	320	6.4645	0.9058	0.9998	0.8380	0.1364	0.8441	0.9689	0.4699
0.0922	4.2763	325	5.9817	0.8992	0.9795	0.9225	0.2273	0.8871	0.9513	0.4229
0.0133	4.3421	330	6.2559	0.8832	0.9980	0.8451	0.1818	0.8387	0.9424	0.4530
0.0045	4.4079	335	6.4340	0.8752	0.9329	0.7887	0.1364	0.8065	0.8644	0.2843
0.0212	4.4737	340	7.2002	0.8778	0.9987	0.7887	0.1136	0.8118	0.9262	0.4243
0.0	4.5395	345	9.4407	0.8712	1.0000	0.7746	0.1136	0.8011	0.9551	0.4895
2.1406	4.6053	350	9.5657	0.8609	0.9999	0.7535	0.1364	0.7796	0.9411	0.4539
0.0044	4.6711	355	8.6190	0.8709	0.9993	0.7535	0.1136	0.7849	0.9282	0.4027
0.0879	4.7368	360	8.0168	0.8759	0.9964	0.8028	0.1364	0.8172	0.9246	0.3964
0.1084	4.8026	365	8.5945	0.8846	0.9995	0.8099	0.1364	0.8226	0.9469	0.4422
0.0	4.8684	370	8.8690	0.8783	0.9995	0.8239	0.1591	0.8280	0.9500	0.4527
0.0002	4.9342	375	8.2028	0.8769	0.9972	0.8451	0.2045	0.8333	0.9370	0.4359
0.0774	5.0	380	7.7547	0.8735	0.9876	0.8380	0.2045	0.8280	0.9051	0.3849
0.0098	5.0658	385	8.9577	0.8724	0.9983	0.8592	0.2273	0.8387	0.9411	0.4464
0.0554	5.1316	390	10.9894	0.8616	0.9999	0.8239	0.2273	0.8118	0.9614	0.5048
0.0	5.1974	395	10.1356	0.8716	1.0000	0.7817	0.1364	0.8011	0.9549	0.4800
0.0	5.2632	400	9.0367	0.8824	0.9997	0.8169	0.1591	0.8226	0.9391	0.4381
0.0001	5.3289	405	8.6214	0.8847	0.9990	0.8380	0.1818	0.8333	0.9283	0.4225
2.5938	5.3947	410	8.3628	0.8850	0.9986	0.8169	0.1591	0.8226	0.9182	0.4012
8.0625	5.4605	415	8.4276	0.8924	0.7957	0.8380	0.1591	0.8387	0.8611	0.2599
8.0242	5.5263	420	10.0505	0.8941	0.3702	0.8310	0.1591	0.8333	0.8021	0.1664
0.0022	5.5921	425	8.0955	0.8978	0.8758	0.8803	0.2045	0.8602	0.8864	0.2898
0.0011	5.6579	430	8.5264	0.9012	0.9996	0.8521	0.1818	0.8441	0.9439	0.4261
0.4772	5.7237	435	9.4212	0.8911	0.9998	0.8732	0.1818	0.8602	0.9617	0.4679
0.0001	5.7895	440	8.6994	0.9031	0.9997	0.8592	0.1818	0.8495	0.9492	0.4279
0.0001	5.8553	445	8.3435	0.9013	0.9994	0.8521	0.1818	0.8441	0.9312	0.3928
0.0025	5.9211	450	7.9349	0.8976	0.9995	0.7887	0.1364	0.8065	0.9122	0.3419
0.0	5.9868	455	7.9755	0.8921	0.9893	0.8521	0.2045	0.8387	0.9111	0.3368
0.0	6.0526	460	7.9633	0.8914	0.8007	0.9155	0.25	0.8763	0.9119	0.3279
0.0001	6.1184	465	8.1330	0.8856	0.8129	0.9014	0.25	0.8656	0.9110	0.3221
0.0	6.1842	470	8.1791	0.8855	0.8366	0.9014	0.25	0.8656	0.9114	0.3249
0.0016	6.25	475	8.3424	0.8856	0.9931	0.8521	0.2045	0.8387	0.9229	0.3578
0.0003	6.3158	480	8.7398	0.8855	0.9994	0.8451	0.2045	0.8333	0.9347	0.3934
0.0309	6.3816	485	8.8993	0.8870	0.9999	0.8169	0.1818	0.8172	0.9438	0.4127
0.0	6.4474	490	9.2989	0.8882	0.9999	0.8239	0.1818	0.8226	0.9505	0.4202
0.0	6.5132	495	9.5010	0.8890	1.0000	0.8239	0.1818	0.8226	0.9523	0.4243
6.0625	6.5789	500	9.5599	0.8904	0.9993	0.8662	0.2273	0.8441	0.9514	0.4199
0.0	6.6447	505	8.8966	0.8883	0.9998	0.8239	0.1818	0.8226	0.9396	0.4013
0.0	6.7105	510	8.3211	0.8916	0.9995	0.7958	0.1364	0.8118	0.9160	0.3525
0.0031	6.7763	515	8.2376	0.8939	0.9992	0.7958	0.1364	0.8118	0.9068	0.3367
0.0	6.8421	520	8.1951	0.8961	0.9977	0.8169	0.1591	0.8226	0.9121	0.3389
0.0	6.9079	525	8.1881	0.8984	0.9980	0.8239	0.1591	0.8280	0.9141	0.3390
0.0057	6.9737	530	8.1538	0.9003	0.9968	0.8451	0.1818	0.8387	0.9152	0.3386
0.0	7.0395	535	8.4027	0.8989	0.9977	0.8239	0.1591	0.8280	0.9110	0.3368
0.0	7.1053	540	8.4496	0.9008	0.9852	0.8803	0.2045	0.8602	0.9135	0.3355
0.0	7.1711	545	8.3727	0.9006	0.9832	0.8873	0.2045	0.8656	0.9131	0.3374
0.0	7.2368	550	8.4559	0.9001	0.9842	0.8873	0.2045	0.8656	0.9137	0.3426
0.0	7.3026	555	8.5881	0.9004	0.9909	0.8803	0.2045	0.8602	0.9178	0.3503
0.0	7.3684	560	8.8236	0.8977	0.9982	0.8732	0.2045	0.8548	0.9324	0.3705
0.0	7.4342	565	9.0579	0.8990	0.9991	0.8732	0.2045	0.8548	0.9407	0.3851
4.2812	7.5	570	8.8891	0.8976	0.9983	0.8732	0.2045	0.8548	0.9365	0.3804
0.0	7.5658	575	8.7430	0.8968	0.9966	0.8803	0.2273	0.8548	0.9301	0.3712
0.0	7.6316	580	8.6918	0.8969	0.9962	0.8732	0.2045	0.8548	0.9264	0.3660
0.0	7.6974	585	8.8465	0.8936	0.9978	0.8662	0.2045	0.8495	0.9280	0.3725
0.2656	7.7632	590	8.9124	0.8942	0.9964	0.8803	0.2273	0.8548	0.9294	0.3717
0.0001	7.8289	595	9.0410	0.8954	0.9979	0.8662	0.2045	0.8495	0.9305	0.3733
0.0	7.8947	600	8.9758	0.8958	0.9985	0.8592	0.2045	0.8441	0.9325	0.3765
0.0	7.9605	605	9.1102	0.8955	0.9983	0.8662	0.2045	0.8495	0.9340	0.3803
0.0123	8.0263	610	9.0449	0.8976	0.9985	0.8732	0.2045	0.8548	0.9346	0.3808
0.0	8.0921	615	9.2282	0.8960	0.9988	0.8732	0.2045	0.8548	0.9366	0.3831
0.0	8.1579	620	9.1741	0.8976	0.9991	0.8662	0.2045	0.8495	0.9379	0.3861
0.003	8.2237	625	9.3382	0.8958	0.9991	0.8732	0.2045	0.8548	0.9380	0.3885
0.0	8.2895	630	9.3798	0.8970	0.9992	0.8732	0.2045	0.8548	0.9402	0.3925
0.0	8.3553	635	9.4814	0.8980	0.9994	0.8732	0.2045	0.8548	0.9424	0.3959
0.0	8.4211	640	9.5738	0.8971	0.9994	0.8732	0.2045	0.8548	0.9418	0.3939
3.7969	8.4868	645	9.5038	0.8960	0.9994	0.8662	0.2045	0.8495	0.9397	0.3894
0.0	8.5526	650	9.3220	0.8963	0.9991	0.8662	0.2045	0.8495	0.9383	0.3881
0.0	8.6184	655	9.1874	0.8956	0.9990	0.8662	0.2045	0.8495	0.9346	0.3802
0.0	8.6842	660	9.2758	0.8968	0.9987	0.8662	0.2045	0.8495	0.9325	0.3779
0.0	8.75	665	9.2191	0.8967	0.9986	0.8662	0.2045	0.8495	0.9318	0.3765
0.0	8.8158	670	9.1030	0.8968	0.9964	0.8873	0.2273	0.8602	0.9302	0.3757
0.0	8.8816	675	9.1426	0.8958	0.9975	0.8662	0.2045	0.8495	0.9301	0.3776
0.0	8.9474	680	9.0891	0.8966	0.9990	0.8592	0.2045	0.8441	0.9312	0.3758
0.0	9.0132	685	9.0907	0.8968	0.9986	0.8662	0.2045	0.8495	0.9312	0.3740
0.0	9.0789	690	9.1227	0.8971	0.9983	0.8662	0.2045	0.8495	0.9304	0.3755
0.0	9.1447	695	9.1793	0.8954	0.9981	0.8592	0.2045	0.8441	0.9287	0.3726
0.0	9.2105	700	9.1335	0.8964	0.9984	0.8662	0.2045	0.8495	0.9286	0.3730
0.0002	9.2763	705	9.1281	0.8956	0.9945	0.8873	0.2273	0.8602	0.9283	0.3710
0.0	9.3421	710	9.1072	0.8961	0.9951	0.8873	0.2273	0.8602	0.9288	0.3735
0.0	9.4079	715	9.1382	0.8956	0.9984	0.8662	0.2045	0.8495	0.9293	0.3711
0.0	9.4737	720	9.0695	0.8969	0.9979	0.8662	0.2045	0.8495	0.9292	0.3731
0.0	9.5395	725	9.1084	0.8960	0.9984	0.8662	0.2045	0.8495	0.9294	0.3729
0.0	9.6053	730	9.0823	0.8966	0.9996	0.8239	0.1591	0.8280	0.9289	0.3708
0.0155	9.6711	735	9.0659	0.8968	0.9971	0.8803	0.2045	0.8602	0.9293	0.3721
0.0003	9.7368	740	9.0718	0.8967	0.9977	0.8662	0.2045	0.8495	0.9294	0.3714
0.0	9.8026	745	9.0564	0.8963	0.9972	0.8732	0.2045	0.8548	0.9275	0.3717
0.0001	9.8684	750	9.0083	0.8972	0.9984	0.8662	0.2045	0.8495	0.9291	0.3705
0.0	9.9342	755	9.0757	0.8964	0.9985	0.8592	0.2045	0.8441	0.9296	0.3714
0.0	10.0	760	9.0810	0.8961	0.9981	0.8662	0.2045	0.8495	0.9292	0.3727

Framework versions

PEFT 0.18.1
Transformers 4.57.6
Pytorch 2.9.1+cu128
Datasets 4.5.0
Tokenizers 0.22.2

leobianco/bosch_RM_Qwen_S12345_LLM_false_STRUCT_false_epo10_lr5e-4_r8_2602050903

bosch_RM_Qwen_S12345_LLM_false_STRUCT_false_epo10_lr5e-4_r8_2602050903

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

leobianco/
bosch_RM_Qwen_S12345_LLM_false_STRUCT_false_epo10_lr5e-4_r8_2602050903